id	sid	tid	token	lemma	pos
fcis-31421	1	1	frontiers	frontier	NOUN
fcis-31421	1	2	in	in	ADP
fcis-31421	1	3	computing	computing	NOUN
fcis-31421	1	4	and	and	CCONJ
fcis-31421	1	5	intelligent	intelligent	ADJ
fcis-31421	1	6	systems	system	NOUN
fcis-31421	1	7	issn	issn	VERB
fcis-31421	1	8	:	:	PUNCT
fcis-31421	1	9	2832	2832	NUM
fcis-31421	1	10	-	-	SYM
fcis-31421	1	11	6024	6024	NUM
fcis-31421	1	12	|	|	NOUN
fcis-31421	1	13	vol	vol	NOUN
fcis-31421	1	14	.	.	PROPN
fcis-31421	2	1	13	13	NUM
fcis-31421	2	2	,	,	PUNCT
fcis-31421	2	3	no	no	INTJ
fcis-31421	2	4	.	.	NOUN
fcis-31421	2	5	1	1	NUM
fcis-31421	2	6	,	,	PUNCT
fcis-31421	2	7	2025	2025	NUM
fcis-31421	2	8	51	51	NUM
fcis-31421	2	9	large‐scale	large‐scale	NOUN
fcis-31421	2	10	fire	fire	NOUN
fcis-31421	2	11	detection	detection	NOUN
fcis-31421	2	12	based	base	VERB
fcis-31421	2	13	on	on	ADP
fcis-31421	2	14	yolov8	yolov8	PROPN
fcis-31421	2	15	lightweight	lightweight	PROPN
fcis-31421	2	16	model	model	NOUN
fcis-31421	2	17	and	and	CCONJ
fcis-31421	2	18	its	its	PRON
fcis-31421	2	19	improved	improved	ADJ
fcis-31421	2	20	system	system	NOUN
fcis-31421	2	21	rui	rui	PROPN
fcis-31421	2	22	wen	wen	PROPN
fcis-31421	2	23	*	*	PUNCT
fcis-31421	2	24	college	college	PROPN
fcis-31421	2	25	of	of	ADP
fcis-31421	2	26	computer	computer	NOUN
fcis-31421	2	27	and	and	CCONJ
fcis-31421	2	28	information	information	NOUN
fcis-31421	2	29	science	science	NOUN
fcis-31421	2	30	,	,	PUNCT
fcis-31421	2	31	college	college	NOUN
fcis-31421	2	32	of	of	ADP
fcis-31421	2	33	software	software	NOUN
fcis-31421	2	34	,	,	PUNCT
fcis-31421	2	35	southwest	southwest	PROPN
fcis-31421	2	36	university	university	NOUN
fcis-31421	2	37	,	,	PUNCT
fcis-31421	2	38	chongqing	chongqing	PROPN
fcis-31421	2	39	,	,	PUNCT
fcis-31421	2	40	400715	400715	NUM
fcis-31421	2	41	,	,	PUNCT
fcis-31421	2	42	china	china	PROPN
fcis-31421	2	43	*	*	PUNCT
fcis-31421	2	44	corresponding	correspond	VERB
fcis-31421	2	45	author	author	NOUN
fcis-31421	2	46	email	email	NOUN
fcis-31421	2	47	:	:	PUNCT
fcis-31421	2	48	wr18223295633@email.swu.edu.cn	wr18223295633@email.swu.edu.cn	PROPN
fcis-31421	2	49	abstract	abstract	NOUN
fcis-31421	2	50	:	:	PUNCT
fcis-31421	2	51	in	in	ADP
fcis-31421	2	52	recent	recent	ADJ
fcis-31421	2	53	years	year	NOUN
fcis-31421	2	54	,	,	PUNCT
fcis-31421	2	55	global	global	ADJ
fcis-31421	2	56	warming	warming	NOUN
fcis-31421	2	57	,	,	PUNCT
fcis-31421	2	58	population	population	NOUN
fcis-31421	2	59	increase	increase	NOUN
fcis-31421	2	60	,	,	PUNCT
fcis-31421	2	61	the	the	DET
fcis-31421	2	62	expansion	expansion	NOUN
fcis-31421	2	63	of	of	ADP
fcis-31421	2	64	human	human	ADJ
fcis-31421	2	65	production	production	NOUN
fcis-31421	2	66	and	and	CCONJ
fcis-31421	2	67	living	living	NOUN
fcis-31421	2	68	areas	area	NOUN
fcis-31421	2	69	and	and	CCONJ
fcis-31421	2	70	many	many	ADJ
fcis-31421	2	71	other	other	ADJ
fcis-31421	2	72	factors	factor	NOUN
fcis-31421	2	73	have	have	AUX
fcis-31421	2	74	led	lead	VERB
fcis-31421	2	75	to	to	ADP
fcis-31421	2	76	the	the	DET
fcis-31421	2	77	increasing	increase	VERB
fcis-31421	2	78	frequency	frequency	NOUN
fcis-31421	2	79	of	of	ADP
fcis-31421	2	80	forest	forest	NOUN
fcis-31421	2	81	fires	fire	NOUN
fcis-31421	2	82	,	,	PUNCT
fcis-31421	2	83	forest	forest	NOUN
fcis-31421	2	84	fire	fire	NOUN
fcis-31421	2	85	prevention	prevention	NOUN
fcis-31421	2	86	and	and	CCONJ
fcis-31421	2	87	response	response	NOUN
fcis-31421	2	88	has	have	AUX
fcis-31421	2	89	become	become	VERB
fcis-31421	2	90	more	more	ADV
fcis-31421	2	91	and	and	CCONJ
fcis-31421	2	92	more	more	ADV
fcis-31421	2	93	important	important	ADJ
fcis-31421	2	94	.	.	PUNCT
fcis-31421	3	1	nowadays	nowadays	ADV
fcis-31421	3	2	,	,	PUNCT
fcis-31421	3	3	computer	computer	NOUN
fcis-31421	3	4	vision	vision	NOUN
fcis-31421	3	5	methods	method	NOUN
fcis-31421	3	6	and	and	CCONJ
fcis-31421	3	7	video	video	NOUN
fcis-31421	3	8	surveillance	surveillance	NOUN
fcis-31421	3	9	cameras	camera	NOUN
fcis-31421	3	10	have	have	VERB
fcis-31421	3	11	a	a	DET
fcis-31421	3	12	wide	wide	ADJ
fcis-31421	3	13	range	range	NOUN
fcis-31421	3	14	of	of	ADP
fcis-31421	3	15	applications	application	NOUN
fcis-31421	3	16	,	,	PUNCT
fcis-31421	3	17	has	have	AUX
fcis-31421	3	18	shifted	shift	VERB
fcis-31421	3	19	from	from	ADP
fcis-31421	3	20	traditional	traditional	ADJ
fcis-31421	3	21	detection	detection	NOUN
fcis-31421	3	22	methods	method	NOUN
fcis-31421	3	23	to	to	PART
fcis-31421	3	24	machine	machine	VERB
fcis-31421	3	25	deep	deep	ADJ
fcis-31421	3	26	learning	learning	NOUN
fcis-31421	3	27	technology	technology	NOUN
fcis-31421	3	28	,	,	PUNCT
fcis-31421	3	29	these	these	DET
fcis-31421	3	30	technologies	technology	NOUN
fcis-31421	3	31	have	have	VERB
fcis-31421	3	32	a	a	DET
fcis-31421	3	33	very	very	ADV
fcis-31421	3	34	important	important	ADJ
fcis-31421	3	35	role	role	NOUN
fcis-31421	3	36	in	in	ADP
fcis-31421	3	37	promoting	promote	VERB
fcis-31421	3	38	and	and	CCONJ
fcis-31421	3	39	significance	significance	NOUN
fcis-31421	3	40	of	of	ADP
fcis-31421	3	41	fire	fire	NOUN
fcis-31421	3	42	detection	detection	NOUN
fcis-31421	3	43	,	,	PUNCT
fcis-31421	3	44	image	image	NOUN
fcis-31421	3	45	detection	detection	NOUN
fcis-31421	3	46	can	can	AUX
fcis-31421	3	47	be	be	AUX
fcis-31421	3	48	extracted	extract	VERB
fcis-31421	3	49	through	through	ADP
fcis-31421	3	50	the	the	DET
fcis-31421	3	51	dataset	dataset	NOUN
fcis-31421	3	52	training	training	NOUN
fcis-31421	3	53	after	after	ADP
fcis-31421	3	54	the	the	DET
fcis-31421	3	55	object	object	NOUN
fcis-31421	3	56	features	feature	VERB
fcis-31421	3	57	,	,	PUNCT
fcis-31421	3	58	compared	compare	VERB
fcis-31421	3	59	with	with	ADP
fcis-31421	3	60	the	the	DET
fcis-31421	3	61	traditional	traditional	ADJ
fcis-31421	3	62	fire	fire	NOUN
fcis-31421	3	63	detection	detection	NOUN
fcis-31421	3	64	methods	method	NOUN
fcis-31421	3	65	,	,	PUNCT
fcis-31421	3	66	it	it	PRON
fcis-31421	3	67	can	can	AUX
fcis-31421	3	68	detect	detect	VERB
fcis-31421	3	69	and	and	CCONJ
fcis-31421	3	70	warn	warn	VERB
fcis-31421	3	71	of	of	ADP
fcis-31421	3	72	fires	fire	NOUN
fcis-31421	3	73	in	in	ADP
fcis-31421	3	74	a	a	DET
fcis-31421	3	75	short	short	ADJ
fcis-31421	3	76	period	period	NOUN
fcis-31421	3	77	of	of	ADP
fcis-31421	3	78	time	time	NOUN
fcis-31421	3	79	,	,	PUNCT
fcis-31421	3	80	and	and	CCONJ
fcis-31421	3	81	the	the	DET
fcis-31421	3	82	coverage	coverage	NOUN
fcis-31421	3	83	of	of	ADP
fcis-31421	3	84	the	the	DET
fcis-31421	3	85	area	area	NOUN
fcis-31421	3	86	is	be	AUX
fcis-31421	3	87	wider	wide	ADJ
fcis-31421	3	88	,	,	PUNCT
fcis-31421	3	89	and	and	CCONJ
fcis-31421	3	90	can	can	AUX
fcis-31421	3	91	realize	realize	VERB
fcis-31421	3	92	a	a	DET
fcis-31421	3	93	wide	wide	ADJ
fcis-31421	3	94	range	range	NOUN
fcis-31421	3	95	of	of	ADP
fcis-31421	3	96	detection	detection	NOUN
fcis-31421	3	97	.	.	PUNCT
fcis-31421	4	1	it	it	PRON
fcis-31421	4	2	can	can	AUX
fcis-31421	4	3	realize	realize	VERB
fcis-31421	4	4	a	a	DET
fcis-31421	4	5	wide	wide	ADJ
fcis-31421	4	6	range	range	NOUN
fcis-31421	4	7	of	of	ADP
fcis-31421	4	8	detection	detection	NOUN
fcis-31421	4	9	.	.	PUNCT
fcis-31421	5	1	yolov8	yolov8	NOUN
fcis-31421	5	2	and	and	CCONJ
fcis-31421	5	3	its	its	PRON
fcis-31421	5	4	improved	improved	ADJ
fcis-31421	5	5	models	model	NOUN
fcis-31421	5	6	have	have	VERB
fcis-31421	5	7	great	great	ADJ
fcis-31421	5	8	advantages	advantage	NOUN
fcis-31421	5	9	in	in	ADP
fcis-31421	5	10	this	this	DET
fcis-31421	5	11	regard	regard	NOUN
fcis-31421	5	12	.	.	PUNCT
fcis-31421	6	1	the	the	DET
fcis-31421	6	2	application	application	NOUN
fcis-31421	6	3	of	of	ADP
fcis-31421	6	4	yolov8	yolov8	NOUN
fcis-31421	6	5	-	-	PUNCT
fcis-31421	6	6	based	base	VERB
fcis-31421	6	7	model	model	NOUN
fcis-31421	6	8	in	in	ADP
fcis-31421	6	9	fire	fire	NOUN
fcis-31421	6	10	detection	detection	NOUN
fcis-31421	6	11	can	can	AUX
fcis-31421	6	12	greatly	greatly	ADV
fcis-31421	6	13	reduce	reduce	VERB
fcis-31421	6	14	the	the	DET
fcis-31421	6	15	consumption	consumption	NOUN
fcis-31421	6	16	of	of	ADP
fcis-31421	6	17	manpower	manpower	NOUN
fcis-31421	6	18	and	and	CCONJ
fcis-31421	6	19	financial	financial	ADJ
fcis-31421	6	20	resources	resource	NOUN
fcis-31421	6	21	for	for	ADP
fcis-31421	6	22	detection	detection	NOUN
fcis-31421	6	23	,	,	PUNCT
fcis-31421	6	24	and	and	CCONJ
fcis-31421	6	25	reduce	reduce	VERB
fcis-31421	6	26	the	the	DET
fcis-31421	6	27	economic	economic	ADJ
fcis-31421	6	28	losses	loss	NOUN
fcis-31421	6	29	caused	cause	VERB
fcis-31421	6	30	by	by	ADP
fcis-31421	6	31	fire	fire	NOUN
fcis-31421	6	32	.	.	PUNCT
fcis-31421	7	1	this	this	DET
fcis-31421	7	2	article	article	NOUN
fcis-31421	7	3	mainly	mainly	ADV
fcis-31421	7	4	tested	test	VERB
fcis-31421	7	5	yolov8	yolov8	NOUN
fcis-31421	7	6	with	with	ADP
fcis-31421	7	7	fire	fire	NOUN
fcis-31421	7	8	smoke	smoke	NOUN
fcis-31421	7	9	detection	detection	NOUN
fcis-31421	7	10	and	and	CCONJ
fcis-31421	7	11	compared	compare	VERB
fcis-31421	7	12	it	it	PRON
fcis-31421	7	13	with	with	ADP
fcis-31421	7	14	convolutional	convolutional	ADJ
fcis-31421	7	15	neural	neural	ADJ
fcis-31421	7	16	network	network	NOUN
fcis-31421	7	17	and	and	CCONJ
fcis-31421	7	18	multimodal	multimodal	NOUN
fcis-31421	7	19	model	model	NOUN
fcis-31421	7	20	as	as	ADV
fcis-31421	7	21	well	well	ADV
fcis-31421	7	22	as	as	ADP
fcis-31421	7	23	based	base	VERB
fcis-31421	7	24	on	on	ADP
fcis-31421	7	25	yolov8	yolov8	NOUN
fcis-31421	7	26	model	model	NOUN
fcis-31421	7	27	by	by	ADP
fcis-31421	7	28	adding	add	VERB
fcis-31421	7	29	modification	modification	NOUN
fcis-31421	7	30	module	module	NOUN
fcis-31421	7	31	to	to	PART
fcis-31421	7	32	achieve	achieve	VERB
fcis-31421	7	33	light	light	ADJ
fcis-31421	7	34	weight	weight	NOUN
fcis-31421	7	35	,	,	PUNCT
fcis-31421	7	36	wide	wide	ADJ
fcis-31421	7	37	range	range	NOUN
fcis-31421	7	38	detection	detection	NOUN
fcis-31421	7	39	and	and	CCONJ
fcis-31421	7	40	other	other	ADJ
fcis-31421	7	41	functions	function	NOUN
fcis-31421	7	42	.	.	PUNCT
fcis-31421	8	1	keywords	keyword	NOUN
fcis-31421	8	2	:	:	PUNCT
fcis-31421	8	3	forest	forest	NOUN
fcis-31421	8	4	fire	fire	NOUN
fcis-31421	8	5	detection	detection	NOUN
fcis-31421	8	6	;	;	PUNCT
fcis-31421	8	7	multimodal	multimodal	ADJ
fcis-31421	8	8	fusion	fusion	NOUN
fcis-31421	8	9	;	;	PUNCT
fcis-31421	8	10	wide	wide	ADJ
fcis-31421	8	11	range	range	NOUN
fcis-31421	8	12	monitoring	monitoring	NOUN
fcis-31421	8	13	;	;	PUNCT
fcis-31421	8	14	lightweight	lightweight	ADJ
fcis-31421	8	15	system	system	NOUN
fcis-31421	8	16	;	;	PUNCT
fcis-31421	8	17	low	low	ADJ
fcis-31421	8	18	latency	latency	NOUN
fcis-31421	8	19	feedback	feedback	NOUN
fcis-31421	8	20	.	.	PUNCT
fcis-31421	9	1	1	1	X
fcis-31421	9	2	.	.	X
fcis-31421	9	3	introduction	introduction	NOUN
fcis-31421	9	4	fire	fire	NOUN
fcis-31421	9	5	is	be	AUX
fcis-31421	9	6	a	a	DET
fcis-31421	9	7	major	major	ADJ
fcis-31421	9	8	natural	natural	ADJ
fcis-31421	9	9	disaster	disaster	NOUN
fcis-31421	9	10	that	that	PRON
fcis-31421	9	11	occurs	occur	VERB
fcis-31421	9	12	frequently	frequently	ADV
fcis-31421	9	13	around	around	ADP
fcis-31421	9	14	the	the	DET
fcis-31421	9	15	globe	globe	NOUN
fcis-31421	9	16	,	,	PUNCT
fcis-31421	9	17	posing	pose	VERB
fcis-31421	9	18	a	a	DET
fcis-31421	9	19	serious	serious	ADJ
fcis-31421	9	20	threat	threat	NOUN
fcis-31421	9	21	to	to	ADP
fcis-31421	9	22	human	human	ADJ
fcis-31421	9	23	society	society	NOUN
fcis-31421	9	24	,	,	PUNCT
fcis-31421	9	25	the	the	DET
fcis-31421	9	26	economy	economy	NOUN
fcis-31421	9	27	and	and	CCONJ
fcis-31421	9	28	the	the	DET
fcis-31421	9	29	ecological	ecological	ADJ
fcis-31421	9	30	environment	environment	NOUN
fcis-31421	9	31	.	.	PUNCT
fcis-31421	10	1	according	accord	VERB
fcis-31421	10	2	to	to	ADP
fcis-31421	10	3	statistics	statistic	NOUN
fcis-31421	10	4	,	,	PUNCT
fcis-31421	10	5	the	the	DET
fcis-31421	10	6	annual	annual	ADJ
fcis-31421	10	7	direct	direct	ADJ
fcis-31421	10	8	economic	economic	ADJ
fcis-31421	10	9	losses	loss	NOUN
fcis-31421	10	10	caused	cause	VERB
fcis-31421	10	11	by	by	ADP
fcis-31421	10	12	fires	fire	NOUN
fcis-31421	10	13	worldwide	worldwide	ADV
fcis-31421	10	14	exceed	exceed	VERB
fcis-31421	10	15	hundreds	hundred	NOUN
fcis-31421	10	16	of	of	ADP
fcis-31421	10	17	billions	billion	NOUN
fcis-31421	10	18	of	of	ADP
fcis-31421	10	19	dollars	dollar	NOUN
fcis-31421	10	20	and	and	CCONJ
fcis-31421	10	21	lead	lead	VERB
fcis-31421	10	22	to	to	ADP
fcis-31421	10	23	tens	ten	NOUN
fcis-31421	10	24	of	of	ADP
fcis-31421	10	25	thousands	thousand	NOUN
fcis-31421	10	26	of	of	ADP
fcis-31421	10	27	casualties	casualty	NOUN
fcis-31421	10	28	,	,	PUNCT
fcis-31421	10	29	as	as	SCONJ
fcis-31421	10	30	shown	show	VERB
fcis-31421	10	31	in	in	ADP
fcis-31421	10	32	table	table	NOUN
fcis-31421	10	33	1	1	NUM
fcis-31421	10	34	,	,	PUNCT
fcis-31421	10	35	the	the	DET
fcis-31421	10	36	loss	loss	NOUN
fcis-31421	10	37	data	datum	NOUN
fcis-31421	10	38	caused	cause	VERB
fcis-31421	10	39	by	by	ADP
fcis-31421	10	40	fires	fire	NOUN
fcis-31421	10	41	from	from	ADP
fcis-31421	10	42	2012	2012	NUM
fcis-31421	10	43	to	to	ADP
fcis-31421	10	44	2020	2020	NUM
fcis-31421	10	45	shows	show	VERB
fcis-31421	10	46	that	that	SCONJ
fcis-31421	10	47	it	it	PRON
fcis-31421	10	48	poses	pose	VERB
fcis-31421	10	49	a	a	DET
fcis-31421	10	50	huge	huge	ADJ
fcis-31421	10	51	threat	threat	NOUN
fcis-31421	10	52	to	to	ADP
fcis-31421	10	53	society	society	NOUN
fcis-31421	10	54	and	and	CCONJ
fcis-31421	10	55	individuals	individual	NOUN
fcis-31421	10	56	.	.	PUNCT
fcis-31421	11	1	table	table	NOUN
fcis-31421	11	2	1	1	NUM
fcis-31421	11	3	.	.	PUNCT
fcis-31421	11	4	consolidated	consolidated	ADJ
fcis-31421	11	5	table	table	NOUN
fcis-31421	11	6	of	of	ADP
fcis-31421	11	7	data	datum	NOUN
fcis-31421	11	8	from	from	ADP
fcis-31421	11	9	the	the	DET
fcis-31421	11	10	china	china	PROPN
fcis-31421	11	11	statistical	statistical	ADJ
fcis-31421	11	12	yearbook	yearbook	NOUN
fcis-31421	11	13	:	:	PUNCT
fcis-31421	11	14	forest	forest	NOUN
fcis-31421	11	15	fires	fire	NOUN
fcis-31421	11	16	nationwide	nationwide	ADV
fcis-31421	11	17	(	(	PUNCT
fcis-31421	11	18	2012	2012	NUM
fcis-31421	11	19	-	-	SYM
fcis-31421	11	20	2020	2020	NUM
fcis-31421	11	21	data	datum	NOUN
fcis-31421	11	22	)	)	PUNCT
fcis-31421	12	1	year	year	NOUN
fcis-31421	12	2	number	number	NOUN
fcis-31421	12	3	of	of	ADP
fcis-31421	12	4	forest	forest	NOUN
fcis-31421	12	5	fires	fire	NOUN
fcis-31421	12	6	(	(	PUNCT
fcis-31421	12	7	times	time	NOUN
fcis-31421	12	8	)	)	PUNCT
fcis-31421	12	9	total	total	ADJ
fcis-31421	12	10	fire	fire	NOUN
fcis-31421	12	11	area	area	NOUN
fcis-31421	12	12	(	(	PUNCT
fcis-31421	12	13	ha	ha	INTJ
fcis-31421	12	14	)	)	PUNCT
fcis-31421	12	15	area	area	NOUN
fcis-31421	12	16	of	of	ADP
fcis-31421	12	17	victimized	victimize	VERB
fcis-31421	12	18	forests	forest	NOUN
fcis-31421	12	19	(	(	PUNCT
fcis-31421	12	20	ha	ha	INTJ
fcis-31421	12	21	)	)	PUNCT
fcis-31421	12	22	number	number	NOUN
fcis-31421	12	23	of	of	ADP
fcis-31421	12	24	casualties	casualty	NOUN
fcis-31421	12	25	depreciation	depreciation	NOUN
fcis-31421	12	26	of	of	ADP
fcis-31421	12	27	other	other	ADJ
fcis-31421	12	28	losses	loss	NOUN
fcis-31421	12	29	(	(	PUNCT
fcis-31421	12	30	in	in	ADP
fcis-31421	12	31	ten	ten	NUM
fcis-31421	12	32	thousand	thousand	NUM
fcis-31421	12	33	yuan	yuan	NOUN
fcis-31421	12	34	)	)	PUNCT
fcis-31421	12	35	2012	2012	NUM
fcis-31421	12	36	3966	3966	NUM
fcis-31421	12	37	2397	2397	NUM
fcis-31421	12	38	1568	1568	NUM
fcis-31421	12	39	1	1	NUM
fcis-31421	12	40	0	0	NUM
fcis-31421	12	41	2013	2013	NUM
fcis-31421	12	42	3929	3929	NUM
fcis-31421	12	43	2347	2347	NUM
fcis-31421	12	44	1582	1582	NUM
fcis-31421	12	45	0	0	NUM
fcis-31421	12	46	0	0	NUM
fcis-31421	12	47	2014	2014	NUM
fcis-31421	12	48	3703	3703	NUM
fcis-31421	12	49	2080	2080	NUM
fcis-31421	12	50	1620	1620	NUM
fcis-31421	12	51	2	2	NUM
fcis-31421	12	52	1	1	NUM
fcis-31421	12	53	2015	2015	NUM
fcis-31421	12	54	2936	2936	NUM
fcis-31421	12	55	1676	1676	NUM
fcis-31421	12	56	1254	1254	NUM
fcis-31421	12	57	6	6	NUM
fcis-31421	12	58	0	0	NUM
fcis-31421	12	59	2016	2016	NUM
fcis-31421	12	60	2034	2034	NUM
fcis-31421	12	61	1340	1340	NUM
fcis-31421	12	62	693	693	NUM
fcis-31421	12	63	1	1	NUM
fcis-31421	12	64	0	0	NUM
fcis-31421	12	65	2017	2017	NUM
fcis-31421	12	66	3223	3223	NUM
fcis-31421	12	67	2258	2258	NUM
fcis-31421	12	68	958	958	NUM
fcis-31421	12	69	4	4	NUM
fcis-31421	12	70	3	3	NUM
fcis-31421	12	71	2018	2018	NUM
fcis-31421	12	72	2478	2478	NUM
fcis-31421	12	73	1579	1579	NUM
fcis-31421	12	74	894	894	NUM
fcis-31421	12	75	3	3	NUM
fcis-31421	12	76	2	2	NUM
fcis-31421	12	77	2019	2019	NUM
fcis-31421	12	78	2345	2345	NUM
fcis-31421	12	79	1534	1534	NUM
fcis-31421	12	80	802	802	NUM
fcis-31421	12	81	8	8	NUM
fcis-31421	12	82	1	1	NUM
fcis-31421	12	83	2020	2020	NUM
fcis-31421	12	84	1153	1153	NUM
fcis-31421	12	85	722	722	NUM
fcis-31421	12	86	424	424	NUM
fcis-31421	12	87	7	7	NUM
fcis-31421	12	88	0	0	NUM
fcis-31421	12	89	fig	fig	NOUN
fcis-31421	12	90	1	1	NUM
fcis-31421	12	91	.	.	PUNCT
fcis-31421	12	92	forest	forest	NOUN
fcis-31421	12	93	fire	fire	NOUN
fcis-31421	12	94	detection	detection	NOUN
fcis-31421	12	95	device	device	NOUN
fcis-31421	12	96	traditional	traditional	ADJ
fcis-31421	12	97	fire	fire	NOUN
fcis-31421	12	98	detection	detection	NOUN
fcis-31421	12	99	technology	technology	NOUN
fcis-31421	12	100	mainly	mainly	ADV
fcis-31421	12	101	relies	rely	VERB
fcis-31421	12	102	on	on	ADP
fcis-31421	12	103	smoke	smoke	NOUN
fcis-31421	12	104	sensors	sensor	NOUN
fcis-31421	12	105	,	,	PUNCT
fcis-31421	12	106	temperature	temperature	NOUN
fcis-31421	12	107	sensors	sensor	NOUN
fcis-31421	12	108	and	and	CCONJ
fcis-31421	12	109	manual	manual	ADJ
fcis-31421	12	110	monitoring	monitoring	NOUN
fcis-31421	12	111	,	,	PUNCT
fcis-31421	12	112	which	which	PRON
fcis-31421	12	113	has	have	VERB
fcis-31421	12	114	problems	problem	NOUN
fcis-31421	12	115	such	such	ADJ
fcis-31421	12	116	as	as	ADP
fcis-31421	12	117	high	high	ADJ
fcis-31421	12	118	response	response	NOUN
fcis-31421	12	119	delay	delay	NOUN
fcis-31421	12	120	,	,	PUNCT
fcis-31421	12	121	high	high	ADJ
fcis-31421	12	122	false	false	ADJ
fcis-31421	12	123	alarm	alarm	NOUN
fcis-31421	12	124	rate	rate	NOUN
fcis-31421	12	125	and	and	CCONJ
fcis-31421	12	126	limited	limited	ADJ
fcis-31421	12	127	coverage	coverage	NOUN
fcis-31421	12	128	.	.	PUNCT
fcis-31421	13	1	with	with	ADP
fcis-31421	13	2	the	the	DET
fcis-31421	13	3	acceleration	acceleration	NOUN
fcis-31421	13	4	of	of	ADP
fcis-31421	13	5	urbanization	urbanization	NOUN
fcis-31421	13	6	and	and	CCONJ
fcis-31421	13	7	the	the	DET
fcis-31421	13	8	increase	increase	NOUN
fcis-31421	13	9	of	of	ADP
fcis-31421	13	10	complex	complex	ADJ
fcis-31421	13	11	building	building	NOUN
fcis-31421	13	12	structures	structure	NOUN
fcis-31421	13	13	,	,	PUNCT
fcis-31421	13	14	the	the	DET
fcis-31421	13	15	demand	demand	NOUN
fcis-31421	13	16	for	for	ADP
fcis-31421	13	17	real	real	ADJ
fcis-31421	13	18	-	-	PUNCT
fcis-31421	13	19	time	time	NOUN
fcis-31421	13	20	,	,	PUNCT
fcis-31421	13	21	accurate	accurate	ADJ
fcis-31421	13	22	and	and	CCONJ
fcis-31421	13	23	robust	robust	ADJ
fcis-31421	13	24	fire	fire	NOUN
fcis-31421	13	25	detection	detection	NOUN
fcis-31421	13	26	is	be	AUX
fcis-31421	13	27	becoming	become	VERB
fcis-31421	13	28	increasingly	increasingly	ADV
fcis-31421	13	29	urgent	urgent	ADJ
fcis-31421	13	30	.	.	PUNCT
fcis-31421	14	1	in	in	ADP
fcis-31421	14	2	this	this	DET
fcis-31421	14	3	context	context	NOUN
fcis-31421	14	4	,	,	PUNCT
fcis-31421	14	5	fire	fire	NOUN
fcis-31421	14	6	detection	detection	NOUN
fcis-31421	14	7	methods	method	NOUN
fcis-31421	14	8	based	base	VERB
fcis-31421	14	9	on	on	ADP
fcis-31421	14	10	computer	computer	NOUN
fcis-31421	14	11	vision	vision	NOUN
fcis-31421	14	12	and	and	CCONJ
fcis-31421	14	13	deep	deep	ADJ
fcis-31421	14	14	learning	learning	NOUN
fcis-31421	14	15	techniques	technique	NOUN
fcis-31421	14	16	have	have	AUX
fcis-31421	14	17	gradually	gradually	ADV
fcis-31421	14	18	become	become	VERB
fcis-31421	14	19	a	a	DET
fcis-31421	14	20	research	research	NOUN
fcis-31421	14	21	hotspot	hotspot	NOUN
fcis-31421	14	22	,	,	PUNCT
fcis-31421	14	23	which	which	PRON
fcis-31421	14	24	can	can	AUX
fcis-31421	14	25	realize	realize	VERB
fcis-31421	14	26	non	non	ADJ
fcis-31421	14	27	-	-	ADJ
fcis-31421	14	28	contact	contact	NOUN
fcis-31421	14	29	and	and	CCONJ
fcis-31421	14	30	high	high	ADJ
fcis-31421	14	31	-	-	PUNCT
fcis-31421	14	32	precision	precision	NOUN
fcis-31421	14	33	fire	fire	NOUN
fcis-31421	14	34	recognition	recognition	NOUN
fcis-31421	14	35	and	and	CCONJ
fcis-31421	14	36	early	early	ADJ
fcis-31421	14	37	warning	warning	NOUN
fcis-31421	14	38	by	by	ADP
fcis-31421	14	39	analyzing	analyze	VERB
fcis-31421	14	40	video	video	NOUN
fcis-31421	14	41	,	,	PUNCT
fcis-31421	14	42	image	image	NOUN
fcis-31421	14	43	and	and	CCONJ
fcis-31421	14	44	multimodal	multimodal	NOUN
fcis-31421	14	45	data	datum	NOUN
fcis-31421	14	46	.	.	PUNCT
fcis-31421	15	1	as	as	SCONJ
fcis-31421	15	2	shown	show	VERB
fcis-31421	15	3	in	in	ADP
fcis-31421	15	4	figure	figure	NOUN
fcis-31421	15	5	1	1	NUM
fcis-31421	15	6	,	,	PUNCT
fcis-31421	15	7	this	this	PRON
fcis-31421	15	8	is	be	AUX
fcis-31421	15	9	a	a	DET
fcis-31421	15	10	forest	forest	NOUN
fcis-31421	15	11	fire	fire	NOUN
fcis-31421	15	12	detection	detection	NOUN
fcis-31421	15	13	device	device	NOUN
fcis-31421	15	14	used	use	VERB
fcis-31421	15	15	for	for	ADP
fcis-31421	15	16	large	large	ADJ
fcis-31421	15	17	-	-	PUNCT
fcis-31421	15	18	scale	scale	NOUN
fcis-31421	15	19	detection	detection	NOUN
fcis-31421	15	20	.	.	PUNCT
fcis-31421	16	1	traditional	traditional	ADJ
fcis-31421	16	2	fire	fire	NOUN
fcis-31421	16	3	detection	detection	NOUN
fcis-31421	16	4	technologies	technology	NOUN
fcis-31421	16	5	mainly	mainly	ADV
fcis-31421	16	6	rely	rely	VERB
fcis-31421	16	7	on	on	ADP
fcis-31421	16	8	physical	physical	ADJ
fcis-31421	16	9	sensors	sensor	NOUN
fcis-31421	16	10	and	and	CCONJ
fcis-31421	16	11	manual	manual	ADJ
fcis-31421	16	12	monitoring	monitoring	NOUN
fcis-31421	16	13	,	,	PUNCT
fcis-31421	16	14	including	include	VERB
fcis-31421	16	15	smoke	smoke	NOUN
fcis-31421	16	16	sensors	sensor	NOUN
fcis-31421	16	17	,	,	PUNCT
fcis-31421	16	18	temperature	temperature	NOUN
fcis-31421	16	19	sensors	sensor	NOUN
fcis-31421	16	20	,	,	PUNCT
fcis-31421	16	21	and	and	CCONJ
fcis-31421	16	22	uv	uv	NOUN
fcis-31421	16	23	/	/	SYM
fcis-31421	16	24	ir	ir	PROPN
fcis-31421	16	25	flame	flame	NOUN
fcis-31421	16	26	detectors	detector	NOUN
fcis-31421	16	27	.	.	PUNCT
fcis-31421	17	1	these	these	DET
fcis-31421	17	2	methods	method	NOUN
fcis-31421	17	3	trigger	trigger	NOUN
fcis-31421	17	4	alarms	alarm	NOUN
fcis-31421	17	5	by	by	ADP
fcis-31421	17	6	capturing	capture	VERB
fcis-31421	17	7	specific	specific	ADJ
fcis-31421	17	8	physical	physical	ADJ
fcis-31421	17	9	signals	signal	NOUN
fcis-31421	17	10	,	,	PUNCT
fcis-31421	17	11	such	such	ADJ
fcis-31421	17	12	as	as	ADP
fcis-31421	17	13	smoke	smoke	NOUN
fcis-31421	17	14	particle	particle	NOUN
fcis-31421	17	15	concentration	concentration	NOUN
fcis-31421	17	16	,	,	PUNCT
fcis-31421	17	17	temperature	temperature	NOUN
fcis-31421	17	18	surge	surge	NOUN
fcis-31421	17	19	,	,	PUNCT
fcis-31421	17	20	flame	flame	NOUN
fcis-31421	17	21	spectrum	spectrum	NOUN
fcis-31421	17	22	,	,	PUNCT
fcis-31421	17	23	etc	etc	X
fcis-31421	17	24	.	.	X
fcis-31421	17	25	,	,	PUNCT
fcis-31421	17	26	with	with	ADP
fcis-31421	17	27	low	low	ADJ
fcis-31421	17	28	cost	cost	NOUN
fcis-31421	17	29	,	,	PUNCT
fcis-31421	17	30	simple	simple	ADJ
fcis-31421	17	31	deployment	deployment	NOUN
fcis-31421	17	32	and	and	CCONJ
fcis-31421	17	33	other	other	ADJ
fcis-31421	17	34	advantages	advantage	NOUN
fcis-31421	17	35	,	,	PUNCT
fcis-31421	17	36	but	but	CCONJ
fcis-31421	17	37	its	its	PRON
fcis-31421	17	38	limitations	limitation	NOUN
fcis-31421	17	39	are	be	AUX
fcis-31421	17	40	significant	significant	ADJ
fcis-31421	17	41	:	:	PUNCT
fcis-31421	17	42	a	a	DET
fcis-31421	17	43	single	single	ADJ
fcis-31421	17	44	dimension	dimension	NOUN
fcis-31421	17	45	of	of	ADP
fcis-31421	17	46	information	information	NOUN
fcis-31421	17	47	,	,	PUNCT
fcis-31421	17	48	relying	rely	VERB
fcis-31421	17	49	only	only	ADV
fcis-31421	17	50	on	on	ADP
fcis-31421	17	51	a	a	DET
fcis-31421	17	52	single	single	ADJ
fcis-31421	17	53	signal	signal	NOUN
fcis-31421	17	54	,	,	PUNCT
fcis-31421	17	55	can	can	AUX
fcis-31421	17	56	not	not	PART
fcis-31421	17	57	be	be	AUX
fcis-31421	17	58	integrated	integrate	VERB
fcis-31421	17	59	to	to	PART
fcis-31421	17	60	determine	determine	VERB
fcis-31421	17	61	the	the	DET
fcis-31421	17	62	characteristics	characteristic	NOUN
fcis-31421	17	63	of	of	ADP
fcis-31421	17	64	the	the	DET
fcis-31421	17	65	fire	fire	NOUN
fcis-31421	17	66	source	source	NOUN
fcis-31421	17	67	;	;	PUNCT
fcis-31421	17	68	response	response	NOUN
fcis-31421	17	69	passive	passive	ADJ
fcis-31421	17	70	and	and	CCONJ
fcis-31421	17	71	delayed	delay	VERB
fcis-31421	17	72	,	,	PUNCT
fcis-31421	17	73	need	need	VERB
fcis-31421	17	74	to	to	PART
fcis-31421	17	75	wait	wait	VERB
fcis-31421	17	76	52	52	NUM
fcis-31421	17	77	for	for	SCONJ
fcis-31421	17	78	the	the	DET
fcis-31421	17	79	physical	physical	ADJ
fcis-31421	17	80	parameters	parameter	NOUN
fcis-31421	17	81	to	to	PART
fcis-31421	17	82	reach	reach	VERB
fcis-31421	17	83	the	the	DET
fcis-31421	17	84	threshold	threshold	NOUN
fcis-31421	17	85	,	,	PUNCT
fcis-31421	17	86	the	the	DET
fcis-31421	17	87	negative	negative	ADJ
fcis-31421	17	88	ignition	ignition	NOUN
fcis-31421	17	89	or	or	CCONJ
fcis-31421	17	90	early	early	ADJ
fcis-31421	17	91	fire	fire	NOUN
fcis-31421	17	92	detection	detection	NOUN
fcis-31421	17	93	lag	lag	NOUN
fcis-31421	17	94	;	;	PUNCT
fcis-31421	17	95	high	high	ADJ
fcis-31421	17	96	false	false	ADJ
fcis-31421	17	97	alarm	alarm	NOUN
fcis-31421	17	98	rate	rate	NOUN
fcis-31421	17	99	,	,	PUNCT
fcis-31421	17	100	susceptible	susceptible	ADJ
fcis-31421	17	101	to	to	ADP
fcis-31421	17	102	environmental	environmental	ADJ
fcis-31421	17	103	interference	interference	NOUN
fcis-31421	17	104	;	;	PUNCT
fcis-31421	17	105	limited	limited	ADJ
fcis-31421	17	106	coverage	coverage	NOUN
fcis-31421	17	107	,	,	PUNCT
fcis-31421	17	108	the	the	DET
fcis-31421	17	109	sensor	sensor	NOUN
fcis-31421	17	110	needs	need	VERB
fcis-31421	17	111	to	to	PART
fcis-31421	17	112	wait	wait	VERB
fcis-31421	17	113	for	for	SCONJ
fcis-31421	17	114	the	the	DET
fcis-31421	17	115	physical	physical	ADJ
fcis-31421	17	116	parameters	parameter	NOUN
fcis-31421	17	117	to	to	PART
fcis-31421	17	118	reach	reach	VERB
fcis-31421	17	119	the	the	DET
fcis-31421	17	120	threshold	threshold	NOUN
fcis-31421	17	121	,	,	PUNCT
fcis-31421	17	122	for	for	ADP
fcis-31421	17	123	the	the	DET
fcis-31421	17	124	negative	negative	ADJ
fcis-31421	17	125	ignition	ignition	NOUN
fcis-31421	17	126	of	of	ADP
fcis-31421	17	127	fire	fire	NOUN
fcis-31421	17	128	or	or	CCONJ
fcis-31421	17	129	early	early	ADJ
fcis-31421	17	130	fire	fire	NOUN
fcis-31421	17	131	detection	detection	NOUN
fcis-31421	17	132	lag	lag	NOUN
fcis-31421	17	133	;	;	PUNCT
fcis-31421	17	134	high	high	ADJ
fcis-31421	17	135	false	false	ADJ
fcis-31421	17	136	alarm	alarm	NOUN
fcis-31421	17	137	rate	rate	NOUN
fcis-31421	17	138	,	,	PUNCT
fcis-31421	17	139	susceptible	susceptible	ADJ
fcis-31421	17	140	to	to	ADP
fcis-31421	17	141	environmental	environmental	ADJ
fcis-31421	17	142	interference	interference	NOUN
fcis-31421	17	143	;	;	PUNCT
fcis-31421	17	144	limited	limited	ADJ
fcis-31421	17	145	coverage	coverage	NOUN
fcis-31421	17	146	,	,	PUNCT
fcis-31421	17	147	the	the	DET
fcis-31421	17	148	sensor	sensor	NOUN
fcis-31421	17	149	needs	need	VERB
fcis-31421	17	150	to	to	PART
fcis-31421	17	151	be	be	AUX
fcis-31421	17	152	in	in	ADP
fcis-31421	17	153	close	close	ADJ
fcis-31421	17	154	contact	contact	NOUN
fcis-31421	17	155	with	with	ADP
fcis-31421	17	156	the	the	DET
fcis-31421	17	157	fire	fire	NOUN
fcis-31421	17	158	source	source	NOUN
fcis-31421	17	159	,	,	PUNCT
fcis-31421	17	160	it	it	PRON
fcis-31421	17	161	is	be	AUX
fcis-31421	17	162	difficult	difficult	ADJ
fcis-31421	17	163	to	to	PART
fcis-31421	17	164	apply	apply	VERB
fcis-31421	17	165	to	to	ADP
fcis-31421	17	166	open	open	ADJ
fcis-31421	17	167	space	space	NOUN
fcis-31421	17	168	or	or	CCONJ
fcis-31421	17	169	complex	complex	ADJ
fcis-31421	17	170	scenarios	scenario	NOUN
fcis-31421	17	171	.	.	PUNCT
fcis-31421	18	1	in	in	ADP
fcis-31421	18	2	addition	addition	NOUN
fcis-31421	18	3	,	,	PUNCT
fcis-31421	18	4	although	although	SCONJ
fcis-31421	18	5	manual	manual	ADJ
fcis-31421	18	6	monitoring	monitoring	NOUN
fcis-31421	18	7	can	can	AUX
fcis-31421	18	8	make	make	VERB
fcis-31421	18	9	empirical	empirical	ADJ
fcis-31421	18	10	judgments	judgment	NOUN
fcis-31421	18	11	,	,	PUNCT
fcis-31421	18	12	there	there	PRON
fcis-31421	18	13	are	be	VERB
fcis-31421	18	14	problems	problem	NOUN
fcis-31421	18	15	such	such	ADJ
fcis-31421	18	16	as	as	ADP
fcis-31421	18	17	high	high	ADJ
fcis-31421	18	18	labor	labor	NOUN
fcis-31421	18	19	costs	cost	NOUN
fcis-31421	18	20	,	,	PUNCT
fcis-31421	18	21	poor	poor	ADJ
fcis-31421	18	22	real	real	ADJ
fcis-31421	18	23	-	-	PUNCT
fcis-31421	18	24	time	time	NOUN
fcis-31421	18	25	performance	performance	NOUN
fcis-31421	18	26	,	,	PUNCT
fcis-31421	18	27	and	and	CCONJ
fcis-31421	18	28	large	large	ADJ
fcis-31421	18	29	subjective	subjective	ADJ
fcis-31421	18	30	errors	error	NOUN
fcis-31421	18	31	.	.	PUNCT
fcis-31421	19	1	the	the	DET
fcis-31421	19	2	traditional	traditional	ADJ
fcis-31421	19	3	method	method	NOUN
fcis-31421	19	4	is	be	AUX
fcis-31421	19	5	essentially	essentially	ADV
fcis-31421	19	6	a	a	DET
fcis-31421	19	7	“	"	PUNCT
fcis-31421	19	8	passive	passive	ADJ
fcis-31421	19	9	response	response	NOUN
fcis-31421	19	10	”	"	PUNCT
fcis-31421	19	11	mechanism	mechanism	NOUN
fcis-31421	19	12	,	,	PUNCT
fcis-31421	19	13	which	which	PRON
fcis-31421	19	14	lacks	lack	VERB
fcis-31421	19	15	the	the	DET
fcis-31421	19	16	multi	multi	ADJ
fcis-31421	19	17	-	-	ADJ
fcis-31421	19	18	dimensional	dimensional	ADJ
fcis-31421	19	19	perception	perception	NOUN
fcis-31421	19	20	of	of	ADP
fcis-31421	19	21	the	the	DET
fcis-31421	19	22	dynamic	dynamic	ADJ
fcis-31421	19	23	characteristics	characteristic	NOUN
fcis-31421	19	24	of	of	ADP
fcis-31421	19	25	the	the	DET
fcis-31421	19	26	fire	fire	NOUN
fcis-31421	19	27	and	and	CCONJ
fcis-31421	19	28	the	the	DET
fcis-31421	19	29	ability	ability	NOUN
fcis-31421	19	30	to	to	PART
fcis-31421	19	31	analyze	analyze	VERB
fcis-31421	19	32	intelligently	intelligently	ADV
fcis-31421	19	33	.	.	PUNCT
fcis-31421	20	1	with	with	ADP
fcis-31421	20	2	the	the	DET
fcis-31421	20	3	breakthrough	breakthrough	NOUN
fcis-31421	20	4	of	of	ADP
fcis-31421	20	5	computer	computer	NOUN
fcis-31421	20	6	vision	vision	NOUN
fcis-31421	20	7	and	and	CCONJ
fcis-31421	20	8	deep	deep	ADJ
fcis-31421	20	9	learning	learning	NOUN
fcis-31421	20	10	technology	technology	NOUN
fcis-31421	20	11	,	,	PUNCT
fcis-31421	20	12	fire	fire	NOUN
fcis-31421	20	13	image	image	NOUN
fcis-31421	20	14	detection	detection	NOUN
fcis-31421	20	15	technology	technology	NOUN
fcis-31421	20	16	based	base	VERB
fcis-31421	20	17	on	on	ADP
fcis-31421	20	18	large	large	ADJ
fcis-31421	20	19	models	model	NOUN
fcis-31421	20	20	has	have	AUX
fcis-31421	20	21	gradually	gradually	ADV
fcis-31421	20	22	become	become	VERB
fcis-31421	20	23	a	a	DET
fcis-31421	20	24	research	research	NOUN
fcis-31421	20	25	focus	focus	NOUN
fcis-31421	20	26	.	.	PUNCT
fcis-31421	21	1	early	early	ADJ
fcis-31421	21	2	research	research	NOUN
fcis-31421	21	3	focused	focus	VERB
fcis-31421	21	4	on	on	ADP
fcis-31421	21	5	flame	flame	ADJ
fcis-31421	21	6	color	color	NOUN
fcis-31421	21	7	model	model	NOUN
fcis-31421	21	8	and	and	CCONJ
fcis-31421	21	9	motion	motion	NOUN
fcis-31421	21	10	feature	feature	NOUN
fcis-31421	21	11	extraction	extraction	NOUN
fcis-31421	21	12	,	,	PUNCT
fcis-31421	21	13	but	but	CCONJ
fcis-31421	21	14	the	the	DET
fcis-31421	21	15	generalization	generalization	NOUN
fcis-31421	21	16	ability	ability	NOUN
fcis-31421	21	17	of	of	ADP
fcis-31421	21	18	artificially	artificially	ADV
fcis-31421	21	19	designed	design	VERB
fcis-31421	21	20	features	feature	NOUN
fcis-31421	21	21	is	be	AUX
fcis-31421	21	22	insufficient	insufficient	ADJ
fcis-31421	21	23	in	in	ADP
fcis-31421	21	24	complex	complex	ADJ
fcis-31421	21	25	lighting	lighting	NOUN
fcis-31421	21	26	and	and	CCONJ
fcis-31421	21	27	firelike	firelike	ADJ
fcis-31421	21	28	interfering	interfere	VERB
fcis-31421	21	29	object	object	NOUN
fcis-31421	21	30	scenes	scene	NOUN
fcis-31421	21	31	.	.	PUNCT
fcis-31421	22	1	compared	compare	VERB
fcis-31421	22	2	to	to	ADP
fcis-31421	22	3	deep	deep	ADJ
fcis-31421	22	4	learningbased	learningbase	VERB
fcis-31421	22	5	image	image	NOUN
fcis-31421	22	6	detection	detection	NOUN
fcis-31421	22	7	techniques	technique	NOUN
fcis-31421	22	8	,	,	PUNCT
fcis-31421	22	9	there	there	PRON
fcis-31421	22	10	is	be	VERB
fcis-31421	22	11	a	a	DET
fcis-31421	22	12	generation	generation	NOUN
fcis-31421	22	13	gap	gap	NOUN
fcis-31421	22	14	between	between	ADP
fcis-31421	22	15	traditional	traditional	ADJ
fcis-31421	22	16	fire	fire	NOUN
fcis-31421	22	17	detection	detection	NOUN
fcis-31421	22	18	methods	method	NOUN
fcis-31421	22	19	in	in	ADP
fcis-31421	22	20	terms	term	NOUN
fcis-31421	22	21	of	of	ADP
fcis-31421	22	22	sensing	sense	VERB
fcis-31421	22	23	dimension	dimension	NOUN
fcis-31421	22	24	,	,	PUNCT
fcis-31421	22	25	scene	scene	NOUN
fcis-31421	22	26	adaptability	adaptability	NOUN
fcis-31421	22	27	,	,	PUNCT
fcis-31421	22	28	and	and	CCONJ
fcis-31421	22	29	functionality	functionality	NOUN
fcis-31421	22	30	expansion	expansion	NOUN
fcis-31421	22	31	.	.	PUNCT
fcis-31421	23	1	first	first	ADV
fcis-31421	23	2	of	of	ADP
fcis-31421	23	3	all	all	PRON
fcis-31421	23	4	,	,	PUNCT
fcis-31421	23	5	traditional	traditional	ADJ
fcis-31421	23	6	sensors	sensor	NOUN
fcis-31421	23	7	can	can	AUX
fcis-31421	23	8	only	only	ADV
fcis-31421	23	9	provide	provide	VERB
fcis-31421	23	10	the	the	DET
fcis-31421	23	11	“	"	PUNCT
fcis-31421	23	12	presence	presence	NOUN
fcis-31421	23	13	or	or	CCONJ
fcis-31421	23	14	absence	absence	NOUN
fcis-31421	23	15	”	"	PUNCT
fcis-31421	23	16	judgment	judgment	NOUN
fcis-31421	23	17	of	of	ADP
fcis-31421	23	18	fire	fire	NOUN
fcis-31421	23	19	,	,	PUNCT
fcis-31421	23	20	while	while	SCONJ
fcis-31421	23	21	the	the	DET
fcis-31421	23	22	large	large	ADJ
fcis-31421	23	23	model	model	NOUN
fcis-31421	23	24	can	can	AUX
fcis-31421	23	25	extract	extract	VERB
fcis-31421	23	26	multi	multi	ADJ
fcis-31421	23	27	-	-	ADJ
fcis-31421	23	28	level	level	ADJ
fcis-31421	23	29	features	feature	NOUN
fcis-31421	23	30	such	such	ADJ
fcis-31421	23	31	as	as	ADP
fcis-31421	23	32	flame	flame	NOUN
fcis-31421	23	33	morphology	morphology	NOUN
fcis-31421	23	34	,	,	PUNCT
fcis-31421	23	35	smoke	smoke	NOUN
fcis-31421	23	36	diffusion	diffusion	NOUN
fcis-31421	23	37	dynamics	dynamic	NOUN
fcis-31421	23	38	,	,	PUNCT
fcis-31421	23	39	etc	etc	X
fcis-31421	23	40	.	.	X
fcis-31421	23	41	from	from	ADP
fcis-31421	23	42	the	the	DET
fcis-31421	23	43	image	image	NOUN
fcis-31421	23	44	to	to	PART
fcis-31421	23	45	realize	realize	VERB
fcis-31421	23	46	the	the	DET
fcis-31421	23	47	localization	localization	NOUN
fcis-31421	23	48	of	of	ADP
fcis-31421	23	49	the	the	DET
fcis-31421	23	50	fire	fire	NOUN
fcis-31421	23	51	source	source	NOUN
fcis-31421	23	52	,	,	PUNCT
fcis-31421	23	53	fire	fire	NOUN
fcis-31421	23	54	grading	grading	NOUN
fcis-31421	23	55	and	and	CCONJ
fcis-31421	23	56	spread	spread	ADJ
fcis-31421	23	57	prediction	prediction	NOUN
fcis-31421	23	58	.	.	PUNCT
fcis-31421	24	1	secondly	secondly	ADV
fcis-31421	24	2	,	,	PUNCT
fcis-31421	24	3	the	the	DET
fcis-31421	24	4	traditional	traditional	ADJ
fcis-31421	24	5	methods	method	NOUN
fcis-31421	24	6	in	in	ADP
fcis-31421	24	7	complex	complex	ADJ
fcis-31421	24	8	environments	environment	NOUN
fcis-31421	24	9	with	with	ADP
fcis-31421	24	10	high	high	ADJ
fcis-31421	24	11	risk	risk	NOUN
fcis-31421	24	12	of	of	ADP
fcis-31421	24	13	failure	failure	NOUN
fcis-31421	24	14	,	,	PUNCT
fcis-31421	24	15	such	such	ADJ
fcis-31421	24	16	as	as	ADP
fcis-31421	24	17	haze	haze	NOUN
fcis-31421	24	18	obstruction	obstruction	NOUN
fcis-31421	24	19	,	,	PUNCT
fcis-31421	24	20	insufficient	insufficient	ADJ
fcis-31421	24	21	light	light	NOUN
fcis-31421	24	22	at	at	ADP
fcis-31421	24	23	night	night	NOUN
fcis-31421	24	24	,	,	PUNCT
fcis-31421	24	25	and	and	CCONJ
fcis-31421	24	26	multimodal	multimodal	NOUN
fcis-31421	24	27	model	model	NOUN
fcis-31421	24	28	through	through	ADP
fcis-31421	24	29	the	the	DET
fcis-31421	24	30	fusion	fusion	NOUN
fcis-31421	24	31	of	of	ADP
fcis-31421	24	32	visible	visible	ADJ
fcis-31421	24	33	light	light	NOUN
fcis-31421	24	34	,	,	PUNCT
fcis-31421	24	35	infrared	infrared	ADJ
fcis-31421	24	36	thermal	thermal	ADJ
fcis-31421	24	37	imaging	imaging	NOUN
fcis-31421	24	38	and	and	CCONJ
fcis-31421	24	39	sensor	sensor	NOUN
fcis-31421	24	40	data	datum	NOUN
fcis-31421	24	41	,	,	PUNCT
fcis-31421	24	42	breaking	break	VERB
fcis-31421	24	43	the	the	DET
fcis-31421	24	44	perception	perception	NOUN
fcis-31421	24	45	bottleneck	bottleneck	NOUN
fcis-31421	24	46	of	of	ADP
fcis-31421	24	47	a	a	DET
fcis-31421	24	48	single	single	ADJ
fcis-31421	24	49	signal	signal	NOUN
fcis-31421	24	50	source	source	NOUN
fcis-31421	24	51	,	,	PUNCT
fcis-31421	24	52	significantly	significantly	ADV
fcis-31421	24	53	improving	improve	VERB
fcis-31421	24	54	the	the	DET
fcis-31421	24	55	robustness	robustness	NOUN
fcis-31421	24	56	.	.	PUNCT
fcis-31421	25	1	in	in	ADP
fcis-31421	25	2	addition	addition	NOUN
fcis-31421	25	3	,	,	PUNCT
fcis-31421	25	4	the	the	DET
fcis-31421	25	5	traditional	traditional	ADJ
fcis-31421	25	6	technology	technology	NOUN
fcis-31421	25	7	is	be	AUX
fcis-31421	25	8	difficult	difficult	ADJ
fcis-31421	25	9	to	to	PART
fcis-31421	25	10	cover	cover	VERB
fcis-31421	25	11	a	a	DET
fcis-31421	25	12	wide	wide	ADJ
fcis-31421	25	13	range	range	NOUN
fcis-31421	25	14	of	of	ADP
fcis-31421	25	15	scenes	scene	NOUN
fcis-31421	25	16	,	,	PUNCT
fcis-31421	25	17	while	while	SCONJ
fcis-31421	25	18	the	the	DET
fcis-31421	25	19	uav	uav	PROPN
fcis-31421	25	20	equipped	equip	VERB
fcis-31421	25	21	with	with	ADP
fcis-31421	25	22	lightweight	lightweight	ADJ
fcis-31421	25	23	yolo	yolo	ADJ
fcis-31421	25	24	model	model	NOUN
fcis-31421	25	25	can	can	AUX
fcis-31421	25	26	carry	carry	VERB
fcis-31421	25	27	out	out	ADP
fcis-31421	25	28	aerial	aerial	ADJ
fcis-31421	25	29	inspection	inspection	NOUN
fcis-31421	25	30	,	,	PUNCT
fcis-31421	25	31	and	and	CCONJ
fcis-31421	25	32	the	the	DET
fcis-31421	25	33	detection	detection	NOUN
fcis-31421	25	34	efficiency	efficiency	NOUN
fcis-31421	25	35	is	be	AUX
fcis-31421	25	36	improved	improve	VERB
fcis-31421	25	37	by	by	ADP
fcis-31421	25	38	tens	ten	NOUN
fcis-31421	25	39	of	of	ADP
fcis-31421	25	40	times	time	NOUN
fcis-31421	25	41	.	.	PUNCT
fcis-31421	26	1	what	what	PRON
fcis-31421	26	2	's	be	AUX
fcis-31421	26	3	more	more	ADJ
fcis-31421	26	4	,	,	PUNCT
fcis-31421	26	5	the	the	DET
fcis-31421	26	6	false	false	ADJ
fcis-31421	26	7	alarm	alarm	NOUN
fcis-31421	26	8	rate	rate	NOUN
fcis-31421	26	9	of	of	ADP
fcis-31421	26	10	traditional	traditional	ADJ
fcis-31421	26	11	methods	method	NOUN
fcis-31421	26	12	is	be	AUX
fcis-31421	26	13	generally	generally	ADV
fcis-31421	26	14	higher	high	ADJ
fcis-31421	26	15	than	than	ADP
fcis-31421	26	16	10	10	NUM
fcis-31421	26	17	%	%	NOUN
fcis-31421	26	18	,	,	PUNCT
fcis-31421	26	19	while	while	SCONJ
fcis-31421	26	20	deep	deep	ADJ
fcis-31421	26	21	learning	learning	NOUN
fcis-31421	26	22	-	-	PUNCT
fcis-31421	26	23	based	base	VERB
fcis-31421	26	24	target	target	NOUN
fcis-31421	26	25	detection	detection	NOUN
fcis-31421	26	26	algorithms	algorithm	NOUN
fcis-31421	26	27	can	can	AUX
fcis-31421	26	28	control	control	VERB
fcis-31421	26	29	the	the	DET
fcis-31421	26	30	false	false	ADJ
fcis-31421	26	31	alarm	alarm	NOUN
fcis-31421	26	32	rate	rate	NOUN
fcis-31421	26	33	below	below	ADP
fcis-31421	26	34	5	5	NUM
fcis-31421	26	35	%	%	NOUN
fcis-31421	26	36	through	through	ADP
fcis-31421	26	37	end	end	NOUN
fcis-31421	26	38	-	-	PUNCT
fcis-31421	26	39	to	to	ADP
fcis-31421	26	40	-	-	PUNCT
fcis-31421	26	41	end	end	NOUN
fcis-31421	26	42	feature	feature	NOUN
fcis-31421	26	43	learning	learning	NOUN
fcis-31421	26	44	,	,	PUNCT
fcis-31421	26	45	while	while	SCONJ
fcis-31421	26	46	supporting	support	VERB
fcis-31421	26	47	early	early	ADJ
fcis-31421	26	48	fire	fire	NOUN
fcis-31421	26	49	warning	warning	NOUN
fcis-31421	26	50	.	.	PUNCT
fcis-31421	27	1	these	these	DET
fcis-31421	27	2	gaps	gap	NOUN
fcis-31421	27	3	highlight	highlight	VERB
fcis-31421	27	4	the	the	DET
fcis-31421	27	5	inevitability	inevitability	NOUN
fcis-31421	27	6	of	of	ADP
fcis-31421	27	7	big	big	ADJ
fcis-31421	27	8	model	model	NOUN
fcis-31421	27	9	technology	technology	NOUN
fcis-31421	27	10	to	to	PART
fcis-31421	27	11	promote	promote	VERB
fcis-31421	27	12	the	the	DET
fcis-31421	27	13	transformation	transformation	NOUN
fcis-31421	27	14	of	of	ADP
fcis-31421	27	15	fire	fire	NOUN
fcis-31421	27	16	detection	detection	NOUN
fcis-31421	27	17	from	from	ADP
fcis-31421	27	18	“	"	PUNCT
fcis-31421	27	19	single	single	ADJ
fcis-31421	27	20	threshold	threshold	NOUN
fcis-31421	27	21	alarm	alarm	NOUN
fcis-31421	27	22	”	"	PUNCT
fcis-31421	27	23	to	to	ADP
fcis-31421	27	24	“	"	PUNCT
fcis-31421	27	25	intelligent	intelligent	ADJ
fcis-31421	27	26	active	active	ADJ
fcis-31421	27	27	prevention	prevention	NOUN
fcis-31421	27	28	and	and	CCONJ
fcis-31421	27	29	control	control	NOUN
fcis-31421	27	30	”	"	PUNCT
fcis-31421	27	31	.	.	PUNCT
fcis-31421	28	1	in	in	ADP
fcis-31421	28	2	recent	recent	ADJ
fcis-31421	28	3	years	year	NOUN
fcis-31421	28	4	,	,	PUNCT
fcis-31421	28	5	the	the	DET
fcis-31421	28	6	development	development	NOUN
fcis-31421	28	7	of	of	ADP
fcis-31421	28	8	machine	machine	NOUN
fcis-31421	28	9	learning	learning	NOUN
fcis-31421	28	10	and	and	CCONJ
fcis-31421	28	11	deep	deep	ADJ
fcis-31421	28	12	learning	learning	NOUN
fcis-31421	28	13	has	have	AUX
fcis-31421	28	14	completely	completely	ADV
fcis-31421	28	15	changed	change	VERB
fcis-31421	28	16	the	the	DET
fcis-31421	28	17	field	field	NOUN
fcis-31421	28	18	of	of	ADP
fcis-31421	28	19	fire	fire	NOUN
fcis-31421	28	20	detection	detection	NOUN
fcis-31421	28	21	.	.	PUNCT
fcis-31421	29	1	the	the	DET
fcis-31421	29	2	model	model	NOUN
fcis-31421	29	3	based	base	VERB
fcis-31421	29	4	on	on	ADP
fcis-31421	29	5	convolutional	convolutional	ADJ
fcis-31421	29	6	neural	neural	ADJ
fcis-31421	29	7	network	network	NOUN
fcis-31421	29	8	(	(	PUNCT
fcis-31421	29	9	cnn	cnn	PROPN
fcis-31421	29	10	)	)	PUNCT
fcis-31421	29	11	performs	perform	VERB
fcis-31421	29	12	very	very	ADV
fcis-31421	29	13	well	well	ADV
fcis-31421	29	14	in	in	ADP
fcis-31421	29	15	image	image	NOUN
fcis-31421	29	16	classification	classification	NOUN
fcis-31421	29	17	and	and	CCONJ
fcis-31421	29	18	target	target	NOUN
fcis-31421	29	19	detection	detection	NOUN
fcis-31421	29	20	tasks	task	NOUN
fcis-31421	29	21	,	,	PUNCT
fcis-31421	29	22	while	while	SCONJ
fcis-31421	29	23	the	the	DET
fcis-31421	29	24	single	single	ADJ
fcis-31421	29	25	-	-	PUNCT
fcis-31421	29	26	stage	stage	NOUN
fcis-31421	29	27	target	target	NOUN
fcis-31421	29	28	detection	detection	NOUN
fcis-31421	29	29	framework	framework	NOUN
fcis-31421	29	30	yolo	yolo	ADJ
fcis-31421	29	31	(	(	PUNCT
fcis-31421	29	32	you	you	PRON
fcis-31421	29	33	only	only	ADV
fcis-31421	29	34	look	look	VERB
fcis-31421	29	35	once	once	ADV
fcis-31421	29	36	)	)	PUNCT
fcis-31421	29	37	has	have	AUX
fcis-31421	29	38	become	become	VERB
fcis-31421	29	39	an	an	DET
fcis-31421	29	40	important	important	ADJ
fcis-31421	29	41	tool	tool	NOUN
fcis-31421	29	42	for	for	ADP
fcis-31421	29	43	fire	fire	NOUN
fcis-31421	29	44	detection	detection	NOUN
fcis-31421	29	45	in	in	ADP
fcis-31421	29	46	dynamic	dynamic	ADJ
fcis-31421	29	47	scenarios	scenario	NOUN
fcis-31421	29	48	by	by	ADP
fcis-31421	29	49	virtue	virtue	NOUN
fcis-31421	29	50	of	of	ADP
fcis-31421	29	51	its	its	PRON
fcis-31421	29	52	high	high	ADJ
fcis-31421	29	53	efficiency	efficiency	NOUN
fcis-31421	29	54	and	and	CCONJ
fcis-31421	29	55	real	real	ADJ
fcis-31421	29	56	-	-	PUNCT
fcis-31421	29	57	time	time	NOUN
fcis-31421	29	58	performance	performance	NOUN
fcis-31421	29	59	.	.	PUNCT
fcis-31421	30	1	meanwhile	meanwhile	ADV
fcis-31421	30	2	,	,	PUNCT
fcis-31421	30	3	the	the	DET
fcis-31421	30	4	deep	deep	ADJ
fcis-31421	30	5	learning	learning	NOUN
fcis-31421	30	6	method	method	NOUN
fcis-31421	30	7	based	base	VERB
fcis-31421	30	8	on	on	ADP
fcis-31421	30	9	multi	multi	ADJ
fcis-31421	30	10	-	-	ADJ
fcis-31421	30	11	source	source	NOUN
fcis-31421	30	12	information	information	NOUN
fcis-31421	30	13	fusion	fusion	NOUN
fcis-31421	30	14	significantly	significantly	ADV
fcis-31421	30	15	improves	improve	VERB
fcis-31421	30	16	the	the	DET
fcis-31421	30	17	detection	detection	NOUN
fcis-31421	30	18	accuracy	accuracy	NOUN
fcis-31421	30	19	and	and	CCONJ
fcis-31421	30	20	environmental	environmental	ADJ
fcis-31421	30	21	adaptability	adaptability	NOUN
fcis-31421	30	22	of	of	ADP
fcis-31421	30	23	fire	fire	NOUN
fcis-31421	30	24	monitoring	monitoring	NOUN
fcis-31421	30	25	systems	system	NOUN
fcis-31421	30	26	by	by	ADP
fcis-31421	30	27	integrating	integrate	VERB
fcis-31421	30	28	multimodal	multimodal	NOUN
fcis-31421	30	29	data	datum	NOUN
fcis-31421	30	30	such	such	ADJ
fcis-31421	30	31	as	as	ADP
fcis-31421	30	32	visual	visual	ADJ
fcis-31421	30	33	images	image	NOUN
fcis-31421	30	34	,	,	PUNCT
fcis-31421	30	35	infrared	infrared	ADJ
fcis-31421	30	36	thermal	thermal	ADJ
fcis-31421	30	37	imaging	imaging	NOUN
fcis-31421	30	38	,	,	PUNCT
fcis-31421	30	39	temperature	temperature	NOUN
fcis-31421	30	40	sensing	sensing	NOUN
fcis-31421	30	41	,	,	PUNCT
fcis-31421	30	42	and	and	CCONJ
fcis-31421	30	43	smoke	smoke	NOUN
fcis-31421	30	44	concentration	concentration	NOUN
fcis-31421	30	45	.	.	PUNCT
fcis-31421	31	1	this	this	DET
fcis-31421	31	2	type	type	NOUN
fcis-31421	31	3	of	of	ADP
fcis-31421	31	4	advanced	advanced	ADJ
fcis-31421	31	5	model	model	NOUN
fcis-31421	31	6	not	not	PART
fcis-31421	31	7	only	only	ADV
fcis-31421	31	8	effectively	effectively	ADV
fcis-31421	31	9	solves	solve	VERB
fcis-31421	31	10	the	the	DET
fcis-31421	31	11	inherent	inherent	ADJ
fcis-31421	31	12	defects	defect	NOUN
fcis-31421	31	13	of	of	ADP
fcis-31421	31	14	traditional	traditional	ADJ
fcis-31421	31	15	detection	detection	NOUN
fcis-31421	31	16	technology	technology	NOUN
fcis-31421	31	17	but	but	CCONJ
fcis-31421	31	18	also	also	ADV
fcis-31421	31	19	opens	open	VERB
fcis-31421	31	20	innovative	innovative	ADJ
fcis-31421	31	21	technological	technological	ADJ
fcis-31421	31	22	paths	path	NOUN
fcis-31421	31	23	for	for	ADP
fcis-31421	31	24	intelligent	intelligent	ADJ
fcis-31421	31	25	fire	fire	NOUN
fcis-31421	31	26	warning	warning	NOUN
fcis-31421	31	27	in	in	ADP
fcis-31421	31	28	complex	complex	ADJ
fcis-31421	31	29	environments	environment	NOUN
fcis-31421	31	30	.	.	PUNCT
fcis-31421	32	1	cnn	cnn	PROPN
fcis-31421	32	2	,	,	PUNCT
fcis-31421	32	3	multimodal	multimodal	NOUN
fcis-31421	32	4	models	model	NOUN
fcis-31421	32	5	and	and	CCONJ
fcis-31421	32	6	yolo	yolo	ADJ
fcis-31421	32	7	series	series	NOUN
fcis-31421	32	8	models	model	NOUN
fcis-31421	32	9	each	each	PRON
fcis-31421	32	10	have	have	VERB
fcis-31421	32	11	their	their	PRON
fcis-31421	32	12	own	own	ADJ
fcis-31421	32	13	advantages	advantage	NOUN
fcis-31421	32	14	and	and	CCONJ
fcis-31421	32	15	limitations	limitation	NOUN
fcis-31421	32	16	in	in	ADP
fcis-31421	32	17	fire	fire	NOUN
fcis-31421	32	18	detection	detection	NOUN
fcis-31421	32	19	.	.	PUNCT
fcis-31421	33	1	cnn	cnn	PROPN
fcis-31421	33	2	is	be	AUX
fcis-31421	33	3	mainly	mainly	ADV
fcis-31421	33	4	used	use	VERB
fcis-31421	33	5	for	for	ADP
fcis-31421	33	6	tasks	task	NOUN
fcis-31421	33	7	such	such	ADJ
fcis-31421	33	8	as	as	ADP
fcis-31421	33	9	static	static	ADJ
fcis-31421	33	10	image	image	NOUN
fcis-31421	33	11	analysis	analysis	NOUN
fcis-31421	33	12	,	,	PUNCT
fcis-31421	33	13	extracts	extract	NOUN
fcis-31421	33	14	flame	flame	NOUN
fcis-31421	33	15	features	feature	VERB
fcis-31421	33	16	through	through	ADP
fcis-31421	33	17	multilayer	multilayer	ADJ
fcis-31421	33	18	convolution	convolution	NOUN
fcis-31421	33	19	,	,	PUNCT
fcis-31421	33	20	and	and	CCONJ
fcis-31421	33	21	performs	perform	VERB
fcis-31421	33	22	well	well	ADV
fcis-31421	33	23	in	in	ADP
fcis-31421	33	24	smoke	smoke	NOUN
fcis-31421	33	25	recognition	recognition	NOUN
fcis-31421	33	26	,	,	PUNCT
fcis-31421	33	27	etc	etc	X
fcis-31421	33	28	.	.	X
fcis-31421	34	1	however	however	ADV
fcis-31421	34	2	,	,	PUNCT
fcis-31421	34	3	it	it	PRON
fcis-31421	34	4	has	have	VERB
fcis-31421	34	5	high	high	ADJ
fcis-31421	34	6	computational	computational	ADJ
fcis-31421	34	7	complexity	complexity	NOUN
fcis-31421	34	8	,	,	PUNCT
fcis-31421	34	9	is	be	AUX
fcis-31421	34	10	weak	weak	ADJ
fcis-31421	34	11	in	in	ADP
fcis-31421	34	12	modeling	model	VERB
fcis-31421	34	13	dynamic	dynamic	ADJ
fcis-31421	34	14	flames	flame	NOUN
fcis-31421	34	15	and	and	CCONJ
fcis-31421	34	16	is	be	AUX
fcis-31421	34	17	prone	prone	ADJ
fcis-31421	34	18	to	to	ADP
fcis-31421	34	19	overfitting	overfitte	VERB
fcis-31421	34	20	,	,	PUNCT
fcis-31421	34	21	but	but	CCONJ
fcis-31421	34	22	it	it	PRON
fcis-31421	34	23	is	be	AUX
fcis-31421	34	24	still	still	ADV
fcis-31421	34	25	the	the	DET
fcis-31421	34	26	basic	basic	ADJ
fcis-31421	34	27	architecture	architecture	NOUN
fcis-31421	34	28	for	for	ADP
fcis-31421	34	29	fire	fire	NOUN
fcis-31421	34	30	feature	feature	NOUN
fcis-31421	34	31	extraction	extraction	NOUN
fcis-31421	34	32	.	.	PUNCT
fcis-31421	35	1	multi	multi	ADJ
fcis-31421	35	2	-	-	ADJ
fcis-31421	35	3	modal	modal	ADJ
fcis-31421	35	4	model	model	NOUN
fcis-31421	35	5	fusion	fusion	NOUN
fcis-31421	35	6	of	of	ADP
fcis-31421	35	7	multi	multi	ADJ
fcis-31421	35	8	-	-	ADJ
fcis-31421	35	9	source	source	NOUN
fcis-31421	35	10	data	datum	NOUN
fcis-31421	35	11	to	to	PART
fcis-31421	35	12	improve	improve	VERB
fcis-31421	35	13	the	the	DET
fcis-31421	35	14	robustness	robustness	NOUN
fcis-31421	35	15	of	of	ADP
fcis-31421	35	16	fire	fire	NOUN
fcis-31421	35	17	detection	detection	NOUN
fcis-31421	35	18	in	in	ADP
fcis-31421	35	19	extreme	extreme	ADJ
fcis-31421	35	20	environments	environment	NOUN
fcis-31421	35	21	,	,	PUNCT
fcis-31421	35	22	using	use	VERB
fcis-31421	35	23	cross	cross	ADJ
fcis-31421	35	24	-	-	ADJ
fcis-31421	35	25	modal	modal	ADJ
fcis-31421	35	26	information	information	NOUN
fcis-31421	35	27	complementary	complementary	ADJ
fcis-31421	35	28	to	to	PART
fcis-31421	35	29	reduce	reduce	VERB
fcis-31421	35	30	the	the	DET
fcis-31421	35	31	false	false	ADJ
fcis-31421	35	32	alarm	alarm	NOUN
fcis-31421	35	33	rate	rate	NOUN
fcis-31421	35	34	,	,	PUNCT
fcis-31421	35	35	however	however	ADV
fcis-31421	35	36	,	,	PUNCT
fcis-31421	35	37	computationally	computationally	ADV
fcis-31421	35	38	complex	complex	ADJ
fcis-31421	35	39	,	,	PUNCT
fcis-31421	35	40	there	there	PRON
fcis-31421	35	41	are	be	VERB
fcis-31421	35	42	spatial	spatial	ADJ
fcis-31421	35	43	and	and	CCONJ
fcis-31421	35	44	temporal	temporal	ADJ
fcis-31421	35	45	alignment	alignment	NOUN
fcis-31421	35	46	,	,	PUNCT
fcis-31421	35	47	high	high	ADJ
fcis-31421	35	48	cost	cost	NOUN
fcis-31421	35	49	of	of	ADP
fcis-31421	35	50	annotation	annotation	NOUN
fcis-31421	35	51	and	and	CCONJ
fcis-31421	35	52	high	high	ADJ
fcis-31421	35	53	hardware	hardware	NOUN
fcis-31421	35	54	resource	resource	NOUN
fcis-31421	35	55	requirements	requirement	NOUN
fcis-31421	35	56	,	,	PUNCT
fcis-31421	35	57	the	the	DET
fcis-31421	35	58	future	future	ADJ
fcis-31421	35	59	focus	focus	NOUN
fcis-31421	35	60	on	on	ADP
fcis-31421	35	61	lightweight	lightweight	ADJ
fcis-31421	35	62	fusion	fusion	NOUN
fcis-31421	35	63	and	and	CCONJ
fcis-31421	35	64	self	self	NOUN
fcis-31421	35	65	-	-	PUNCT
fcis-31421	35	66	supervised	supervise	VERB
fcis-31421	35	67	learning	learning	NOUN
fcis-31421	35	68	.	.	PUNCT
fcis-31421	36	1	yolo	yolo	PROPN
fcis-31421	36	2	series	series	PROPN
fcis-31421	36	3	models	model	NOUN
fcis-31421	36	4	to	to	ADP
fcis-31421	36	5	single	single	ADJ
fcis-31421	36	6	-	-	PUNCT
fcis-31421	36	7	stage	stage	NOUN
fcis-31421	36	8	detection	detection	NOUN
fcis-31421	36	9	framework	framework	NOUN
fcis-31421	36	10	of	of	ADP
fcis-31421	36	11	high	high	ADJ
fcis-31421	36	12	efficiency	efficiency	NOUN
fcis-31421	36	13	,	,	PUNCT
fcis-31421	36	14	suitable	suitable	ADJ
fcis-31421	36	15	for	for	ADP
fcis-31421	36	16	real	real	ADJ
fcis-31421	36	17	-	-	PUNCT
fcis-31421	36	18	time	time	NOUN
fcis-31421	36	19	video	video	NOUN
fcis-31421	36	20	surveillance	surveillance	NOUN
fcis-31421	36	21	and	and	CCONJ
fcis-31421	36	22	mobile	mobile	ADJ
fcis-31421	36	23	deployment	deployment	NOUN
fcis-31421	36	24	,	,	PUNCT
fcis-31421	36	25	can	can	AUX
fcis-31421	36	26	synchronize	synchronize	VERB
fcis-31421	36	27	the	the	PRON
fcis-31421	36	28	it	it	PRON
fcis-31421	36	29	can	can	AUX
fcis-31421	36	30	detect	detect	VERB
fcis-31421	36	31	multiple	multiple	ADJ
fcis-31421	36	32	types	type	NOUN
fcis-31421	36	33	of	of	ADP
fcis-31421	36	34	targets	target	NOUN
fcis-31421	36	35	and	and	CCONJ
fcis-31421	36	36	support	support	VERB
fcis-31421	36	37	real	real	ADJ
fcis-31421	36	38	-	-	PUNCT
fcis-31421	36	39	time	time	NOUN
fcis-31421	36	40	processing	processing	NOUN
fcis-31421	36	41	,	,	PUNCT
fcis-31421	36	42	this	this	DET
fcis-31421	36	43	article	article	NOUN
fcis-31421	36	44	used	use	VERB
fcis-31421	36	45	the	the	DET
fcis-31421	36	46	dataset	dataset	NOUN
fcis-31421	36	47	of	of	ADP
fcis-31421	36	48	fire	fire	NOUN
fcis-31421	36	49	smoke	smoke	NOUN
fcis-31421	36	50	detection	detection	NOUN
fcis-31421	36	51	to	to	PART
fcis-31421	36	52	train	train	VERB
fcis-31421	36	53	the	the	DET
fcis-31421	36	54	relevant	relevant	ADJ
fcis-31421	36	55	modules	module	NOUN
fcis-31421	36	56	of	of	ADP
fcis-31421	36	57	yolov8	yolov8	NOUN
fcis-31421	36	58	and	and	CCONJ
fcis-31421	36	59	achieved	achieve	VERB
fcis-31421	36	60	good	good	ADJ
fcis-31421	36	61	results	result	NOUN
fcis-31421	36	62	,	,	PUNCT
fcis-31421	36	63	but	but	CCONJ
fcis-31421	36	64	it	it	PRON
fcis-31421	36	65	has	have	VERB
fcis-31421	36	66	a	a	DET
fcis-31421	36	67	high	high	ADJ
fcis-31421	36	68	leakage	leakage	NOUN
fcis-31421	36	69	detection	detection	NOUN
fcis-31421	36	70	rate	rate	NOUN
fcis-31421	36	71	for	for	ADP
fcis-31421	36	72	small	small	ADJ
fcis-31421	36	73	-	-	PUNCT
fcis-31421	36	74	scale	scale	NOUN
fcis-31421	36	75	flames	flame	NOUN
fcis-31421	36	76	and	and	CCONJ
fcis-31421	36	77	sparse	sparse	ADJ
fcis-31421	36	78	smoke	smoke	NOUN
fcis-31421	36	79	,	,	PUNCT
fcis-31421	36	80	limited	limited	ADJ
fcis-31421	36	81	generalization	generalization	NOUN
fcis-31421	36	82	ability	ability	NOUN
fcis-31421	36	83	,	,	PUNCT
fcis-31421	36	84	and	and	CCONJ
fcis-31421	36	85	is	be	AUX
fcis-31421	36	86	still	still	ADV
fcis-31421	36	87	affected	affect	VERB
fcis-31421	36	88	by	by	ADP
fcis-31421	36	89	target	target	NOUN
fcis-31421	36	90	size	size	NOUN
fcis-31421	36	91	and	and	CCONJ
fcis-31421	36	92	data	datum	NOUN
fcis-31421	36	93	distribution	distribution	NOUN
fcis-31421	36	94	despite	despite	SCONJ
fcis-31421	36	95	improvements	improvement	NOUN
fcis-31421	36	96	.	.	PUNCT
fcis-31421	37	1	in	in	ADP
fcis-31421	37	2	summary	summary	NOUN
fcis-31421	37	3	,	,	PUNCT
fcis-31421	37	4	the	the	DET
fcis-31421	37	5	above	above	ADJ
fcis-31421	37	6	three	three	NUM
fcis-31421	37	7	provide	provide	VERB
fcis-31421	37	8	a	a	DET
fcis-31421	37	9	theoretical	theoretical	ADJ
fcis-31421	37	10	basis	basis	NOUN
fcis-31421	37	11	for	for	ADP
fcis-31421	37	12	low	low	ADJ
fcis-31421	37	13	-	-	PUNCT
fcis-31421	37	14	latency	latency	NOUN
fcis-31421	37	15	large	large	ADJ
fcis-31421	37	16	-	-	PUNCT
fcis-31421	37	17	scale	scale	NOUN
fcis-31421	37	18	fire	fire	NOUN
fcis-31421	37	19	detection	detection	NOUN
fcis-31421	37	20	and	and	CCONJ
fcis-31421	37	21	provide	provide	VERB
fcis-31421	37	22	a	a	DET
fcis-31421	37	23	reference	reference	NOUN
fcis-31421	37	24	direction	direction	NOUN
fcis-31421	37	25	for	for	SCONJ
fcis-31421	37	26	scholars	scholar	NOUN
fcis-31421	37	27	to	to	PART
fcis-31421	37	28	improve	improve	VERB
fcis-31421	37	29	the	the	DET
fcis-31421	37	30	training	training	NOUN
fcis-31421	37	31	of	of	ADP
fcis-31421	37	32	the	the	DET
fcis-31421	37	33	model	model	NOUN
fcis-31421	37	34	when	when	SCONJ
fcis-31421	37	35	doing	do	VERB
fcis-31421	37	36	related	relate	VERB
fcis-31421	37	37	research	research	NOUN
fcis-31421	37	38	,	,	PUNCT
fcis-31421	37	39	and	and	CCONJ
fcis-31421	37	40	also	also	ADV
fcis-31421	37	41	provide	provide	VERB
fcis-31421	37	42	them	they	PRON
fcis-31421	37	43	with	with	ADP
fcis-31421	37	44	a	a	DET
fcis-31421	37	45	part	part	NOUN
fcis-31421	37	46	of	of	ADP
fcis-31421	37	47	the	the	DET
fcis-31421	37	48	examples	example	NOUN
fcis-31421	37	49	of	of	ADP
fcis-31421	37	50	previous	previous	ADJ
fcis-31421	37	51	scholars	scholar	NOUN
fcis-31421	37	52	as	as	ADP
fcis-31421	37	53	a	a	DET
fcis-31421	37	54	reference	reference	NOUN
fcis-31421	37	55	.	.	PUNCT
fcis-31421	38	1	2	2	X
fcis-31421	38	2	.	.	X
fcis-31421	38	3	related	relate	VERB
fcis-31421	38	4	work	work	NOUN
fcis-31421	38	5	2.1	2.1	NUM
fcis-31421	38	6	.	.	PUNCT
fcis-31421	39	1	cnn	cnn	PROPN
fcis-31421	39	2	convolutional	convolutional	ADJ
fcis-31421	39	3	neural	neural	ADJ
fcis-31421	39	4	network	network	NOUN
fcis-31421	39	5	(	(	PUNCT
fcis-31421	39	6	cnn	cnn	PROPN
fcis-31421	39	7	)	)	PUNCT
fcis-31421	39	8	is	be	AUX
fcis-31421	39	9	a	a	DET
fcis-31421	39	10	deep	deep	ADJ
fcis-31421	39	11	learning	learning	NOUN
fcis-31421	39	12	model	model	NOUN
fcis-31421	39	13	designed	design	VERB
fcis-31421	39	14	for	for	ADP
fcis-31421	39	15	processing	process	VERB
fcis-31421	39	16	grid	grid	NOUN
fcis-31421	39	17	-	-	PUNCT
fcis-31421	39	18	like	like	ADJ
fcis-31421	39	19	data	datum	NOUN
fcis-31421	39	20	,	,	PUNCT
fcis-31421	39	21	the	the	DET
fcis-31421	39	22	core	core	NOUN
fcis-31421	39	23	idea	idea	NOUN
fcis-31421	39	24	is	be	AUX
fcis-31421	39	25	to	to	PART
fcis-31421	39	26	extract	extract	VERB
fcis-31421	39	27	spatial	spatial	ADJ
fcis-31421	39	28	features	feature	NOUN
fcis-31421	39	29	through	through	ADP
fcis-31421	39	30	local	local	ADJ
fcis-31421	39	31	perception	perception	NOUN
fcis-31421	39	32	and	and	CCONJ
fcis-31421	39	33	weight	weight	NOUN
fcis-31421	39	34	sharing	sharing	NOUN
fcis-31421	39	35	,	,	PUNCT
fcis-31421	39	36	which	which	PRON
fcis-31421	39	37	mainly	mainly	ADV
fcis-31421	39	38	contains	contain	VERB
fcis-31421	39	39	the	the	DET
fcis-31421	39	40	following	follow	VERB
fcis-31421	39	41	components	component	NOUN
fcis-31421	39	42	.	.	PUNCT
fcis-31421	40	1	convolutional	convolutional	ADJ
fcis-31421	40	2	layer	layer	NOUN
fcis-31421	40	3	:	:	PUNCT
fcis-31421	40	4	sliding	slide	VERB
fcis-31421	40	5	scanning	scanning	NOUN
fcis-31421	40	6	of	of	ADP
fcis-31421	40	7	input	input	NOUN
fcis-31421	40	8	image	image	NOUN
fcis-31421	40	9	using	use	VERB
fcis-31421	40	10	convolutional	convolutional	ADJ
fcis-31421	40	11	kernel	kernel	NOUN
fcis-31421	40	12	to	to	PART
fcis-31421	40	13	extract	extract	VERB
fcis-31421	40	14	local	local	ADJ
fcis-31421	40	15	features	feature	NOUN
fcis-31421	40	16	such	such	ADJ
fcis-31421	40	17	as	as	ADP
fcis-31421	40	18	edges	edge	NOUN
fcis-31421	40	19	,	,	PUNCT
fcis-31421	40	20	texture	texture	NOUN
fcis-31421	40	21	etc	etc	X
fcis-31421	40	22	.	.	X
fcis-31421	40	23	reduces	reduce	VERB
fcis-31421	40	24	the	the	DET
fcis-31421	40	25	number	number	NOUN
fcis-31421	40	26	of	of	ADP
fcis-31421	40	27	parameters	parameter	NOUN
fcis-31421	40	28	and	and	CCONJ
fcis-31421	40	29	preserves	preserve	VERB
fcis-31421	40	30	spatial	spatial	ADJ
fcis-31421	40	31	information	information	NOUN
fcis-31421	40	32	through	through	ADP
fcis-31421	40	33	local	local	ADJ
fcis-31421	40	34	concatenation	concatenation	NOUN
fcis-31421	40	35	,	,	PUNCT
fcis-31421	40	36	suitable	suitable	ADJ
fcis-31421	40	37	for	for	ADP
fcis-31421	40	38	image	image	NOUN
fcis-31421	40	39	translation	translation	NOUN
fcis-31421	40	40	invariance	invariance	NOUN
fcis-31421	40	41	.	.	PUNCT
fcis-31421	41	1	pooling	pool	VERB
fcis-31421	41	2	layer	layer	NOUN
fcis-31421	41	3	:	:	PUNCT
fcis-31421	41	4	downsampling	downsample	VERB
fcis-31421	41	5	the	the	DET
fcis-31421	41	6	feature	feature	NOUN
fcis-31421	41	7	map	map	NOUN
fcis-31421	41	8	to	to	PART
fcis-31421	41	9	reduce	reduce	VERB
fcis-31421	41	10	computational	computational	ADJ
fcis-31421	41	11	complexity	complexity	NOUN
fcis-31421	41	12	and	and	CCONJ
fcis-31421	41	13	enhance	enhance	VERB
fcis-31421	41	14	feature	feature	NOUN
fcis-31421	41	15	robustness	robustness	NOUN
fcis-31421	41	16	.	.	PUNCT
fcis-31421	42	1	compress	compress	VERB
fcis-31421	42	2	the	the	DET
fcis-31421	42	3	feature	feature	NOUN
fcis-31421	42	4	dimension	dimension	NOUN
fcis-31421	42	5	to	to	PART
fcis-31421	42	6	prevent	prevent	VERB
fcis-31421	42	7	overfitting	overfitting	NOUN
fcis-31421	42	8	and	and	CCONJ
fcis-31421	42	9	improve	improve	VERB
fcis-31421	42	10	the	the	DET
fcis-31421	42	11	model	model	NOUN
fcis-31421	42	12	's	's	PART
fcis-31421	42	13	tolerance	tolerance	NOUN
fcis-31421	42	14	to	to	ADP
fcis-31421	42	15	small	small	ADJ
fcis-31421	42	16	deformations	deformation	NOUN
fcis-31421	42	17	.	.	PUNCT
fcis-31421	43	1	activation	activation	NOUN
fcis-31421	43	2	function	function	NOUN
fcis-31421	43	3	:	:	PUNCT
fcis-31421	43	4	solving	solve	VERB
fcis-31421	43	5	the	the	DET
fcis-31421	43	6	gradient	gradient	NOUN
fcis-31421	43	7	vanishing	vanishing	NOUN
fcis-31421	43	8	problem	problem	NOUN
fcis-31421	43	9	and	and	CCONJ
fcis-31421	43	10	accelerating	accelerate	VERB
fcis-31421	43	11	convergence	convergence	NOUN
fcis-31421	43	12	.	.	PUNCT
fcis-31421	44	1	fully	fully	ADV
fcis-31421	44	2	connected	connected	ADJ
fcis-31421	44	3	layer	layer	NOUN
fcis-31421	44	4	:	:	PUNCT
fcis-31421	44	5	mapping	map	VERB
fcis-31421	44	6	high	high	ADJ
fcis-31421	44	7	-	-	PUNCT
fcis-31421	44	8	level	level	NOUN
fcis-31421	44	9	features	feature	NOUN
fcis-31421	44	10	to	to	ADP
fcis-31421	44	11	classification	classification	NOUN
fcis-31421	44	12	labels	label	NOUN
fcis-31421	44	13	to	to	PART
fcis-31421	44	14	complete	complete	VERB
fcis-31421	44	15	the	the	DET
fcis-31421	44	16	final	final	ADJ
fcis-31421	44	17	decision	decision	NOUN
fcis-31421	44	18	is	be	AUX
fcis-31421	44	19	usually	usually	ADV
fcis-31421	44	20	located	locate	VERB
fcis-31421	44	21	at	at	ADP
fcis-31421	44	22	the	the	DET
fcis-31421	44	23	end	end	NOUN
fcis-31421	44	24	of	of	ADP
fcis-31421	44	25	the	the	DET
fcis-31421	44	26	network	network	NOUN
fcis-31421	44	27	,	,	PUNCT
fcis-31421	44	28	connecting	connect	VERB
fcis-31421	44	29	all	all	DET
fcis-31421	44	30	neurons	neuron	NOUN
fcis-31421	44	31	.	.	PUNCT
fcis-31421	45	1	cnn	cnn	PROPN
fcis-31421	45	2	,	,	PUNCT
fcis-31421	45	3	as	as	ADP
fcis-31421	45	4	a	a	DET
fcis-31421	45	5	cornerstone	cornerstone	NOUN
fcis-31421	45	6	model	model	NOUN
fcis-31421	45	7	of	of	ADP
fcis-31421	45	8	computer	computer	NOUN
fcis-31421	45	9	vision	vision	NOUN
fcis-31421	45	10	,	,	PUNCT
fcis-31421	45	11	significantly	significantly	ADV
fcis-31421	45	12	improves	improve	VERB
fcis-31421	45	13	detection	detection	NOUN
fcis-31421	45	14	accuracy	accuracy	NOUN
fcis-31421	45	15	and	and	CCONJ
fcis-31421	45	16	scene	scene	NOUN
fcis-31421	45	17	adaptation	adaptation	NOUN
fcis-31421	45	18	through	through	ADP
fcis-31421	45	19	automatic	automatic	ADJ
fcis-31421	45	20	feature	feature	NOUN
fcis-31421	45	21	learning	learning	NOUN
fcis-31421	45	22	and	and	CCONJ
fcis-31421	45	23	end	end	NOUN
fcis-31421	45	24	-	-	PUNCT
fcis-31421	45	25	to	to	ADP
fcis-31421	45	26	-	-	PUNCT
fcis-31421	45	27	end	end	NOUN
fcis-31421	45	28	optimization	optimization	NOUN
fcis-31421	45	29	in	in	ADP
fcis-31421	45	30	fire	fire	NOUN
fcis-31421	45	31	detection	detection	NOUN
fcis-31421	45	32	.	.	PUNCT
fcis-31421	46	1	however	however	ADV
fcis-31421	46	2	,	,	PUNCT
fcis-31421	46	3	its	its	PRON
fcis-31421	46	4	limitations	limitation	NOUN
fcis-31421	46	5	in	in	ADP
fcis-31421	46	6	realtime	realtime	NOUN
fcis-31421	46	7	,	,	PUNCT
fcis-31421	46	8	dynamic	dynamic	ADJ
fcis-31421	46	9	modeling	modeling	NOUN
fcis-31421	46	10	and	and	CCONJ
fcis-31421	46	11	small	small	ADJ
fcis-31421	46	12	target	target	NOUN
fcis-31421	46	13	detection	detection	NOUN
fcis-31421	46	14	still	still	ADV
fcis-31421	46	15	need	need	VERB
fcis-31421	46	16	to	to	PART
fcis-31421	46	17	be	be	AUX
fcis-31421	46	18	further	far	ADV
fcis-31421	46	19	broken	break	VERB
fcis-31421	46	20	through	through	ADP
fcis-31421	46	21	by	by	ADP
fcis-31421	46	22	combining	combine	VERB
fcis-31421	46	23	lightweight	lightweight	ADJ
fcis-31421	46	24	design	design	NOUN
fcis-31421	46	25	,	,	PUNCT
fcis-31421	46	26	time	time	NOUN
fcis-31421	46	27	-	-	PUNCT
fcis-31421	46	28	series	series	NOUN
fcis-31421	46	29	network	network	NOUN
fcis-31421	46	30	and	and	CCONJ
fcis-31421	46	31	multimodal	multimodal	ADJ
fcis-31421	46	32	fusion	fusion	NOUN
fcis-31421	46	33	techniques	technique	NOUN
fcis-31421	46	34	.	.	PUNCT
fcis-31421	47	1	in	in	ADP
fcis-31421	47	2	recent	recent	ADJ
fcis-31421	47	3	years	year	NOUN
fcis-31421	47	4	,	,	PUNCT
fcis-31421	47	5	many	many	ADJ
fcis-31421	47	6	scholars	scholar	NOUN
fcis-31421	47	7	have	have	AUX
fcis-31421	47	8	made	make	VERB
fcis-31421	47	9	a	a	DET
fcis-31421	47	10	lot	lot	NOUN
fcis-31421	47	11	of	of	ADP
fcis-31421	47	12	improvements	improvement	NOUN
fcis-31421	47	13	based	base	VERB
fcis-31421	47	14	on	on	ADP
fcis-31421	47	15	cnn	cnn	PROPN
fcis-31421	47	16	to	to	PART
fcis-31421	47	17	improve	improve	VERB
fcis-31421	47	18	the	the	DET
fcis-31421	47	19	performance	performance	NOUN
fcis-31421	47	20	of	of	ADP
fcis-31421	47	21	detecting	detect	VERB
fcis-31421	47	22	fire	fire	NOUN
fcis-31421	47	23	to	to	ADP
fcis-31421	47	24	some	some	DET
fcis-31421	47	25	extent	extent	NOUN
fcis-31421	47	26	.	.	PUNCT
fcis-31421	48	1	the	the	DET
fcis-31421	48	2	khan	khan	PROPN
fcis-31421	48	3	muhammad	muhammad	PROPN
fcis-31421	48	4	research	research	NOUN
fcis-31421	48	5	team	team	NOUN
fcis-31421	48	6	used	use	VERB
fcis-31421	48	7	an	an	DET
fcis-31421	48	8	improved	improved	ADJ
fcis-31421	48	9	squeezenet	squeezenet	NOUN
fcis-31421	48	10	architecture	architecture	NOUN
fcis-31421	48	11	for	for	ADP
fcis-31421	48	12	model	model	NOUN
fcis-31421	48	13	53	53	NUM
fcis-31421	48	14	fine	fine	ADV
fcis-31421	48	15	-	-	PUNCT
fcis-31421	48	16	tuning	tuning	NOUN
fcis-31421	48	17	and	and	CCONJ
fcis-31421	48	18	developed	develop	VERB
fcis-31421	48	19	a	a	DET
fcis-31421	48	20	fire	fire	NOUN
fcis-31421	48	21	detection	detection	NOUN
fcis-31421	48	22	system	system	NOUN
fcis-31421	48	23	called	call	VERB
fcis-31421	48	24	cnnfire	cnnfire	NOUN
fcis-31421	48	25	.	.	PUNCT
fcis-31421	49	1	this	this	DET
fcis-31421	49	2	study	study	NOUN
fcis-31421	49	3	innovatively	innovatively	ADV
fcis-31421	49	4	proposes	propose	VERB
fcis-31421	49	5	an	an	DET
fcis-31421	49	6	intelligent	intelligent	ADJ
fcis-31421	49	7	feature	feature	NOUN
fcis-31421	49	8	selection	selection	NOUN
fcis-31421	49	9	algorithm	algorithm	NOUN
fcis-31421	49	10	that	that	PRON
fcis-31421	49	11	can	can	AUX
fcis-31421	49	12	automatically	automatically	ADV
fcis-31421	49	13	identify	identify	VERB
fcis-31421	49	14	and	and	CCONJ
fcis-31421	49	15	extract	extract	VERB
fcis-31421	49	16	the	the	DET
fcis-31421	49	17	most	most	ADV
fcis-31421	49	18	discriminative	discriminative	ADJ
fcis-31421	49	19	feature	feature	NOUN
fcis-31421	49	20	maps	map	NOUN
fcis-31421	49	21	for	for	ADP
fcis-31421	49	22	fire	fire	NOUN
fcis-31421	49	23	areas	area	NOUN
fcis-31421	49	24	in	in	ADP
fcis-31421	49	25	convolutional	convolutional	ADJ
fcis-31421	49	26	layers	layer	NOUN
fcis-31421	49	27	.	.	PUNCT
fcis-31421	50	1	these	these	DET
fcis-31421	50	2	feature	feature	NOUN
fcis-31421	50	3	maps	map	NOUN
fcis-31421	50	4	filtered	filter	VERB
fcis-31421	50	5	by	by	ADP
fcis-31421	50	6	algorithms	algorithm	NOUN
fcis-31421	50	7	demonstrate	demonstrate	VERB
fcis-31421	50	8	superior	superior	ADJ
fcis-31421	50	9	accuracy	accuracy	NOUN
fcis-31421	50	10	in	in	ADP
fcis-31421	50	11	fire	fire	NOUN
fcis-31421	50	12	segmentation	segmentation	NOUN
fcis-31421	50	13	tasks	task	NOUN
fcis-31421	50	14	compared	compare	VERB
fcis-31421	50	15	to	to	ADP
fcis-31421	50	16	traditional	traditional	ADJ
fcis-31421	50	17	manually	manually	ADV
fcis-31421	50	18	designed	design	VERB
fcis-31421	50	19	features	feature	NOUN
fcis-31421	50	20	.	.	PUNCT
fcis-31421	51	1	based	base	VERB
fcis-31421	51	2	on	on	ADP
fcis-31421	51	3	the	the	DET
fcis-31421	51	4	segmentation	segmentation	NOUN
fcis-31421	51	5	results	result	NOUN
fcis-31421	51	6	generated	generate	VERB
fcis-31421	51	7	from	from	ADP
fcis-31421	51	8	these	these	DET
fcis-31421	51	9	feature	feature	NOUN
fcis-31421	51	10	maps	map	NOUN
fcis-31421	51	11	,	,	PUNCT
fcis-31421	51	12	researchers	researcher	NOUN
fcis-31421	51	13	can	can	AUX
fcis-31421	51	14	deeply	deeply	ADV
fcis-31421	51	15	analyze	analyze	VERB
fcis-31421	51	16	the	the	DET
fcis-31421	51	17	dynamic	dynamic	ADJ
fcis-31421	51	18	characteristics	characteristic	NOUN
fcis-31421	51	19	of	of	ADP
fcis-31421	51	20	fires	fire	NOUN
fcis-31421	51	21	,	,	PUNCT
fcis-31421	51	22	including	include	VERB
fcis-31421	51	23	key	key	ADJ
fcis-31421	51	24	parameters	parameter	NOUN
fcis-31421	51	25	such	such	ADJ
fcis-31421	51	26	as	as	ADP
fcis-31421	51	27	fire	fire	NOUN
fcis-31421	51	28	spread	spread	VERB
fcis-31421	51	29	speed	speed	NOUN
fcis-31421	51	30	.	.	PUNCT
fcis-31421	52	1	in	in	ADP
fcis-31421	52	2	addition	addition	NOUN
fcis-31421	52	3	,	,	PUNCT
fcis-31421	52	4	this	this	DET
fcis-31421	52	5	method	method	NOUN
fcis-31421	52	6	can	can	AUX
fcis-31421	52	7	effectively	effectively	ADV
fcis-31421	52	8	evaluate	evaluate	VERB
fcis-31421	52	9	the	the	DET
fcis-31421	52	10	development	development	NOUN
fcis-31421	52	11	stage	stage	NOUN
fcis-31421	52	12	and	and	CCONJ
fcis-31421	52	13	combustion	combustion	NOUN
fcis-31421	52	14	intensity	intensity	NOUN
fcis-31421	52	15	of	of	ADP
fcis-31421	52	16	a	a	DET
fcis-31421	52	17	fire	fire	NOUN
fcis-31421	52	18	,	,	PUNCT
fcis-31421	52	19	providing	provide	VERB
fcis-31421	52	20	a	a	DET
fcis-31421	52	21	more	more	ADV
fcis-31421	52	22	comprehensive	comprehensive	ADJ
fcis-31421	52	23	analytical	analytical	ADJ
fcis-31421	52	24	dimension	dimension	NOUN
fcis-31421	52	25	for	for	ADP
fcis-31421	52	26	fire	fire	NOUN
fcis-31421	52	27	monitoring[1	monitoring[1	NOUN
fcis-31421	52	28	]	]	PUNCT
fcis-31421	52	29	.	.	PUNCT
fcis-31421	53	1	but	but	CCONJ
fcis-31421	53	2	there	there	PRON
fcis-31421	53	3	are	be	VERB
fcis-31421	53	4	also	also	ADV
fcis-31421	53	5	motion	motion	NOUN
fcis-31421	53	6	-	-	PUNCT
fcis-31421	53	7	based	base	VERB
fcis-31421	53	8	fire	fire	NOUN
fcis-31421	53	9	detection	detection	NOUN
fcis-31421	53	10	schemes	scheme	NOUN
fcis-31421	53	11	that	that	PRON
fcis-31421	53	12	may	may	AUX
fcis-31421	53	13	not	not	PART
fcis-31421	53	14	work	work	VERB
fcis-31421	53	15	properly	properly	ADV
fcis-31421	53	16	if	if	SCONJ
fcis-31421	53	17	there	there	PRON
fcis-31421	53	18	is	be	VERB
fcis-31421	53	19	cloudy	cloudy	ADJ
fcis-31421	53	20	and	and	CCONJ
fcis-31421	53	21	complex	complex	ADJ
fcis-31421	53	22	weather	weather	NOUN
fcis-31421	53	23	.	.	PUNCT
fcis-31421	54	1	and	and	CCONJ
fcis-31421	54	2	the	the	DET
fcis-31421	54	3	false	false	ADJ
fcis-31421	54	4	positive	positive	ADJ
fcis-31421	54	5	score	score	NOUN
fcis-31421	54	6	of	of	ADP
fcis-31421	54	7	11.67	11.67	NUM
fcis-31421	54	8	%	%	NOUN
fcis-31421	54	9	is	be	AUX
fcis-31421	54	10	still	still	ADV
fcis-31421	54	11	high	high	ADJ
fcis-31421	54	12	and	and	CCONJ
fcis-31421	54	13	the	the	DET
fcis-31421	54	14	accuracy	accuracy	NOUN
fcis-31421	54	15	needs	need	VERB
fcis-31421	54	16	to	to	PART
fcis-31421	54	17	be	be	AUX
fcis-31421	54	18	further	far	ADV
fcis-31421	54	19	improved	improve	VERB
fcis-31421	54	20	.	.	PUNCT
fcis-31421	55	1	to	to	PART
fcis-31421	55	2	achieve	achieve	VERB
fcis-31421	55	3	high	high	ADJ
fcis-31421	55	4	accuracy	accuracy	NOUN
fcis-31421	55	5	and	and	CCONJ
fcis-31421	55	6	low	low	ADJ
fcis-31421	55	7	false	false	ADJ
fcis-31421	55	8	alarm	alarm	NOUN
fcis-31421	55	9	rate	rate	NOUN
fcis-31421	55	10	.	.	PUNCT
fcis-31421	56	1	also	also	ADV
fcis-31421	56	2	,	,	PUNCT
fcis-31421	56	3	khan	khan	PROPN
fcis-31421	56	4	muhammad	muhammad	PROPN
fcis-31421	56	5	et	et	PROPN
fcis-31421	56	6	al	al	PROPN
fcis-31421	56	7	.	.	PROPN
fcis-31421	56	8	used	use	VERB
fcis-31421	56	9	a	a	DET
fcis-31421	56	10	googlenet	googlenet	NOUN
fcis-31421	56	11	-	-	PUNCT
fcis-31421	56	12	like	like	ADJ
fcis-31421	56	13	model	model	NOUN
fcis-31421	56	14	and	and	CCONJ
fcis-31421	56	15	applied	apply	VERB
fcis-31421	56	16	a	a	DET
fcis-31421	56	17	transfer	transfer	NOUN
fcis-31421	56	18	learning	learn	VERB
fcis-31421	56	19	strategy	strategy	NOUN
fcis-31421	57	1	[	[	X
fcis-31421	57	2	2	2	NUM
fcis-31421	57	3	]	]	PUNCT
fcis-31421	57	4	.	.	PUNCT
fcis-31421	58	1	googlenet	googlenet	PROPN
fcis-31421	58	2	is	be	AUX
fcis-31421	58	3	an	an	DET
fcis-31421	58	4	ideal	ideal	ADJ
fcis-31421	58	5	choice	choice	NOUN
fcis-31421	58	6	due	due	ADP
fcis-31421	58	7	to	to	ADP
fcis-31421	58	8	its	its	PRON
fcis-31421	58	9	excellent	excellent	ADJ
fcis-31421	58	10	classification	classification	NOUN
fcis-31421	58	11	performance	performance	NOUN
fcis-31421	58	12	,	,	PUNCT
fcis-31421	58	13	compact	compact	ADJ
fcis-31421	58	14	model	model	NOUN
fcis-31421	58	15	structure	structure	NOUN
fcis-31421	58	16	,	,	PUNCT
fcis-31421	58	17	and	and	CCONJ
fcis-31421	58	18	good	good	ADJ
fcis-31421	58	19	adaptability	adaptability	NOUN
fcis-31421	58	20	to	to	ADP
fcis-31421	58	21	memory	memory	NOUN
fcis-31421	58	22	limited	limited	ADJ
fcis-31421	58	23	hardware	hardware	NOUN
fcis-31421	58	24	.	.	PUNCT
fcis-31421	59	1	through	through	ADP
fcis-31421	59	2	the	the	DET
fcis-31421	59	3	application	application	NOUN
fcis-31421	59	4	of	of	ADP
fcis-31421	59	5	transfer	transfer	NOUN
fcis-31421	59	6	learning	learn	VERB
fcis-31421	59	7	technology	technology	NOUN
fcis-31421	59	8	,	,	PUNCT
fcis-31421	59	9	researchers	researcher	NOUN
fcis-31421	59	10	have	have	AUX
fcis-31421	59	11	successfully	successfully	ADV
fcis-31421	59	12	improved	improve	VERB
fcis-31421	59	13	the	the	DET
fcis-31421	59	14	accuracy	accuracy	NOUN
fcis-31421	59	15	of	of	ADP
fcis-31421	59	16	flame	flame	NOUN
fcis-31421	59	17	detection	detection	NOUN
fcis-31421	59	18	from	from	ADP
fcis-31421	59	19	88.41	88.41	NUM
fcis-31421	59	20	%	%	NOUN
fcis-31421	59	21	to	to	ADP
fcis-31421	59	22	94	94	NUM
fcis-31421	59	23	%	%	NOUN
fcis-31421	59	24	,	,	PUNCT
fcis-31421	59	25	achieving	achieve	VERB
fcis-31421	59	26	significant	significant	ADJ
fcis-31421	59	27	progress	progress	NOUN
fcis-31421	59	28	.	.	PUNCT
fcis-31421	60	1	however	however	ADV
fcis-31421	60	2	,	,	PUNCT
fcis-31421	60	3	there	there	PRON
fcis-31421	60	4	is	be	VERB
fcis-31421	60	5	still	still	ADV
fcis-31421	60	6	room	room	NOUN
fcis-31421	60	7	for	for	ADP
fcis-31421	60	8	improvement	improvement	NOUN
fcis-31421	60	9	in	in	ADP
fcis-31421	60	10	reducing	reduce	VERB
fcis-31421	60	11	false	false	ADJ
fcis-31421	60	12	alarm	alarm	NOUN
fcis-31421	60	13	rates	rate	NOUN
fcis-31421	60	14	using	use	VERB
fcis-31421	60	15	this	this	DET
fcis-31421	60	16	method	method	NOUN
fcis-31421	60	17	,	,	PUNCT
fcis-31421	60	18	which	which	PRON
fcis-31421	60	19	points	point	VERB
fcis-31421	60	20	the	the	DET
fcis-31421	60	21	way	way	NOUN
fcis-31421	60	22	for	for	ADP
fcis-31421	60	23	future	future	ADJ
fcis-31421	60	24	research	research	NOUN
fcis-31421	60	25	.	.	PUNCT
fcis-31421	61	1	barmpoutis	barmpoutis	PROPN
fcis-31421	61	2	p	p	PROPN
fcis-31421	61	3	proposed	propose	VERB
fcis-31421	61	4	an	an	DET
fcis-31421	61	5	innovative	innovative	ADJ
fcis-31421	61	6	early	early	ADJ
fcis-31421	61	7	fire	fire	NOUN
fcis-31421	61	8	detection	detection	NOUN
fcis-31421	61	9	method	method	NOUN
fcis-31421	61	10	that	that	PRON
fcis-31421	61	11	combines	combine	VERB
fcis-31421	61	12	deep	deep	ADJ
fcis-31421	61	13	learning	learning	NOUN
fcis-31421	61	14	and	and	CCONJ
fcis-31421	61	15	multi	multi	ADJ
fcis-31421	61	16	-	-	ADJ
fcis-31421	61	17	dimensional	dimensional	ADJ
fcis-31421	61	18	texture	texture	ADJ
fcis-31421	61	19	feature	feature	NOUN
fcis-31421	61	20	analysis	analysis	NOUN
fcis-31421	61	21	techniques	technique	NOUN
fcis-31421	61	22	based	base	VERB
fcis-31421	61	23	on	on	ADP
fcis-31421	61	24	linear	linear	PROPN
fcis-31421	61	25	dynamic	dynamic	ADJ
fcis-31421	61	26	systems	system	NOUN
fcis-31421	61	27	(	(	PUNCT
fcis-31421	61	28	lds).this	lds).this	DET
fcis-31421	61	29	study	study	NOUN
fcis-31421	61	30	adopts	adopt	VERB
fcis-31421	61	31	the	the	DET
fcis-31421	61	32	regional	regional	ADJ
fcis-31421	61	33	convolutional	convolutional	ADJ
fcis-31421	61	34	neural	neural	ADJ
fcis-31421	61	35	network	network	NOUN
fcis-31421	61	36	(	(	PUNCT
fcis-31421	61	37	r	r	NOUN
fcis-31421	61	38	-	-	PUNCT
fcis-31421	61	39	cnn	cnn	NOUN
fcis-31421	61	40	)	)	PUNCT
fcis-31421	61	41	architecture	architecture	NOUN
fcis-31421	61	42	,	,	PUNCT
fcis-31421	61	43	which	which	PRON
fcis-31421	61	44	synchronously	synchronously	ADV
fcis-31421	61	45	optimizes	optimize	VERB
fcis-31421	61	46	target	target	NOUN
fcis-31421	61	47	classification	classification	NOUN
fcis-31421	61	48	and	and	CCONJ
fcis-31421	61	49	bounding	bound	VERB
fcis-31421	61	50	box	box	NOUN
fcis-31421	61	51	regression	regression	NOUN
fcis-31421	61	52	tasks	task	NOUN
fcis-31421	61	53	through	through	ADP
fcis-31421	61	54	end	end	NOUN
fcis-31421	61	55	-	-	PUNCT
fcis-31421	61	56	to	to	ADP
fcis-31421	61	57	-	-	PUNCT
fcis-31421	61	58	end	end	NOUN
fcis-31421	61	59	training	training	NOUN
fcis-31421	61	60	mechanisms	mechanism	NOUN
fcis-31421	61	61	,	,	PUNCT
fcis-31421	61	62	achieving	achieve	VERB
fcis-31421	61	63	precise	precise	ADJ
fcis-31421	61	64	localization	localization	NOUN
fcis-31421	61	65	and	and	CCONJ
fcis-31421	61	66	recognition	recognition	NOUN
fcis-31421	61	67	of	of	ADP
fcis-31421	61	68	fire	fire	NOUN
fcis-31421	61	69	areas[3	areas[3	PROPN
fcis-31421	61	70	]	]	PUNCT
fcis-31421	61	71	.	.	PUNCT
fcis-31421	62	1	such	such	DET
fcis-31421	62	2	an	an	DET
fcis-31421	62	3	approach	approach	NOUN
fcis-31421	62	4	significantly	significantly	ADV
fcis-31421	62	5	reduces	reduce	VERB
fcis-31421	62	6	the	the	DET
fcis-31421	62	7	overall	overall	ADJ
fcis-31421	62	8	computational	computational	ADJ
fcis-31421	62	9	complexity	complexity	NOUN
fcis-31421	62	10	while	while	SCONJ
fcis-31421	62	11	improving	improve	VERB
fcis-31421	62	12	the	the	DET
fcis-31421	62	13	performance	performance	NOUN
fcis-31421	62	14	by	by	ADP
fcis-31421	62	15	significantly	significantly	ADV
fcis-31421	62	16	reducing	reduce	VERB
fcis-31421	62	17	the	the	DET
fcis-31421	62	18	false	false	ADJ
fcis-31421	62	19	positive	positive	ADJ
fcis-31421	62	20	rate	rate	NOUN
fcis-31421	62	21	due	due	ADP
fcis-31421	62	22	to	to	ADP
fcis-31421	62	23	flame	flame	ADJ
fcis-31421	62	24	color	color	NOUN
fcis-31421	62	25	targets	target	NOUN
fcis-31421	62	26	while	while	SCONJ
fcis-31421	62	27	maintaining	maintain	VERB
fcis-31421	62	28	a	a	DET
fcis-31421	62	29	high	high	ADJ
fcis-31421	62	30	true	true	ADJ
fcis-31421	62	31	positive	positive	ADJ
fcis-31421	62	32	rate	rate	NOUN
fcis-31421	62	33	.	.	PUNCT
fcis-31421	63	1	the	the	DET
fcis-31421	63	2	faisal	faisal	PROPN
fcis-31421	63	3	saeed	saeed	PROPN
fcis-31421	63	4	research	research	PROPN
fcis-31421	63	5	team	team	NOUN
fcis-31421	63	6	has	have	AUX
fcis-31421	63	7	developed	develop	VERB
fcis-31421	63	8	a	a	DET
fcis-31421	63	9	multimodal	multimodal	ADJ
fcis-31421	63	10	deep	deep	ADJ
fcis-31421	63	11	neural	neural	ADJ
fcis-31421	63	12	network	network	NOUN
fcis-31421	63	13	architecture	architecture	NOUN
fcis-31421	63	14	that	that	PRON
fcis-31421	63	15	integrates	integrate	VERB
fcis-31421	63	16	three	three	NUM
fcis-31421	63	17	core	core	NOUN
fcis-31421	63	18	model	model	NOUN
fcis-31421	63	19	components	component	NOUN
fcis-31421	63	20	:	:	PUNCT
fcis-31421	63	21	a	a	DET
fcis-31421	63	22	hybrid	hybrid	NOUN
fcis-31421	63	23	classifier	classifier	NOUN
fcis-31421	63	24	based	base	VERB
fcis-31421	63	25	on	on	ADP
fcis-31421	63	26	adaboost	adaboost	ADJ
fcis-31421	63	27	algorithm	algorithm	NOUN
fcis-31421	63	28	and	and	CCONJ
fcis-31421	63	29	multi	multi	ADJ
fcis-31421	63	30	-	-	ADJ
fcis-31421	63	31	layer	layer	ADJ
fcis-31421	63	32	perceptron	perceptron	NOUN
fcis-31421	63	33	(	(	PUNCT
fcis-31421	63	34	mlp	mlp	NOUN
fcis-31421	63	35	)	)	PUNCT
fcis-31421	63	36	ensemble	ensemble	ADJ
fcis-31421	63	37	;	;	PUNCT
fcis-31421	63	38	combining	combine	VERB
fcis-31421	63	39	adaboost	adaboost	ADV
fcis-31421	63	40	with	with	ADP
fcis-31421	63	41	local	local	ADJ
fcis-31421	63	42	binary	binary	ADJ
fcis-31421	63	43	pattern	pattern	NOUN
fcis-31421	63	44	(	(	PUNCT
fcis-31421	63	45	lbp	lbp	NOUN
fcis-31421	63	46	)	)	PUNCT
fcis-31421	63	47	feature	feature	NOUN
fcis-31421	63	48	extraction	extraction	NOUN
fcis-31421	63	49	model	model	NOUN
fcis-31421	63	50	;	;	PUNCT
fcis-31421	63	51	deep	deep	ADJ
fcis-31421	63	52	convolutional	convolutional	ADJ
fcis-31421	63	53	neural	neural	ADJ
fcis-31421	63	54	networks	network	NOUN
fcis-31421	63	55	.	.	PUNCT
fcis-31421	64	1	to	to	PART
fcis-31421	64	2	optimize	optimize	VERB
fcis-31421	64	3	detection	detection	NOUN
fcis-31421	64	4	performance	performance	NOUN
fcis-31421	64	5	,	,	PUNCT
fcis-31421	64	6	researchers	researcher	NOUN
fcis-31421	64	7	specifically	specifically	ADV
fcis-31421	64	8	used	use	VERB
fcis-31421	64	9	cnn	cnn	PROPN
fcis-31421	64	10	for	for	ADP
fcis-31421	64	11	secondary	secondary	ADJ
fcis-31421	64	12	classification	classification	NOUN
fcis-31421	64	13	validation	validation	NOUN
fcis-31421	64	14	of	of	ADP
fcis-31421	64	15	candidate	candidate	NOUN
fcis-31421	64	16	regions	region	NOUN
fcis-31421	64	17	(	(	PUNCT
fcis-31421	64	18	rois	rois	PROPN
fcis-31421	64	19	)	)	PUNCT
fcis-31421	64	20	.	.	PUNCT
fcis-31421	65	1	this	this	DET
fcis-31421	65	2	strategy	strategy	NOUN
fcis-31421	65	3	not	not	PART
fcis-31421	65	4	only	only	ADV
fcis-31421	65	5	significantly	significantly	ADV
fcis-31421	65	6	reduced	reduce	VERB
fcis-31421	65	7	the	the	DET
fcis-31421	65	8	system	system	NOUN
fcis-31421	65	9	's	's	PART
fcis-31421	65	10	false	false	ADJ
fcis-31421	65	11	alarm	alarm	NOUN
fcis-31421	65	12	rate	rate	NOUN
fcis-31421	65	13	,	,	PUNCT
fcis-31421	65	14	but	but	CCONJ
fcis-31421	65	15	also	also	ADV
fcis-31421	65	16	improved	improve	VERB
fcis-31421	65	17	the	the	DET
fcis-31421	65	18	overall	overall	ADJ
fcis-31421	65	19	classification	classification	NOUN
fcis-31421	65	20	accuracy	accuracy	NOUN
fcis-31421	65	21	to	to	ADP
fcis-31421	65	22	an	an	DET
fcis-31421	65	23	excellent	excellent	ADJ
fcis-31421	65	24	level	level	NOUN
fcis-31421	65	25	of	of	ADP
fcis-31421	65	26	97.8	97.8	NUM
fcis-31421	65	27	%	%	NOUN
fcis-31421	65	28	[	[	X
fcis-31421	65	29	4	4	NUM
fcis-31421	65	30	]	]	PUNCT
fcis-31421	65	31	.	.	PUNCT
fcis-31421	66	1	however	however	ADV
fcis-31421	66	2	,	,	PUNCT
fcis-31421	66	3	more	more	ADJ
fcis-31421	66	4	datasets	dataset	NOUN
fcis-31421	66	5	are	be	AUX
fcis-31421	66	6	used	use	VERB
fcis-31421	66	7	,	,	PUNCT
fcis-31421	66	8	the	the	DET
fcis-31421	66	9	training	training	NOUN
fcis-31421	66	10	period	period	NOUN
fcis-31421	66	11	is	be	AUX
fcis-31421	66	12	longer	long	ADJ
fcis-31421	66	13	,	,	PUNCT
fcis-31421	66	14	and	and	CCONJ
fcis-31421	66	15	the	the	DET
fcis-31421	66	16	mixing	mixing	NOUN
fcis-31421	66	17	of	of	ADP
fcis-31421	66	18	multiple	multiple	ADJ
fcis-31421	66	19	models	model	NOUN
fcis-31421	66	20	may	may	AUX
fcis-31421	66	21	lead	lead	VERB
fcis-31421	66	22	to	to	ADP
fcis-31421	66	23	a	a	DET
fcis-31421	66	24	more	more	ADV
fcis-31421	66	25	complex	complex	ADJ
fcis-31421	66	26	system	system	NOUN
fcis-31421	66	27	that	that	PRON
fcis-31421	66	28	is	be	AUX
fcis-31421	66	29	not	not	PART
fcis-31421	66	30	suitable	suitable	ADJ
fcis-31421	66	31	for	for	ADP
fcis-31421	66	32	popularization	popularization	NOUN
fcis-31421	66	33	.	.	PUNCT
fcis-31421	67	1	although	although	SCONJ
fcis-31421	67	2	there	there	PRON
fcis-31421	67	3	are	be	VERB
fcis-31421	67	4	a	a	DET
fcis-31421	67	5	lot	lot	NOUN
fcis-31421	67	6	of	of	ADP
fcis-31421	67	7	research	research	NOUN
fcis-31421	67	8	with	with	ADP
fcis-31421	67	9	more	more	ADV
fcis-31421	67	10	obvious	obvious	ADJ
fcis-31421	67	11	improvements	improvement	NOUN
fcis-31421	67	12	on	on	ADP
fcis-31421	67	13	the	the	DET
fcis-31421	67	14	basis	basis	NOUN
fcis-31421	67	15	of	of	ADP
fcis-31421	67	16	cnn	cnn	PROPN
fcis-31421	67	17	,	,	PUNCT
fcis-31421	67	18	there	there	PRON
fcis-31421	67	19	still	still	ADV
fcis-31421	67	20	exists	exist	VERB
fcis-31421	67	21	the	the	DET
fcis-31421	67	22	problem	problem	NOUN
fcis-31421	67	23	of	of	ADP
fcis-31421	67	24	misjudging	misjudge	VERB
fcis-31421	67	25	fire	fire	NOUN
fcis-31421	67	26	detection	detection	NOUN
fcis-31421	67	27	in	in	ADP
fcis-31421	67	28	dynamic	dynamic	ADJ
fcis-31421	67	29	scenes	scene	NOUN
fcis-31421	67	30	and	and	CCONJ
fcis-31421	67	31	variable	variable	ADJ
fcis-31421	67	32	and	and	CCONJ
fcis-31421	67	33	complex	complex	ADJ
fcis-31421	67	34	weather	weather	NOUN
fcis-31421	67	35	.	.	PUNCT
fcis-31421	68	1	moreover	moreover	ADV
fcis-31421	68	2	,	,	PUNCT
fcis-31421	68	3	the	the	DET
fcis-31421	68	4	deep	deep	ADJ
fcis-31421	68	5	network	network	NOUN
fcis-31421	68	6	structure	structure	NOUN
fcis-31421	68	7	of	of	ADP
fcis-31421	68	8	cnn	cnn	PROPN
fcis-31421	68	9	contains	contain	VERB
fcis-31421	68	10	a	a	DET
fcis-31421	68	11	large	large	ADJ
fcis-31421	68	12	number	number	NOUN
fcis-31421	68	13	of	of	ADP
fcis-31421	68	14	convolutional	convolutional	ADJ
fcis-31421	68	15	layers	layer	NOUN
fcis-31421	68	16	and	and	CCONJ
fcis-31421	68	17	fully	fully	ADV
fcis-31421	68	18	connected	connected	ADJ
fcis-31421	68	19	layers	layer	NOUN
fcis-31421	68	20	,	,	PUNCT
fcis-31421	68	21	with	with	ADP
fcis-31421	68	22	high	high	ADJ
fcis-31421	68	23	computational	computational	ADJ
fcis-31421	68	24	complexity	complexity	NOUN
fcis-31421	68	25	and	and	CCONJ
fcis-31421	68	26	limited	limited	ADJ
fcis-31421	68	27	inference	inference	NOUN
fcis-31421	68	28	speed	speed	NOUN
fcis-31421	68	29	,	,	PUNCT
fcis-31421	68	30	even	even	ADV
fcis-31421	68	31	if	if	SCONJ
fcis-31421	68	32	there	there	PRON
fcis-31421	68	33	is	be	VERB
fcis-31421	68	34	an	an	DET
fcis-31421	68	35	improvement	improvement	NOUN
fcis-31421	68	36	on	on	ADP
fcis-31421	68	37	the	the	DET
fcis-31421	68	38	basis	basis	NOUN
fcis-31421	68	39	of	of	ADP
fcis-31421	68	40	this	this	PRON
fcis-31421	68	41	,	,	PUNCT
fcis-31421	68	42	it	it	PRON
fcis-31421	68	43	is	be	AUX
fcis-31421	68	44	still	still	ADV
fcis-31421	68	45	difficult	difficult	ADJ
fcis-31421	68	46	to	to	PART
fcis-31421	68	47	achieve	achieve	VERB
fcis-31421	68	48	real	real	ADJ
fcis-31421	68	49	-	-	PUNCT
fcis-31421	68	50	time	time	NOUN
fcis-31421	68	51	detection	detection	NOUN
fcis-31421	68	52	,	,	PUNCT
fcis-31421	68	53	which	which	PRON
fcis-31421	68	54	is	be	AUX
fcis-31421	68	55	likely	likely	ADJ
fcis-31421	68	56	to	to	PART
fcis-31421	68	57	lead	lead	VERB
fcis-31421	68	58	to	to	ADP
fcis-31421	68	59	the	the	DET
fcis-31421	68	60	detection	detection	NOUN
fcis-31421	68	61	of	of	ADP
fcis-31421	68	62	the	the	DET
fcis-31421	68	63	fire	fire	NOUN
fcis-31421	68	64	after	after	SCONJ
fcis-31421	68	65	it	it	PRON
fcis-31421	68	66	spreads	spread	VERB
fcis-31421	68	67	to	to	PART
fcis-31421	68	68	trigger	trigger	VERB
fcis-31421	68	69	the	the	DET
fcis-31421	68	70	alarm	alarm	NOUN
fcis-31421	68	71	,	,	PUNCT
fcis-31421	68	72	and	and	CCONJ
fcis-31421	68	73	fail	fail	VERB
fcis-31421	68	74	to	to	PART
fcis-31421	68	75	achieve	achieve	VERB
fcis-31421	68	76	the	the	DET
fcis-31421	68	77	effect	effect	NOUN
fcis-31421	68	78	of	of	ADP
fcis-31421	68	79	early	early	ADJ
fcis-31421	68	80	warning	warning	NOUN
fcis-31421	68	81	.	.	PUNCT
fcis-31421	69	1	and	and	CCONJ
fcis-31421	69	2	the	the	DET
fcis-31421	69	3	cnn	cnn	PROPN
fcis-31421	69	4	high	high	ADJ
fcis-31421	69	5	-	-	PUNCT
fcis-31421	69	6	level	level	NOUN
fcis-31421	69	7	feature	feature	NOUN
fcis-31421	69	8	map	map	NOUN
fcis-31421	69	9	has	have	VERB
fcis-31421	69	10	a	a	DET
fcis-31421	69	11	low	low	ADJ
fcis-31421	69	12	resolution	resolution	NOUN
fcis-31421	69	13	after	after	ADP
fcis-31421	69	14	many	many	ADJ
fcis-31421	69	15	pooling	pooling	NOUN
fcis-31421	69	16	,	,	PUNCT
fcis-31421	69	17	and	and	CCONJ
fcis-31421	69	18	the	the	DET
fcis-31421	69	19	detail	detail	NOUN
fcis-31421	69	20	information	information	NOUN
fcis-31421	69	21	of	of	ADP
fcis-31421	69	22	small	small	ADJ
fcis-31421	69	23	targets	target	NOUN
fcis-31421	69	24	such	such	ADJ
fcis-31421	69	25	as	as	ADP
fcis-31421	69	26	long	long	ADJ
fcis-31421	69	27	-	-	PUNCT
fcis-31421	69	28	distance	distance	NOUN
fcis-31421	69	29	small	small	ADJ
fcis-31421	69	30	fires	fire	NOUN
fcis-31421	69	31	is	be	AUX
fcis-31421	69	32	seriously	seriously	ADV
fcis-31421	69	33	lost	lose	VERB
fcis-31421	69	34	.	.	PUNCT
fcis-31421	70	1	in	in	ADP
fcis-31421	70	2	the	the	DET
fcis-31421	70	3	early	early	ADJ
fcis-31421	70	4	stage	stage	NOUN
fcis-31421	70	5	of	of	ADP
fcis-31421	70	6	the	the	DET
fcis-31421	70	7	flame	flame	NOUN
fcis-31421	70	8	,	,	PUNCT
fcis-31421	70	9	long	long	ADJ
fcis-31421	70	10	-	-	PUNCT
fcis-31421	70	11	distance	distance	NOUN
fcis-31421	70	12	small	small	ADJ
fcis-31421	70	13	fires	fire	NOUN
fcis-31421	70	14	only	only	ADV
fcis-31421	70	15	account	account	VERB
fcis-31421	70	16	for	for	ADP
fcis-31421	70	17	a	a	DET
fcis-31421	70	18	few	few	ADJ
fcis-31421	70	19	pixels	pixel	NOUN
fcis-31421	70	20	of	of	ADP
fcis-31421	70	21	the	the	DET
fcis-31421	70	22	image	image	NOUN
fcis-31421	70	23	,	,	PUNCT
fcis-31421	70	24	and	and	CCONJ
fcis-31421	70	25	it	it	PRON
fcis-31421	70	26	is	be	AUX
fcis-31421	70	27	difficult	difficult	ADJ
fcis-31421	70	28	for	for	SCONJ
fcis-31421	70	29	cnn	cnn	PROPN
fcis-31421	70	30	to	to	PART
fcis-31421	70	31	carry	carry	VERB
fcis-31421	70	32	out	out	ADP
fcis-31421	70	33	effective	effective	ADJ
fcis-31421	70	34	recognition	recognition	NOUN
fcis-31421	70	35	.	.	PUNCT
fcis-31421	71	1	2.2	2.2	NUM
fcis-31421	71	2	.	.	PUNCT
fcis-31421	71	3	multimodal	multimodal	NOUN
fcis-31421	71	4	:	:	PUNCT
fcis-31421	71	5	multimodal	multimodal	ADJ
fcis-31421	71	6	deep	deep	ADJ
fcis-31421	71	7	learning	learning	NOUN
fcis-31421	71	8	refers	refer	VERB
fcis-31421	71	9	to	to	ADP
fcis-31421	71	10	improving	improve	VERB
fcis-31421	71	11	the	the	DET
fcis-31421	71	12	perception	perception	NOUN
fcis-31421	71	13	and	and	CCONJ
fcis-31421	71	14	decision	decision	NOUN
fcis-31421	71	15	-	-	PUNCT
fcis-31421	71	16	making	make	VERB
fcis-31421	71	17	ability	ability	NOUN
fcis-31421	71	18	of	of	ADP
fcis-31421	71	19	a	a	DET
fcis-31421	71	20	model	model	NOUN
fcis-31421	71	21	by	by	ADP
fcis-31421	71	22	jointly	jointly	ADV
fcis-31421	71	23	learning	learn	VERB
fcis-31421	71	24	information	information	NOUN
fcis-31421	71	25	from	from	ADP
fcis-31421	71	26	multiple	multiple	ADJ
fcis-31421	71	27	heterogeneous	heterogeneous	ADJ
fcis-31421	71	28	data	datum	NOUN
fcis-31421	71	29	.	.	PUNCT
fcis-31421	72	1	in	in	ADP
fcis-31421	72	2	fire	fire	NOUN
fcis-31421	72	3	detection	detection	NOUN
fcis-31421	72	4	,	,	PUNCT
fcis-31421	72	5	modalities	modality	NOUN
fcis-31421	72	6	usually	usually	ADV
fcis-31421	72	7	include	include	VERB
fcis-31421	72	8	:	:	PUNCT
fcis-31421	72	9	visual	visual	ADJ
fcis-31421	72	10	modalities	modality	NOUN
fcis-31421	72	11	:	:	PUNCT
fcis-31421	72	12	visible	visible	ADJ
fcis-31421	72	13	camera	camera	NOUN
fcis-31421	72	14	images	image	NOUN
fcis-31421	72	15	,	,	PUNCT
fcis-31421	72	16	infrared	infrared	ADJ
fcis-31421	72	17	thermal	thermal	ADJ
fcis-31421	72	18	imaging	imaging	NOUN
fcis-31421	72	19	.	.	PUNCT
fcis-31421	73	1	physical	physical	ADJ
fcis-31421	73	2	modalities	modality	NOUN
fcis-31421	73	3	:	:	PUNCT
fcis-31421	73	4	temperature	temperature	NOUN
fcis-31421	73	5	sensor	sensor	NOUN
fcis-31421	73	6	data	datum	NOUN
fcis-31421	73	7	,	,	PUNCT
fcis-31421	73	8	smoke	smoke	NOUN
fcis-31421	73	9	concentration	concentration	NOUN
fcis-31421	73	10	sensor	sensor	NOUN
fcis-31421	73	11	signals	signal	NOUN
fcis-31421	73	12	.	.	PUNCT
fcis-31421	74	1	environmental	environmental	ADJ
fcis-31421	74	2	modalities	modality	NOUN
fcis-31421	74	3	:	:	PUNCT
fcis-31421	74	4	meteorological	meteorological	ADJ
fcis-31421	74	5	data	datum	NOUN
fcis-31421	74	6	such	such	ADJ
fcis-31421	74	7	as	as	ADP
fcis-31421	74	8	wind	wind	NOUN
fcis-31421	74	9	speed	speed	NOUN
fcis-31421	74	10	,	,	PUNCT
fcis-31421	74	11	humidity	humidity	NOUN
fcis-31421	74	12	,	,	PUNCT
fcis-31421	74	13	geographic	geographic	ADJ
fcis-31421	74	14	information	information	NOUN
fcis-31421	74	15	system	system	NOUN
fcis-31421	74	16	(	(	PUNCT
fcis-31421	74	17	gis	gis	NOUN
fcis-31421	74	18	)	)	PUNCT
fcis-31421	74	19	maps	map	NOUN
fcis-31421	74	20	.	.	PUNCT
fcis-31421	75	1	different	different	ADJ
fcis-31421	75	2	modalities	modality	NOUN
fcis-31421	75	3	provide	provide	VERB
fcis-31421	75	4	complementary	complementary	ADJ
fcis-31421	75	5	information	information	NOUN
fcis-31421	75	6	,	,	PUNCT
fcis-31421	75	7	and	and	CCONJ
fcis-31421	75	8	through	through	ADP
fcis-31421	75	9	cross	cross	ADJ
fcis-31421	75	10	-	-	ADJ
fcis-31421	75	11	modal	modal	ADJ
fcis-31421	75	12	association	association	NOUN
fcis-31421	75	13	learning	learning	NOUN
fcis-31421	75	14	,	,	PUNCT
fcis-31421	75	15	it	it	PRON
fcis-31421	75	16	breaks	break	VERB
fcis-31421	75	17	through	through	ADP
fcis-31421	75	18	the	the	DET
fcis-31421	75	19	perceptual	perceptual	ADJ
fcis-31421	75	20	limitations	limitation	NOUN
fcis-31421	75	21	of	of	ADP
fcis-31421	75	22	a	a	DET
fcis-31421	75	23	single	single	ADJ
fcis-31421	75	24	data	datum	NOUN
fcis-31421	75	25	source	source	NOUN
fcis-31421	75	26	,	,	PUNCT
fcis-31421	75	27	allowing	allow	VERB
fcis-31421	75	28	for	for	ADP
fcis-31421	75	29	more	more	ADV
fcis-31421	75	30	accurate	accurate	ADJ
fcis-31421	75	31	detection	detection	NOUN
fcis-31421	75	32	,	,	PUNCT
fcis-31421	75	33	while	while	SCONJ
fcis-31421	75	34	supporting	support	VERB
fcis-31421	75	35	multitask	multitask	ADJ
fcis-31421	75	36	output	output	NOUN
fcis-31421	75	37	.	.	PUNCT
fcis-31421	76	1	in	in	ADP
fcis-31421	76	2	recent	recent	ADJ
fcis-31421	76	3	years	year	NOUN
fcis-31421	76	4	,	,	PUNCT
fcis-31421	76	5	several	several	ADJ
fcis-31421	76	6	scholars	scholar	NOUN
fcis-31421	76	7	have	have	AUX
fcis-31421	76	8	also	also	ADV
fcis-31421	76	9	conducted	conduct	VERB
fcis-31421	76	10	research	research	NOUN
fcis-31421	76	11	on	on	ADP
fcis-31421	76	12	multimodal	multimodal	NOUN
fcis-31421	76	13	deep	deep	ADJ
fcis-31421	76	14	learning	learning	NOUN
fcis-31421	76	15	for	for	ADP
fcis-31421	76	16	fire	fire	NOUN
fcis-31421	76	17	detection	detection	NOUN
fcis-31421	76	18	.	.	PUNCT
fcis-31421	77	1	for	for	ADP
fcis-31421	77	2	example	example	NOUN
fcis-31421	77	3	,	,	PUNCT
fcis-31421	77	4	sharma	sharma	PROPN
fcis-31421	77	5	a	a	PRON
fcis-31421	77	6	et	et	PROPN
fcis-31421	77	7	al	al	PROPN
fcis-31421	77	8	.	.	PROPN
fcis-31421	78	1	collected	collect	VERB
fcis-31421	78	2	multimodal	multimodal	NOUN
fcis-31421	78	3	datasets	dataset	NOUN
fcis-31421	78	4	using	use	VERB
fcis-31421	78	5	thermal	thermal	ADJ
fcis-31421	78	6	imaging	imaging	NOUN
fcis-31421	78	7	cameras	camera	NOUN
fcis-31421	78	8	that	that	PRON
fcis-31421	78	9	can	can	AUX
fcis-31421	78	10	detect	detect	VERB
fcis-31421	78	11	temperature	temperature	NOUN
fcis-31421	78	12	changes	change	NOUN
fcis-31421	78	13	.	.	PUNCT
fcis-31421	79	1	this	this	DET
fcis-31421	79	2	study	study	NOUN
fcis-31421	79	3	adopts	adopt	VERB
fcis-31421	79	4	a	a	DET
fcis-31421	79	5	multimodal	multimodal	ADJ
fcis-31421	79	6	data	datum	NOUN
fcis-31421	79	7	fusion	fusion	NOUN
fcis-31421	79	8	strategy	strategy	NOUN
fcis-31421	79	9	to	to	PART
fcis-31421	79	10	improve	improve	VERB
fcis-31421	79	11	detection	detection	NOUN
fcis-31421	79	12	performance	performance	NOUN
fcis-31421	79	13	through	through	ADP
fcis-31421	79	14	collaborative	collaborative	ADJ
fcis-31421	79	15	training	training	NOUN
fcis-31421	79	16	of	of	ADP
fcis-31421	79	17	thermal	thermal	ADJ
fcis-31421	79	18	imaging	imaging	ADJ
fcis-31421	79	19	visual	visual	ADJ
fcis-31421	79	20	data	datum	NOUN
fcis-31421	79	21	and	and	CCONJ
fcis-31421	79	22	fire	fire	NOUN
fcis-31421	79	23	sensor	sensor	NOUN
fcis-31421	79	24	data	datum	NOUN
fcis-31421	79	25	.	.	PUNCT
fcis-31421	80	1	in	in	ADP
fcis-31421	80	2	terms	term	NOUN
fcis-31421	80	3	of	of	ADP
fcis-31421	80	4	visual	visual	ADJ
fcis-31421	80	5	modality	modality	NOUN
fcis-31421	80	6	processing	processing	NOUN
fcis-31421	80	7	,	,	PUNCT
fcis-31421	80	8	the	the	DET
fcis-31421	80	9	research	research	NOUN
fcis-31421	80	10	team	team	NOUN
fcis-31421	80	11	trained	train	VERB
fcis-31421	80	12	multiple	multiple	ADJ
fcis-31421	80	13	improved	improve	VERB
fcis-31421	80	14	convolutional	convolutional	ADJ
fcis-31421	80	15	neural	neural	ADJ
fcis-31421	80	16	network	network	NOUN
fcis-31421	80	17	(	(	PUNCT
fcis-31421	80	18	cnn	cnn	PROPN
fcis-31421	80	19	)	)	PUNCT
fcis-31421	80	20	architectures	architecture	NOUN
fcis-31421	80	21	using	use	VERB
fcis-31421	80	22	image	image	NOUN
fcis-31421	80	23	datasets	dataset	NOUN
fcis-31421	80	24	collected	collect	VERB
fcis-31421	80	25	by	by	ADP
fcis-31421	80	26	thermal	thermal	ADJ
fcis-31421	80	27	imaging	imaging	NOUN
fcis-31421	80	28	cameras	camera	NOUN
fcis-31421	80	29	.	.	PUNCT
fcis-31421	81	1	at	at	ADP
fcis-31421	81	2	the	the	DET
fcis-31421	81	3	same	same	ADJ
fcis-31421	81	4	time	time	NOUN
fcis-31421	81	5	,	,	PUNCT
fcis-31421	81	6	for	for	ADP
fcis-31421	81	7	sensor	sensor	NOUN
fcis-31421	81	8	data	datum	NOUN
fcis-31421	81	9	streams	stream	NOUN
fcis-31421	81	10	,	,	PUNCT
fcis-31421	81	11	we	we	PRON
fcis-31421	81	12	innovatively	innovatively	ADV
fcis-31421	81	13	adopted	adopt	VERB
fcis-31421	81	14	bidirectional	bidirectional	ADJ
fcis-31421	81	15	long	long	ADJ
fcis-31421	81	16	short	short	ADJ
fcis-31421	81	17	-	-	PUNCT
fcis-31421	81	18	term	term	NOUN
fcis-31421	81	19	memory	memory	NOUN
fcis-31421	81	20	networks	network	NOUN
fcis-31421	81	21	(	(	PUNCT
fcis-31421	81	22	bilstm	bilstm	NOUN
fcis-31421	81	23	dense	dense	ADJ
fcis-31421	81	24	)	)	PUNCT
fcis-31421	81	25	and	and	CCONJ
fcis-31421	81	26	densely	densely	ADV
fcis-31421	81	27	connected	connect	VERB
fcis-31421	81	28	long	long	ADJ
fcis-31421	81	29	short	short	ADJ
fcis-31421	81	30	-	-	PUNCT
fcis-31421	81	31	term	term	NOUN
fcis-31421	81	32	memory	memory	NOUN
fcis-31421	81	33	networks	network	NOUN
fcis-31421	81	34	(	(	PUNCT
fcis-31421	81	35	lstm	lstm	NOUN
fcis-31421	81	36	densedensenet	densedensenet	PROPN
fcis-31421	81	37	201	201	NUM
fcis-31421	81	38	)	)	PUNCT
fcis-31421	81	39	for	for	ADP
fcis-31421	81	40	temporal	temporal	ADJ
fcis-31421	81	41	feature	feature	NOUN
fcis-31421	81	42	modeling	modeling	NOUN
fcis-31421	81	43	[	[	X
fcis-31421	81	44	5	5	NUM
fcis-31421	81	45	]	]	PUNCT
fcis-31421	81	46	.	.	PUNCT
fcis-31421	82	1	the	the	DET
fcis-31421	82	2	experimental	experimental	ADJ
fcis-31421	82	3	results	result	NOUN
fcis-31421	82	4	show	show	VERB
fcis-31421	82	5	that	that	SCONJ
fcis-31421	82	6	this	this	DET
fcis-31421	82	7	multi	multi	ADJ
fcis-31421	82	8	-	-	ADJ
fcis-31421	82	9	source	source	ADJ
fcis-31421	82	10	data	data	NOUN
fcis-31421	82	11	fusion	fusion	NOUN
fcis-31421	82	12	method	method	NOUN
fcis-31421	82	13	exhibits	exhibit	VERB
fcis-31421	82	14	significant	significant	ADJ
fcis-31421	82	15	performance	performance	NOUN
fcis-31421	82	16	advantages	advantage	NOUN
fcis-31421	82	17	:	:	PUNCT
fcis-31421	82	18	the	the	DET
fcis-31421	82	19	cnn	cnn	PROPN
fcis-31421	82	20	model	model	NOUN
fcis-31421	82	21	maintains	maintain	VERB
fcis-31421	82	22	a	a	DET
fcis-31421	82	23	classification	classification	NOUN
fcis-31421	82	24	accuracy	accuracy	NOUN
fcis-31421	82	25	of	of	ADP
fcis-31421	82	26	over	over	ADP
fcis-31421	82	27	0.99	0.99	NUM
fcis-31421	82	28	in	in	ADP
fcis-31421	82	29	visual	visual	ADJ
fcis-31421	82	30	modality	modality	NOUN
fcis-31421	82	31	,	,	PUNCT
fcis-31421	82	32	while	while	SCONJ
fcis-31421	82	33	the	the	DET
fcis-31421	82	34	three	three	NUM
fcis-31421	82	35	temporal	temporal	ADJ
fcis-31421	82	36	models	model	NOUN
fcis-31421	82	37	of	of	ADP
fcis-31421	82	38	dense	dense	ADJ
fcis-31421	82	39	,	,	PUNCT
fcis-31421	82	40	bilstm	bilstm	NOUN
fcis-31421	82	41	dense	dense	ADJ
fcis-31421	82	42	,	,	PUNCT
fcis-31421	82	43	and	and	CCONJ
fcis-31421	82	44	lstm	lstm	NOUN
fcis-31421	82	45	densedensenet	densedensenet	PROPN
fcis-31421	82	46	201	201	NUM
fcis-31421	82	47	perform	perform	VERB
fcis-31421	82	48	better	well	ADV
fcis-31421	82	49	in	in	ADP
fcis-31421	82	50	sensor	sensor	NOUN
fcis-31421	82	51	data	datum	NOUN
fcis-31421	82	52	processing	processing	NOUN
fcis-31421	82	53	than	than	ADP
fcis-31421	82	54	traditional	traditional	ADJ
fcis-31421	82	55	bilstm	bilstm	NOUN
fcis-31421	82	56	dense	dense	ADJ
fcis-31421	82	57	models	model	NOUN
fcis-31421	82	58	,	,	PUNCT
fcis-31421	82	59	with	with	ADP
fcis-31421	82	60	the	the	DET
fcis-31421	82	61	highest	high	ADJ
fcis-31421	82	62	accuracy	accuracy	NOUN
fcis-31421	82	63	reaching	reach	VERB
fcis-31421	82	64	93.39	93.39	NUM
fcis-31421	82	65	.	.	PUNCT
fcis-31421	83	1	the	the	DET
fcis-31421	83	2	system	system	NOUN
fcis-31421	83	3	ultimately	ultimately	ADV
fcis-31421	83	4	achieved	achieve	VERB
fcis-31421	83	5	a	a	DET
fcis-31421	83	6	verification	verification	NOUN
fcis-31421	83	7	accuracy	accuracy	NOUN
fcis-31421	83	8	of	of	ADP
fcis-31421	83	9	99.7	99.7	NUM
fcis-31421	83	10	%	%	NOUN
fcis-31421	83	11	,	,	PUNCT
fcis-31421	83	12	and	and	CCONJ
fcis-31421	83	13	the	the	DET
fcis-31421	83	14	verification	verification	NOUN
fcis-31421	83	15	loss	loss	NOUN
fcis-31421	83	16	was	be	AUX
fcis-31421	83	17	maintained	maintain	VERB
fcis-31421	83	18	at	at	ADP
fcis-31421	83	19	an	an	DET
fcis-31421	83	20	extremely	extremely	ADV
fcis-31421	83	21	low	low	ADJ
fcis-31421	83	22	level	level	NOUN
fcis-31421	83	23	,	,	PUNCT
fcis-31421	83	24	fully	fully	ADV
fcis-31421	83	25	demonstrating	demonstrate	VERB
fcis-31421	83	26	the	the	DET
fcis-31421	83	27	effectiveness	effectiveness	NOUN
fcis-31421	83	28	of	of	ADP
fcis-31421	83	29	this	this	DET
fcis-31421	83	30	scheme	scheme	NOUN
fcis-31421	83	31	in	in	ADP
fcis-31421	83	32	improving	improve	VERB
fcis-31421	83	33	detection	detection	NOUN
fcis-31421	83	34	accuracy	accuracy	NOUN
fcis-31421	83	35	and	and	CCONJ
fcis-31421	83	36	reducing	reduce	VERB
fcis-31421	83	37	false	false	ADJ
fcis-31421	83	38	alarm	alarm	NOUN
fcis-31421	83	39	rates	rate	NOUN
fcis-31421	83	40	.	.	PUNCT
fcis-31421	84	1	the	the	DET
fcis-31421	84	2	bhamra	bhamra	PROPN
fcis-31421	84	3	j	j	PROPN
fcis-31421	84	4	k	k	PROPN
fcis-31421	84	5	research	research	NOUN
fcis-31421	84	6	team	team	NOUN
fcis-31421	84	7	has	have	AUX
fcis-31421	84	8	developed	develop	VERB
fcis-31421	84	9	a	a	DET
fcis-31421	84	10	multimodal	multimodal	ADJ
fcis-31421	84	11	smokenet	smokenet	NOUN
fcis-31421	84	12	system	system	NOUN
fcis-31421	84	13	and	and	CCONJ
fcis-31421	84	14	its	its	PRON
fcis-31421	84	15	integrated	integrate	VERB
fcis-31421	84	16	version	version	NOUN
fcis-31421	84	17	smokenet	smokenet	NOUN
fcis-31421	84	18	ensemble	ensemble	ADJ
fcis-31421	84	19	,	,	PUNCT
fcis-31421	84	20	specifically	specifically	ADV
fcis-31421	84	21	designed	design	VERB
fcis-31421	84	22	for	for	ADP
fcis-31421	84	23	multi	multi	ADJ
fcis-31421	84	24	-	-	ADJ
fcis-31421	84	25	source	source	ADJ
fcis-31421	84	26	monitoring	monitoring	NOUN
fcis-31421	84	27	of	of	ADP
fcis-31421	84	28	wildfire	wildfire	NOUN
fcis-31421	84	29	smoke	smoke	NOUN
fcis-31421	84	30	.	.	PUNCT
fcis-31421	85	1	this	this	DET
fcis-31421	85	2	system	system	NOUN
fcis-31421	85	3	innovatively	innovatively	ADV
fcis-31421	85	4	integrates	integrate	VERB
fcis-31421	85	5	three	three	NUM
fcis-31421	85	6	heterogeneous	heterogeneous	ADJ
fcis-31421	85	7	data	datum	NOUN
fcis-31421	85	8	sources	source	NOUN
fcis-31421	85	9	:	:	PUNCT
fcis-31421	85	10	satellite	satellite	NOUN
fcis-31421	85	11	remote	remote	ADJ
fcis-31421	85	12	sensing	sense	VERB
fcis-31421	85	13	data	datum	NOUN
fcis-31421	85	14	,	,	PUNCT
fcis-31421	85	15	meteorological	meteorological	ADJ
fcis-31421	85	16	sensor	sensor	NOUN
fcis-31421	85	17	parameters	parameter	NOUN
fcis-31421	85	18	,	,	PUNCT
fcis-31421	85	19	and	and	CCONJ
fcis-31421	85	20	optical	optical	ADJ
fcis-31421	85	21	camera	camera	NOUN
fcis-31421	85	22	images	image	NOUN
fcis-31421	85	23	.	.	PUNCT
fcis-31421	86	1	based	base	VERB
fcis-31421	86	2	on	on	ADP
fcis-31421	86	3	the	the	DET
fcis-31421	86	4	original	original	ADJ
fcis-31421	86	5	smokenet	smokenet	NOUN
fcis-31421	86	6	architecture	architecture	NOUN
fcis-31421	86	7	,	,	PUNCT
fcis-31421	86	8	the	the	DET
fcis-31421	86	9	study	study	NOUN
fcis-31421	86	10	systematically	systematically	ADV
fcis-31421	86	11	evaluated	evaluate	VERB
fcis-31421	86	12	the	the	DET
fcis-31421	86	13	improvement	improvement	NOUN
fcis-31421	86	14	effect	effect	NOUN
fcis-31421	86	15	of	of	ADP
fcis-31421	86	16	multisource	multisource	ADJ
fcis-31421	86	17	information	information	NOUN
fcis-31421	86	18	fusion	fusion	NOUN
fcis-31421	86	19	on	on	ADP
fcis-31421	86	20	fire	fire	NOUN
fcis-31421	86	21	detection	detection	NOUN
fcis-31421	86	22	accuracy	accuracy	NOUN
fcis-31421	86	23	and	and	CCONJ
fcis-31421	86	24	timeliness	timeliness	NOUN
fcis-31421	86	25	by	by	ADP
fcis-31421	86	26	introducing	introduce	VERB
fcis-31421	86	27	meteorological	meteorological	ADJ
fcis-31421	86	28	observation	observation	NOUN
fcis-31421	86	29	and	and	CCONJ
fcis-31421	86	30	satellite	satellite	NOUN
fcis-31421	86	31	fire	fire	NOUN
fcis-31421	86	32	point	point	NOUN
fcis-31421	86	33	detection	detection	NOUN
fcis-31421	86	34	data	datum	NOUN
fcis-31421	86	35	.	.	PUNCT
fcis-31421	87	1	the	the	DET
fcis-31421	87	2	experimental	experimental	ADJ
fcis-31421	87	3	design	design	NOUN
fcis-31421	87	4	focuses	focus	VERB
fcis-31421	87	5	on	on	ADP
fcis-31421	87	6	analyzing	analyze	VERB
fcis-31421	87	7	the	the	DET
fcis-31421	87	8	impact	impact	NOUN
fcis-31421	87	9	mechanism	mechanism	NOUN
fcis-31421	87	10	of	of	ADP
fcis-31421	87	11	new	new	ADJ
fcis-31421	87	12	data	datum	NOUN
fcis-31421	87	13	54	54	NUM
fcis-31421	87	14	dimensions	dimension	NOUN
fcis-31421	87	15	on	on	ADP
fcis-31421	87	16	system	system	NOUN
fcis-31421	87	17	performance	performance	NOUN
fcis-31421	87	18	indicators	indicator	NOUN
fcis-31421	87	19	,	,	PUNCT
fcis-31421	87	20	providing	provide	VERB
fcis-31421	87	21	important	important	ADJ
fcis-31421	87	22	basis	basis	NOUN
fcis-31421	87	23	for	for	ADP
fcis-31421	87	24	the	the	DET
fcis-31421	87	25	optimization	optimization	NOUN
fcis-31421	87	26	of	of	ADP
fcis-31421	87	27	multimodal	multimodal	NOUN
fcis-31421	87	28	fire	fire	NOUN
fcis-31421	87	29	warning	warning	NOUN
fcis-31421	87	30	systems	system	NOUN
fcis-31421	87	31	[	[	X
fcis-31421	87	32	6	6	NUM
fcis-31421	87	33	]	]	PUNCT
fcis-31421	87	34	.	.	PUNCT
fcis-31421	88	1	in	in	ADP
fcis-31421	88	2	their	their	PRON
fcis-31421	88	3	experiments	experiment	NOUN
fcis-31421	88	4	,	,	PUNCT
fcis-31421	88	5	they	they	PRON
fcis-31421	88	6	used	use	VERB
fcis-31421	88	7	three	three	NUM
fcis-31421	88	8	model	model	NOUN
fcis-31421	88	9	architectures	architecture	NOUN
fcis-31421	88	10	,	,	PUNCT
fcis-31421	88	11	smokeynet	smokeynet	NOUN
fcis-31421	88	12	,	,	PUNCT
fcis-31421	88	13	smokeynet	smokeynet	NOUN
fcis-31421	88	14	ensemble	ensemble	ADJ
fcis-31421	88	15	,	,	PUNCT
fcis-31421	88	16	and	and	CCONJ
fcis-31421	88	17	multimodal	multimodal	NOUN
fcis-31421	88	18	smokeynet	smokeynet	NOUN
fcis-31421	88	19	.	.	PUNCT
fcis-31421	89	1	the	the	DET
fcis-31421	89	2	fusion	fusion	NOUN
fcis-31421	89	3	of	of	ADP
fcis-31421	89	4	the	the	DET
fcis-31421	89	5	three	three	NUM
fcis-31421	89	6	models	model	NOUN
fcis-31421	89	7	resulted	result	VERB
fcis-31421	89	8	in	in	ADP
fcis-31421	89	9	an	an	DET
fcis-31421	89	10	average	average	ADJ
fcis-31421	89	11	increase	increase	NOUN
fcis-31421	89	12	in	in	ADP
fcis-31421	89	13	the	the	DET
fcis-31421	89	14	speed	speed	NOUN
fcis-31421	89	15	of	of	ADP
fcis-31421	89	16	detecting	detect	VERB
fcis-31421	89	17	smoke	smoke	NOUN
fcis-31421	89	18	by	by	ADP
fcis-31421	89	19	13.6	13.6	NUM
fcis-31421	89	20	%	%	NOUN
fcis-31421	89	21	and	and	CCONJ
fcis-31421	89	22	a	a	DET
fcis-31421	89	23	reduction	reduction	NOUN
fcis-31421	89	24	in	in	ADP
fcis-31421	89	25	the	the	DET
fcis-31421	89	26	standard	standard	ADJ
fcis-31421	89	27	deviation	deviation	NOUN
fcis-31421	89	28	by	by	ADP
fcis-31421	89	29	0.32	0.32	NUM
fcis-31421	89	30	.	.	PUNCT
fcis-31421	90	1	the	the	DET
fcis-31421	90	2	results	result	NOUN
fcis-31421	90	3	are	be	AUX
fcis-31421	90	4	summarized	summarize	VERB
fcis-31421	90	5	in	in	ADP
fcis-31421	90	6	the	the	DET
fcis-31421	90	7	following	follow	VERB
fcis-31421	90	8	table	table	NOUN
fcis-31421	90	9	.	.	PUNCT
fcis-31421	91	1	the	the	DET
fcis-31421	91	2	toulouse	toulouse	NOUN
fcis-31421	91	3	t	t	PROPN
fcis-31421	91	4	research	research	NOUN
fcis-31421	91	5	team	team	NOUN
fcis-31421	91	6	has	have	AUX
fcis-31421	91	7	developed	develop	VERB
fcis-31421	91	8	an	an	DET
fcis-31421	91	9	innovative	innovative	ADJ
fcis-31421	91	10	multimodal	multimodal	NOUN
fcis-31421	91	11	3d	3d	NUM
fcis-31421	91	12	visual	visual	ADJ
fcis-31421	91	13	analysis	analysis	NOUN
fcis-31421	91	14	system	system	NOUN
fcis-31421	91	15	specifically	specifically	ADV
fcis-31421	91	16	designed	design	VERB
fcis-31421	91	17	for	for	ADP
fcis-31421	91	18	geometric	geometric	ADJ
fcis-31421	91	19	feature	feature	NOUN
fcis-31421	91	20	extraction	extraction	NOUN
fcis-31421	91	21	in	in	ADP
fcis-31421	91	22	fire	fire	NOUN
fcis-31421	91	23	scenes	scene	NOUN
fcis-31421	91	24	.	.	PUNCT
fcis-31421	92	1	the	the	DET
fcis-31421	92	2	system	system	NOUN
fcis-31421	92	3	achieves	achieve	VERB
fcis-31421	92	4	accurate	accurate	ADJ
fcis-31421	92	5	modeling	modeling	NOUN
fcis-31421	92	6	of	of	ADP
fcis-31421	92	7	the	the	DET
fcis-31421	92	8	three	three	NUM
fcis-31421	92	9	-	-	PUNCT
fcis-31421	92	10	dimensional	dimensional	ADJ
fcis-31421	92	11	spatial	spatial	ADJ
fcis-31421	92	12	characteristics	characteristic	NOUN
fcis-31421	92	13	of	of	ADP
fcis-31421	92	14	fires	fire	NOUN
fcis-31421	92	15	by	by	ADP
fcis-31421	92	16	integrating	integrate	VERB
fcis-31421	92	17	data	datum	NOUN
fcis-31421	92	18	collected	collect	VERB
fcis-31421	92	19	from	from	ADP
fcis-31421	92	20	multiple	multiple	ADJ
fcis-31421	92	21	3d	3d	NUM
fcis-31421	92	22	vision	vision	NOUN
fcis-31421	92	23	devices	device	NOUN
fcis-31421	92	24	.	.	PUNCT
fcis-31421	93	1	based	base	VERB
fcis-31421	93	2	on	on	ADP
fcis-31421	93	3	the	the	DET
fcis-31421	93	4	registered	register	VERB
fcis-31421	93	5	3d	3d	NUM
fcis-31421	93	6	fire	fire	NOUN
fcis-31421	93	7	point	point	NOUN
fcis-31421	93	8	data	datum	NOUN
fcis-31421	93	9	,	,	PUNCT
fcis-31421	93	10	researchers	researcher	NOUN
fcis-31421	93	11	have	have	AUX
fcis-31421	93	12	established	establish	VERB
fcis-31421	93	13	a	a	DET
fcis-31421	93	14	complete	complete	ADJ
fcis-31421	93	15	method	method	NOUN
fcis-31421	93	16	for	for	ADP
fcis-31421	93	17	calculating	calculate	VERB
fcis-31421	93	18	the	the	DET
fcis-31421	93	19	geometric	geometric	ADJ
fcis-31421	93	20	features	feature	NOUN
fcis-31421	93	21	of	of	ADP
fcis-31421	93	22	fires	fire	NOUN
fcis-31421	93	23	.	.	PUNCT
fcis-31421	94	1	this	this	DET
fcis-31421	94	2	method	method	NOUN
fcis-31421	94	3	extends	extend	VERB
fcis-31421	94	4	the	the	DET
fcis-31421	94	5	monocular	monocular	ADJ
fcis-31421	94	6	stereo	stereo	NOUN
fcis-31421	94	7	vision	vision	NOUN
fcis-31421	94	8	technology	technology	NOUN
fcis-31421	94	9	previously	previously	ADV
fcis-31421	94	10	developed	develop	VERB
fcis-31421	94	11	by	by	ADP
fcis-31421	94	12	the	the	DET
fcis-31421	94	13	author	author	NOUN
fcis-31421	94	14	.	.	PUNCT
fcis-31421	95	1	by	by	ADP
fcis-31421	95	2	integrating	integrate	VERB
fcis-31421	95	3	the	the	DET
fcis-31421	95	4	three	three	NUM
fcis-31421	95	5	-	-	PUNCT
fcis-31421	95	6	dimensional	dimensional	ADJ
fcis-31421	95	7	point	point	NOUN
fcis-31421	95	8	cloud	cloud	NOUN
fcis-31421	95	9	data	datum	NOUN
fcis-31421	95	10	obtained	obtain	VERB
fcis-31421	95	11	from	from	ADP
fcis-31421	95	12	the	the	DET
fcis-31421	95	13	dual	dual	ADJ
fcis-31421	95	14	vision	vision	NOUN
fcis-31421	95	15	system	system	NOUN
fcis-31421	95	16	,	,	PUNCT
fcis-31421	95	17	it	it	PRON
fcis-31421	95	18	can	can	AUX
fcis-31421	95	19	accurately	accurately	ADV
fcis-31421	95	20	reconstruct	reconstruct	VERB
fcis-31421	95	21	the	the	DET
fcis-31421	95	22	spatial	spatial	ADJ
fcis-31421	95	23	characteristics	characteristic	NOUN
fcis-31421	95	24	of	of	ADP
fcis-31421	95	25	the	the	DET
fcis-31421	95	26	fire	fire	NOUN
fcis-31421	95	27	scene	scene	NOUN
fcis-31421	95	28	,	,	PUNCT
fcis-31421	95	29	including	include	VERB
fcis-31421	95	30	key	key	ADJ
fcis-31421	95	31	parameters	parameter	NOUN
fcis-31421	95	32	such	such	ADJ
fcis-31421	95	33	as	as	ADP
fcis-31421	95	34	the	the	DET
fcis-31421	95	35	propagation	propagation	NOUN
fcis-31421	95	36	plane	plane	NOUN
fcis-31421	95	37	parameters	parameter	NOUN
fcis-31421	95	38	of	of	ADP
fcis-31421	95	39	the	the	DET
fcis-31421	95	40	fire	fire	NOUN
fcis-31421	95	41	front	front	NOUN
fcis-31421	95	42	,	,	PUNCT
fcis-31421	95	43	spatial	spatial	ADJ
fcis-31421	95	44	position	position	NOUN
fcis-31421	95	45	coordinates	coordinate	NOUN
fcis-31421	95	46	,	,	PUNCT
fcis-31421	95	47	spread	spread	ADJ
fcis-31421	95	48	rate	rate	NOUN
fcis-31421	95	49	,	,	PUNCT
fcis-31421	95	50	flame	flame	NOUN
fcis-31421	95	51	height	height	NOUN
fcis-31421	95	52	,	,	PUNCT
fcis-31421	95	53	burning	burn	VERB
fcis-31421	95	54	surface	surface	NOUN
fcis-31421	95	55	morphology	morphology	NOUN
fcis-31421	95	56	and	and	CCONJ
fcis-31421	95	57	tilt	tilt	NOUN
fcis-31421	95	58	angle	angle	NOUN
fcis-31421	95	59	,	,	PUNCT
fcis-31421	95	60	fire	fire	NOUN
fcis-31421	95	61	perimeter	perimeter	NOUN
fcis-31421	95	62	,	,	PUNCT
fcis-31421	95	63	burning	burning	NOUN
fcis-31421	95	64	area	area	NOUN
fcis-31421	95	65	,	,	PUNCT
fcis-31421	95	66	main	main	ADJ
fcis-31421	95	67	direction	direction	NOUN
fcis-31421	95	68	of	of	ADP
fcis-31421	95	69	fire	fire	NOUN
fcis-31421	95	70	spread	spread	NOUN
fcis-31421	95	71	,	,	PUNCT
fcis-31421	95	72	and	and	CCONJ
fcis-31421	95	73	overall	overall	ADJ
fcis-31421	95	74	fire	fire	NOUN
fcis-31421	95	75	volume	volume	NOUN
fcis-31421	95	76	estimation	estimation	NOUN
fcis-31421	95	77	[	[	X
fcis-31421	95	78	7	7	NUM
fcis-31421	95	79	]	]	PUNCT
fcis-31421	95	80	.	.	PUNCT
fcis-31421	96	1	such	such	ADJ
fcis-31421	96	2	multimodal	multimodal	NOUN
fcis-31421	96	3	fusion	fusion	NOUN
fcis-31421	96	4	is	be	AUX
fcis-31421	96	5	not	not	PART
fcis-31421	96	6	only	only	ADV
fcis-31421	96	7	able	able	ADJ
fcis-31421	96	8	to	to	PART
fcis-31421	96	9	detect	detect	VERB
fcis-31421	96	10	fire	fire	NOUN
fcis-31421	96	11	effectively	effectively	ADV
fcis-31421	96	12	,	,	PUNCT
fcis-31421	96	13	but	but	CCONJ
fcis-31421	96	14	also	also	ADV
fcis-31421	96	15	able	able	ADJ
fcis-31421	96	16	to	to	PART
fcis-31421	96	17	predict	predict	VERB
fcis-31421	96	18	flame	flame	NOUN
fcis-31421	96	19	propagation	propagation	NOUN
fcis-31421	96	20	,	,	PUNCT
fcis-31421	96	21	which	which	PRON
fcis-31421	96	22	is	be	AUX
fcis-31421	96	23	very	very	ADV
fcis-31421	96	24	powerful	powerful	ADJ
fcis-31421	96	25	.	.	PUNCT
fcis-31421	97	1	however	however	ADV
fcis-31421	97	2	,	,	PUNCT
fcis-31421	97	3	two	two	NUM
fcis-31421	97	4	multispectral	multispectral	ADJ
fcis-31421	97	5	cameras	camera	NOUN
fcis-31421	97	6	(	(	PUNCT
fcis-31421	97	7	jai	jai	PROPN
fcis-31421	97	8	ad-080	ad-080	PROPN
fcis-31421	97	9	ge	ge	PROPN
fcis-31421	97	10	)	)	PUNCT
fcis-31421	97	11	are	be	AUX
fcis-31421	97	12	used	use	VERB
fcis-31421	97	13	in	in	ADP
fcis-31421	97	14	the	the	DET
fcis-31421	97	15	stereo	stereo	PROPN
fcis-31421	97	16	vision	vision	NOUN
fcis-31421	97	17	system	system	NOUN
fcis-31421	97	18	,	,	PUNCT
fcis-31421	97	19	which	which	PRON
fcis-31421	97	20	makes	make	VERB
fcis-31421	97	21	the	the	DET
fcis-31421	97	22	cost	cost	NOUN
fcis-31421	97	23	much	much	ADV
fcis-31421	97	24	higher	high	ADJ
fcis-31421	97	25	compared	compare	VERB
fcis-31421	97	26	to	to	ADP
fcis-31421	97	27	other	other	ADJ
fcis-31421	97	28	large	large	ADJ
fcis-31421	97	29	-	-	PUNCT
fcis-31421	97	30	model	model	NOUN
fcis-31421	97	31	detections	detection	NOUN
fcis-31421	97	32	,	,	PUNCT
fcis-31421	97	33	and	and	CCONJ
fcis-31421	97	34	this	this	DET
fcis-31421	97	35	high	high	ADJ
fcis-31421	97	36	cost	cost	NOUN
fcis-31421	97	37	makes	make	VERB
fcis-31421	97	38	it	it	PRON
fcis-31421	97	39	difficult	difficult	ADJ
fcis-31421	97	40	to	to	PART
fcis-31421	97	41	detect	detect	VERB
fcis-31421	97	42	fires	fire	NOUN
fcis-31421	97	43	on	on	ADP
fcis-31421	97	44	a	a	DET
fcis-31421	97	45	large	large	ADJ
fcis-31421	97	46	scale	scale	NOUN
fcis-31421	97	47	.	.	PUNCT
fcis-31421	98	1	chenyu	chenyu	PROPN
fcis-31421	98	2	chaoxia	chaoxia	PROPN
fcis-31421	98	3	's	's	PART
fcis-31421	98	4	research	research	NOUN
fcis-31421	98	5	team	team	NOUN
fcis-31421	98	6	has	have	AUX
fcis-31421	98	7	innovatively	innovatively	ADV
fcis-31421	98	8	developed	develop	VERB
fcis-31421	98	9	a	a	DET
fcis-31421	98	10	flame	flame	NOUN
fcis-31421	98	11	detection	detection	NOUN
fcis-31421	98	12	framework	framework	NOUN
fcis-31421	98	13	that	that	PRON
fcis-31421	98	14	integrates	integrate	VERB
fcis-31421	98	15	multimodal	multimodal	ADJ
fcis-31421	98	16	information	information	NOUN
fcis-31421	98	17	.	.	PUNCT
fcis-31421	99	1	this	this	DET
fcis-31421	99	2	model	model	NOUN
fcis-31421	99	3	uses	use	VERB
fcis-31421	99	4	thermal	thermal	ADJ
fcis-31421	99	5	imaging	imaging	NOUN
fcis-31421	99	6	projection	projection	NOUN
fcis-31421	99	7	technology	technology	NOUN
fcis-31421	99	8	to	to	PART
fcis-31421	99	9	preliminarily	preliminarily	ADV
fcis-31421	99	10	locate	locate	VERB
fcis-31421	99	11	the	the	DET
fcis-31421	99	12	flame	flame	NOUN
fcis-31421	99	13	area	area	NOUN
fcis-31421	99	14	and	and	CCONJ
fcis-31421	99	15	capture	capture	VERB
fcis-31421	99	16	the	the	DET
fcis-31421	99	17	surrounding	surround	VERB
fcis-31421	99	18	features	feature	NOUN
fcis-31421	99	19	of	of	ADP
fcis-31421	99	20	the	the	DET
fcis-31421	99	21	flame	flame	NOUN
fcis-31421	99	22	through	through	ADP
fcis-31421	99	23	neighborhood	neighborhood	NOUN
fcis-31421	99	24	sampling	sampling	NOUN
fcis-31421	99	25	mechanism	mechanism	NOUN
fcis-31421	99	26	.	.	PUNCT
fcis-31421	100	1	the	the	DET
fcis-31421	100	2	research	research	NOUN
fcis-31421	100	3	team	team	NOUN
fcis-31421	100	4	has	have	AUX
fcis-31421	100	5	specially	specially	ADV
fcis-31421	100	6	designed	design	VERB
fcis-31421	100	7	an	an	DET
fcis-31421	100	8	attention	attention	NOUN
fcis-31421	100	9	guidance	guidance	NOUN
fcis-31421	100	10	mechanism	mechanism	NOUN
fcis-31421	100	11	based	base	VERB
fcis-31421	100	12	on	on	ADP
fcis-31421	100	13	index	index	NOUN
fcis-31421	100	14	matching	matching	NOUN
fcis-31421	100	15	,	,	PUNCT
fcis-31421	100	16	using	use	VERB
fcis-31421	100	17	attention	attention	NOUN
fcis-31421	100	18	weight	weight	NOUN
fcis-31421	100	19	maps	map	NOUN
fcis-31421	100	20	generated	generate	VERB
fcis-31421	100	21	from	from	ADP
fcis-31421	100	22	thermal	thermal	ADJ
fcis-31421	100	23	imaging	imaging	NOUN
fcis-31421	100	24	data	datum	NOUN
fcis-31421	100	25	to	to	PART
fcis-31421	100	26	enhance	enhance	VERB
fcis-31421	100	27	the	the	DET
fcis-31421	100	28	ability	ability	NOUN
fcis-31421	100	29	to	to	PART
fcis-31421	100	30	extract	extract	VERB
fcis-31421	100	31	regional	regional	ADJ
fcis-31421	100	32	features	feature	NOUN
fcis-31421	100	33	of	of	ADP
fcis-31421	100	34	visible	visible	ADJ
fcis-31421	100	35	light	light	ADJ
fcis-31421	100	36	modes	mode	NOUN
fcis-31421	100	37	.	.	PUNCT
fcis-31421	101	1	the	the	DET
fcis-31421	101	2	experimental	experimental	ADJ
fcis-31421	101	3	results	result	NOUN
fcis-31421	101	4	show	show	VERB
fcis-31421	101	5	that	that	SCONJ
fcis-31421	101	6	this	this	DET
fcis-31421	101	7	multimodal	multimodal	ADJ
fcis-31421	101	8	fusion	fusion	NOUN
fcis-31421	101	9	method	method	NOUN
fcis-31421	101	10	significantly	significantly	ADV
fcis-31421	101	11	outperforms	outperform	VERB
fcis-31421	101	12	the	the	DET
fcis-31421	101	13	traditional	traditional	ADJ
fcis-31421	101	14	single	single	ADJ
fcis-31421	101	15	modal	modal	NOUN
fcis-31421	101	16	faster	fast	ADJ
fcis-31421	101	17	r	r	NOUN
fcis-31421	101	18	-	-	PUNCT
fcis-31421	101	19	cnn	cnn	PROPN
fcis-31421	101	20	model	model	NOUN
fcis-31421	101	21	in	in	ADP
fcis-31421	101	22	terms	term	NOUN
fcis-31421	101	23	of	of	ADP
fcis-31421	101	24	detection	detection	NOUN
fcis-31421	101	25	performance	performance	NOUN
fcis-31421	101	26	[	[	X
fcis-31421	101	27	8	8	NUM
fcis-31421	101	28	]	]	PUNCT
fcis-31421	101	29	.	.	PUNCT
fcis-31421	102	1	however	however	ADV
fcis-31421	102	2	,	,	PUNCT
fcis-31421	102	3	the	the	DET
fcis-31421	102	4	device	device	NOUN
fcis-31421	102	5	requires	require	VERB
fcis-31421	102	6	that	that	SCONJ
fcis-31421	102	7	the	the	DET
fcis-31421	102	8	positional	positional	ADJ
fcis-31421	102	9	offsets	offset	NOUN
fcis-31421	102	10	and	and	CCONJ
fcis-31421	102	11	orientations	orientation	NOUN
fcis-31421	102	12	of	of	ADP
fcis-31421	102	13	the	the	DET
fcis-31421	102	14	different	different	ADJ
fcis-31421	102	15	modal	modal	ADJ
fcis-31421	102	16	cameras	camera	NOUN
fcis-31421	102	17	are	be	AUX
fcis-31421	102	18	not	not	PART
fcis-31421	102	19	too	too	ADV
fcis-31421	102	20	large	large	ADJ
fcis-31421	102	21	and	and	CCONJ
fcis-31421	102	22	that	that	SCONJ
fcis-31421	102	23	the	the	DET
fcis-31421	102	24	flame	flame	NOUN
fcis-31421	102	25	target	target	NOUN
fcis-31421	102	26	is	be	AUX
fcis-31421	102	27	within	within	ADP
fcis-31421	102	28	the	the	DET
fcis-31421	102	29	field	field	NOUN
fcis-31421	102	30	of	of	ADP
fcis-31421	102	31	view	view	NOUN
fcis-31421	102	32	of	of	ADP
fcis-31421	102	33	both	both	DET
fcis-31421	102	34	cameras	camera	NOUN
fcis-31421	102	35	.	.	PUNCT
fcis-31421	103	1	further	further	ADJ
fcis-31421	103	2	research	research	NOUN
fcis-31421	103	3	is	be	AUX
fcis-31421	103	4	needed	need	VERB
fcis-31421	103	5	for	for	ADP
fcis-31421	103	6	the	the	DET
fcis-31421	103	7	case	case	NOUN
fcis-31421	103	8	where	where	SCONJ
fcis-31421	103	9	the	the	DET
fcis-31421	103	10	multimodal	multimodal	NOUN
fcis-31421	103	11	images	image	NOUN
fcis-31421	103	12	are	be	AUX
fcis-31421	103	13	very	very	ADV
fcis-31421	103	14	poorly	poorly	ADV
fcis-31421	103	15	aligned	align	VERB
fcis-31421	103	16	or	or	CCONJ
fcis-31421	103	17	unpaired	unpaired	ADJ
fcis-31421	103	18	.	.	PUNCT
fcis-31421	104	1	multi	multi	ADJ
fcis-31421	104	2	-	-	ADJ
fcis-31421	104	3	modal	modal	ADJ
fcis-31421	104	4	fusion	fusion	NOUN
fcis-31421	104	5	has	have	VERB
fcis-31421	104	6	a	a	DET
fcis-31421	104	7	very	very	ADV
fcis-31421	104	8	strong	strong	ADJ
fcis-31421	104	9	performance	performance	NOUN
fcis-31421	104	10	breakthrough	breakthrough	VERB
fcis-31421	104	11	single	single	ADJ
fcis-31421	104	12	-	-	PUNCT
fcis-31421	104	13	modal	modal	NOUN
fcis-31421	104	14	perception	perception	NOUN
fcis-31421	104	15	limitations	limitation	NOUN
fcis-31421	104	16	,	,	PUNCT
fcis-31421	104	17	in	in	ADP
fcis-31421	104	18	haze	haze	NOUN
fcis-31421	104	19	,	,	PUNCT
fcis-31421	104	20	night	night	NOUN
fcis-31421	104	21	and	and	CCONJ
fcis-31421	104	22	other	other	ADJ
fcis-31421	104	23	scenes	scene	NOUN
fcis-31421	104	24	,	,	PUNCT
fcis-31421	104	25	through	through	ADP
fcis-31421	104	26	the	the	DET
fcis-31421	104	27	infrared	infrared	ADJ
fcis-31421	104	28	and	and	CCONJ
fcis-31421	104	29	sensor	sensor	NOUN
fcis-31421	104	30	data	datum	NOUN
fcis-31421	104	31	can	can	AUX
fcis-31421	104	32	also	also	ADV
fcis-31421	104	33	maintain	maintain	VERB
fcis-31421	104	34	detection	detection	NOUN
fcis-31421	104	35	capabilities	capability	NOUN
fcis-31421	104	36	.	.	PUNCT
fcis-31421	105	1	in	in	ADP
fcis-31421	105	2	fire	fire	NOUN
fcis-31421	105	3	detection	detection	NOUN
fcis-31421	105	4	,	,	PUNCT
fcis-31421	105	5	it	it	PRON
fcis-31421	105	6	can	can	AUX
fcis-31421	105	7	also	also	ADV
fcis-31421	105	8	support	support	VERB
fcis-31421	105	9	multi	multi	ADJ
fcis-31421	105	10	-	-	ADJ
fcis-31421	105	11	task	task	ADJ
fcis-31421	105	12	outputs	output	NOUN
fcis-31421	105	13	such	such	ADJ
fcis-31421	105	14	as	as	ADP
fcis-31421	105	15	fire	fire	NOUN
fcis-31421	105	16	source	source	NOUN
fcis-31421	105	17	localization	localization	NOUN
fcis-31421	105	18	,	,	PUNCT
fcis-31421	105	19	fire	fire	NOUN
fcis-31421	105	20	classification	classification	NOUN
fcis-31421	105	21	,	,	PUNCT
fcis-31421	105	22	and	and	CCONJ
fcis-31421	105	23	spread	spread	VERB
fcis-31421	105	24	prediction	prediction	NOUN
fcis-31421	105	25	.	.	PUNCT
fcis-31421	106	1	however	however	ADV
fcis-31421	106	2	,	,	PUNCT
fcis-31421	106	3	multimodal	multimodal	ADJ
fcis-31421	106	4	parallel	parallel	NOUN
fcis-31421	106	5	processing	processing	NOUN
fcis-31421	106	6	requires	require	VERB
fcis-31421	106	7	gpu	gpu	NOUN
fcis-31421	106	8	cluster	cluster	NOUN
fcis-31421	106	9	support	support	NOUN
fcis-31421	106	10	,	,	PUNCT
fcis-31421	106	11	difficult	difficult	ADJ
fcis-31421	106	12	to	to	PART
fcis-31421	106	13	deploy	deploy	VERB
fcis-31421	106	14	in	in	ADP
fcis-31421	106	15	the	the	DET
fcis-31421	106	16	edge	edge	NOUN
fcis-31421	106	17	devices	device	NOUN
fcis-31421	106	18	,	,	PUNCT
fcis-31421	106	19	such	such	ADJ
fcis-31421	106	20	as	as	ADP
fcis-31421	106	21	transformer	transformer	ADJ
fcis-31421	106	22	fusion	fusion	NOUN
fcis-31421	106	23	model	model	NOUN
fcis-31421	106	24	training	training	NOUN
fcis-31421	106	25	time	time	NOUN
fcis-31421	106	26	consuming	consume	VERB
fcis-31421	106	27	3	3	NUM
fcis-31421	106	28	to	to	PART
fcis-31421	106	29	5	5	NUM
fcis-31421	106	30	times	time	NOUN
fcis-31421	106	31	that	that	DET
fcis-31421	106	32	of	of	ADP
fcis-31421	106	33	a	a	DET
fcis-31421	106	34	unimodal	unimodal	ADJ
fcis-31421	106	35	cnn	cnn	NOUN
fcis-31421	106	36	,	,	PUNCT
fcis-31421	106	37	and	and	CCONJ
fcis-31421	106	38	at	at	ADP
fcis-31421	106	39	the	the	DET
fcis-31421	106	40	same	same	ADJ
fcis-31421	106	41	time	time	NOUN
fcis-31421	106	42	its	its	PRON
fcis-31421	106	43	application	application	NOUN
fcis-31421	106	44	to	to	PART
fcis-31421	106	45	detect	detect	VERB
fcis-31421	106	46	fires	fire	NOUN
fcis-31421	106	47	will	will	AUX
fcis-31421	106	48	make	make	VERB
fcis-31421	106	49	the	the	DET
fcis-31421	106	50	cost	cost	NOUN
fcis-31421	106	51	very	very	ADV
fcis-31421	106	52	high	high	ADJ
fcis-31421	106	53	,	,	PUNCT
fcis-31421	106	54	for	for	ADP
fcis-31421	106	55	example	example	NOUN
fcis-31421	106	56	,	,	PUNCT
fcis-31421	106	57	the	the	DET
fcis-31421	106	58	camera	camera	NOUN
fcis-31421	106	59	,	,	PUNCT
fcis-31421	106	60	you	you	PRON
fcis-31421	106	61	need	need	VERB
fcis-31421	106	62	to	to	PART
fcis-31421	106	63	label	label	VERB
fcis-31421	106	64	the	the	DET
fcis-31421	106	65	flame	flame	NOUN
fcis-31421	106	66	region	region	NOUN
fcis-31421	106	67	in	in	ADP
fcis-31421	106	68	the	the	DET
fcis-31421	106	69	image	image	NOUN
fcis-31421	106	70	,	,	PUNCT
fcis-31421	106	71	the	the	DET
fcis-31421	106	72	sensor	sensor	NOUN
fcis-31421	106	73	value	value	NOUN
fcis-31421	106	74	and	and	CCONJ
fcis-31421	106	75	its	its	PRON
fcis-31421	106	76	correlation	correlation	NOUN
fcis-31421	106	77	relationship	relationship	NOUN
fcis-31421	106	78	labeling	labeling	NOUN
fcis-31421	106	79	costs	cost	NOUN
fcis-31421	106	80	,	,	PUNCT
fcis-31421	106	81	which	which	PRON
fcis-31421	106	82	makes	make	VERB
fcis-31421	106	83	the	the	DET
fcis-31421	106	84	use	use	NOUN
fcis-31421	106	85	of	of	ADP
fcis-31421	106	86	its	its	PRON
fcis-31421	106	87	this	this	PRON
fcis-31421	106	88	makes	make	VERB
fcis-31421	106	89	it	it	PRON
fcis-31421	106	90	very	very	ADV
fcis-31421	106	91	difficult	difficult	ADJ
fcis-31421	106	92	to	to	PART
fcis-31421	106	93	detect	detect	VERB
fcis-31421	106	94	large	large	ADJ
fcis-31421	106	95	-	-	PUNCT
fcis-31421	106	96	scale	scale	NOUN
fcis-31421	106	97	fires	fire	NOUN
fcis-31421	106	98	,	,	PUNCT
fcis-31421	106	99	and	and	CCONJ
fcis-31421	106	100	the	the	DET
fcis-31421	106	101	maintenance	maintenance	NOUN
fcis-31421	106	102	cost	cost	NOUN
fcis-31421	106	103	is	be	AUX
fcis-31421	106	104	also	also	ADV
fcis-31421	106	105	very	very	ADV
fcis-31421	106	106	high	high	ADJ
fcis-31421	106	107	.	.	PUNCT
fcis-31421	107	1	2.3	2.3	NUM
fcis-31421	107	2	.	.	PUNCT
fcis-31421	108	1	yolo	yolo	ADJ
fcis-31421	108	2	yolo	yolo	ADV
fcis-31421	108	3	(	(	PUNCT
fcis-31421	108	4	you	you	PRON
fcis-31421	108	5	only	only	ADV
fcis-31421	108	6	look	look	VERB
fcis-31421	108	7	once	once	ADV
fcis-31421	108	8	)	)	PUNCT
fcis-31421	108	9	it	it	PRON
fcis-31421	108	10	is	be	AUX
fcis-31421	108	11	a	a	DET
fcis-31421	108	12	single	single	ADJ
fcis-31421	108	13	-	-	PUNCT
fcis-31421	108	14	stage	stage	NOUN
fcis-31421	108	15	target	target	NOUN
fcis-31421	108	16	detection	detection	NOUN
fcis-31421	108	17	algorithm	algorithm	NOUN
fcis-31421	108	18	whose	whose	DET
fcis-31421	108	19	core	core	NOUN
fcis-31421	108	20	idea	idea	NOUN
fcis-31421	108	21	is	be	AUX
fcis-31421	108	22	to	to	PART
fcis-31421	108	23	transform	transform	VERB
fcis-31421	108	24	target	target	NOUN
fcis-31421	108	25	detection	detection	NOUN
fcis-31421	108	26	into	into	ADP
fcis-31421	108	27	a	a	DET
fcis-31421	108	28	regression	regression	NOUN
fcis-31421	108	29	problem	problem	NOUN
fcis-31421	108	30	and	and	CCONJ
fcis-31421	108	31	predict	predict	VERB
fcis-31421	108	32	bounding	bounding	NOUN
fcis-31421	108	33	box	box	NOUN
fcis-31421	108	34	and	and	CCONJ
fcis-31421	108	35	category	category	NOUN
fcis-31421	108	36	probabilities	probability	NOUN
fcis-31421	108	37	directly	directly	ADV
fcis-31421	108	38	on	on	ADP
fcis-31421	108	39	the	the	DET
fcis-31421	108	40	image	image	NOUN
fcis-31421	108	41	.	.	PUNCT
fcis-31421	109	1	compared	compare	VERB
fcis-31421	109	2	to	to	ADP
fcis-31421	109	3	two	two	NUM
fcis-31421	109	4	-	-	PUNCT
fcis-31421	109	5	stage	stage	NOUN
fcis-31421	109	6	models	model	NOUN
fcis-31421	109	7	(	(	PUNCT
fcis-31421	109	8	e.g.	e.g.	ADV
fcis-31421	109	9	,	,	PUNCT
fcis-31421	109	10	faster	fast	ADJ
fcis-31421	109	11	r	r	NOUN
fcis-31421	109	12	-	-	PUNCT
fcis-31421	109	13	cnn	cnn	PROPN
fcis-31421	109	14	)	)	PUNCT
fcis-31421	109	15	,	,	PUNCT
fcis-31421	109	16	yolo	yolo	PROPN
fcis-31421	109	17	is	be	AUX
fcis-31421	109	18	known	know	VERB
fcis-31421	109	19	for	for	ADP
fcis-31421	109	20	its	its	PRON
fcis-31421	109	21	high	high	ADJ
fcis-31421	109	22	speed	speed	NOUN
fcis-31421	109	23	and	and	CCONJ
fcis-31421	109	24	end	end	NOUN
fcis-31421	109	25	-	-	PUNCT
fcis-31421	109	26	to	to	ADP
fcis-31421	109	27	-	-	PUNCT
fcis-31421	109	28	end	end	NOUN
fcis-31421	109	29	optimization	optimization	NOUN
fcis-31421	109	30	,	,	PUNCT
fcis-31421	109	31	which	which	PRON
fcis-31421	109	32	is	be	AUX
fcis-31421	109	33	especially	especially	ADV
fcis-31421	109	34	suitable	suitable	ADJ
fcis-31421	109	35	for	for	ADP
fcis-31421	109	36	fire	fire	NOUN
fcis-31421	109	37	detection	detection	NOUN
fcis-31421	109	38	in	in	ADP
fcis-31421	109	39	scenarios	scenario	NOUN
fcis-31421	109	40	with	with	ADP
fcis-31421	109	41	high	high	ADJ
fcis-31421	109	42	real	real	ADJ
fcis-31421	109	43	-	-	PUNCT
fcis-31421	109	44	time	time	NOUN
fcis-31421	109	45	requirements	requirement	NOUN
fcis-31421	109	46	.	.	PUNCT
fcis-31421	110	1	in	in	ADP
fcis-31421	110	2	recent	recent	ADJ
fcis-31421	110	3	years	year	NOUN
fcis-31421	110	4	,	,	PUNCT
fcis-31421	110	5	the	the	DET
fcis-31421	110	6	iterative	iterative	NOUN
fcis-31421	110	7	upgrades	upgrade	NOUN
fcis-31421	110	8	of	of	ADP
fcis-31421	110	9	yolo	yolo	NOUN
fcis-31421	110	10	have	have	AUX
fcis-31421	110	11	also	also	ADV
fcis-31421	110	12	gradually	gradually	ADV
fcis-31421	110	13	improved	improve	VERB
fcis-31421	110	14	the	the	DET
fcis-31421	110	15	detection	detection	NOUN
fcis-31421	110	16	results	result	NOUN
fcis-31421	110	17	.	.	PUNCT
fcis-31421	111	1	yolov3	yolov3	PROPN
fcis-31421	111	2	/	/	SYM
fcis-31421	111	3	v4	v4	PROPN
fcis-31421	111	4	:	:	PUNCT
fcis-31421	111	5	introduces	introduce	VERB
fcis-31421	111	6	multi	multi	ADJ
fcis-31421	111	7	-	-	ADJ
fcis-31421	111	8	scale	scale	ADJ
fcis-31421	111	9	prediction	prediction	NOUN
fcis-31421	111	10	(	(	PUNCT
fcis-31421	111	11	fpn	fpn	ADJ
fcis-31421	111	12	)	)	PUNCT
fcis-31421	111	13	and	and	CCONJ
fcis-31421	111	14	ciou	ciou	NOUN
fcis-31421	111	15	loss	loss	NOUN
fcis-31421	111	16	function	function	NOUN
fcis-31421	111	17	to	to	PART
fcis-31421	111	18	improve	improve	VERB
fcis-31421	111	19	the	the	DET
fcis-31421	111	20	detection	detection	NOUN
fcis-31421	111	21	of	of	ADP
fcis-31421	111	22	small	small	ADJ
fcis-31421	111	23	targets	target	NOUN
fcis-31421	111	24	in	in	ADP
fcis-31421	111	25	flame	flame	NOUN
fcis-31421	111	26	and	and	CCONJ
fcis-31421	111	27	smoke	smoke	NOUN
fcis-31421	111	28	.	.	PUNCT
fcis-31421	112	1	yolov5	yolov5	NOUN
fcis-31421	112	2	:	:	PUNCT
fcis-31421	112	3	adopts	adopt	VERB
fcis-31421	112	4	pytorch	pytorch	NOUN
fcis-31421	112	5	framework	framework	NOUN
fcis-31421	112	6	,	,	PUNCT
fcis-31421	112	7	supports	support	VERB
fcis-31421	112	8	lightweight	lightweight	ADJ
fcis-31421	112	9	deployment	deployment	NOUN
fcis-31421	112	10	,	,	PUNCT
fcis-31421	112	11	and	and	CCONJ
fcis-31421	112	12	becomes	become	VERB
fcis-31421	112	13	the	the	DET
fcis-31421	112	14	mainstream	mainstream	NOUN
fcis-31421	112	15	choice	choice	NOUN
fcis-31421	112	16	for	for	ADP
fcis-31421	112	17	fire	fire	NOUN
fcis-31421	112	18	mobile	mobile	ADJ
fcis-31421	112	19	detection	detection	NOUN
fcis-31421	112	20	.	.	PUNCT
fcis-31421	113	1	yolov7	yolov7	NOUN
fcis-31421	113	2	/	/	SYM
fcis-31421	113	3	v8	v8	PROPN
fcis-31421	113	4	:	:	PUNCT
fcis-31421	113	5	optimizes	optimize	VERB
fcis-31421	113	6	dynamic	dynamic	ADJ
fcis-31421	113	7	label	label	NOUN
fcis-31421	113	8	assignment	assignment	NOUN
fcis-31421	113	9	and	and	CCONJ
fcis-31421	113	10	re	re	ADJ
fcis-31421	113	11	-	-	ADJ
fcis-31421	113	12	parameterized	parameterized	ADJ
fcis-31421	113	13	convolution	convolution	NOUN
fcis-31421	113	14	to	to	PART
fcis-31421	113	15	reduce	reduce	VERB
fcis-31421	113	16	false	false	ADJ
fcis-31421	113	17	alarm	alarm	NOUN
fcis-31421	113	18	rate	rate	NOUN
fcis-31421	113	19	in	in	ADP
fcis-31421	113	20	complex	complex	ADJ
fcis-31421	113	21	backgrounds	background	NOUN
fcis-31421	113	22	.	.	PUNCT
fcis-31421	114	1	because	because	SCONJ
fcis-31421	114	2	of	of	ADP
fcis-31421	114	3	the	the	DET
fcis-31421	114	4	excellent	excellent	ADJ
fcis-31421	114	5	performance	performance	NOUN
fcis-31421	114	6	of	of	ADP
fcis-31421	114	7	yolo	yolo	NOUN
fcis-31421	114	8	in	in	ADP
fcis-31421	114	9	realtime	realtime	ADJ
fcis-31421	114	10	detection	detection	NOUN
fcis-31421	114	11	,	,	PUNCT
fcis-31421	114	12	a	a	DET
fcis-31421	114	13	very	very	ADV
fcis-31421	114	14	large	large	ADJ
fcis-31421	114	15	number	number	NOUN
fcis-31421	114	16	of	of	ADP
fcis-31421	114	17	scholars	scholar	NOUN
fcis-31421	114	18	have	have	AUX
fcis-31421	114	19	conducted	conduct	VERB
fcis-31421	114	20	research	research	NOUN
fcis-31421	114	21	in	in	ADP
fcis-31421	114	22	this	this	DET
fcis-31421	114	23	area	area	NOUN
fcis-31421	114	24	in	in	ADP
fcis-31421	114	25	recent	recent	ADJ
fcis-31421	114	26	years	year	NOUN
fcis-31421	114	27	.	.	PUNCT
fcis-31421	115	1	more	more	ADV
fcis-31421	115	2	classically	classically	ADV
fcis-31421	115	3	,	,	PUNCT
fcis-31421	115	4	talaat	talaat	PROPN
fcis-31421	115	5	f	f	AUX
fcis-31421	115	6	m	m	VERB
fcis-31421	115	7	et	et	PROPN
fcis-31421	115	8	al	al	PROPN
fcis-31421	115	9	.	.	PROPN
fcis-31421	115	10	used	use	VERB
fcis-31421	115	11	the	the	DET
fcis-31421	115	12	smart	smart	ADJ
fcis-31421	115	13	fire	fire	NOUN
fcis-31421	115	14	detection	detection	NOUN
fcis-31421	115	15	(	(	PUNCT
fcis-31421	115	16	sfd	sfd	NOUN
fcis-31421	115	17	)	)	PUNCT
fcis-31421	115	18	algorithm	algorithm	NOUN
fcis-31421	115	19	to	to	PART
fcis-31421	115	20	detect	detect	VERB
fcis-31421	115	21	fires	fire	NOUN
fcis-31421	115	22	in	in	ADP
fcis-31421	115	23	real	real	ADJ
fcis-31421	115	24	time	time	NOUN
fcis-31421	115	25	using	use	VERB
fcis-31421	115	26	computer	computer	NOUN
fcis-31421	115	27	vision	vision	NOUN
fcis-31421	115	28	from	from	ADP
fcis-31421	115	29	live	live	ADJ
fcis-31421	115	30	cameras	camera	NOUN
fcis-31421	115	31	or	or	CCONJ
fcis-31421	115	32	pre	pre	ADJ
fcis-31421	115	33	-	-	ADJ
fcis-31421	115	34	recorded	recorded	ADJ
fcis-31421	115	35	video	video	NOUN
fcis-31421	115	36	files	file	NOUN
fcis-31421	115	37	in	in	ADP
fcis-31421	115	38	real	real	ADJ
fcis-31421	115	39	[	[	X
fcis-31421	115	40	9	9	NUM
fcis-31421	115	41	]	]	PUNCT
fcis-31421	115	42	.	.	PUNCT
fcis-31421	116	1	the	the	DET
fcis-31421	116	2	sfd	sfd	PROPN
fcis-31421	116	3	algorithm	algorithm	PROPN
fcis-31421	116	4	is	be	AUX
fcis-31421	116	5	an	an	DET
fcis-31421	116	6	efficient	efficient	ADJ
fcis-31421	116	7	real	real	ADJ
fcis-31421	116	8	-	-	PUNCT
fcis-31421	116	9	time	time	NOUN
fcis-31421	116	10	fire	fire	NOUN
fcis-31421	116	11	detection	detection	NOUN
fcis-31421	116	12	solution	solution	NOUN
fcis-31421	116	13	that	that	PRON
fcis-31421	116	14	can	can	AUX
fcis-31421	116	15	quickly	quickly	ADV
fcis-31421	116	16	respond	respond	VERB
fcis-31421	116	17	to	to	ADP
fcis-31421	116	18	potential	potential	ADJ
fcis-31421	116	19	fire	fire	NOUN
fcis-31421	116	20	risks	risk	NOUN
fcis-31421	116	21	.	.	PUNCT
fcis-31421	117	1	this	this	DET
fcis-31421	117	2	method	method	NOUN
fcis-31421	117	3	adopts	adopt	VERB
fcis-31421	117	4	the	the	DET
fcis-31421	117	5	yolov8	yolov8	NOUN
fcis-31421	117	6	architecture	architecture	NOUN
fcis-31421	117	7	and	and	CCONJ
fcis-31421	117	8	can	can	AUX
fcis-31421	117	9	achieve	achieve	VERB
fcis-31421	117	10	fast	fast	ADJ
fcis-31421	117	11	and	and	CCONJ
fcis-31421	117	12	accurate	accurate	ADJ
fcis-31421	117	13	target	target	NOUN
fcis-31421	117	14	recognition	recognition	NOUN
fcis-31421	117	15	without	without	ADP
fcis-31421	117	16	relying	rely	VERB
fcis-31421	117	17	on	on	ADP
fcis-31421	117	18	regional	regional	ADJ
fcis-31421	117	19	suggestion	suggestion	NOUN
fcis-31421	117	20	networks	network	NOUN
fcis-31421	117	21	.	.	PUNCT
fcis-31421	118	1	through	through	ADP
fcis-31421	118	2	system	system	NOUN
fcis-31421	118	3	optimization	optimization	NOUN
fcis-31421	118	4	,	,	PUNCT
fcis-31421	118	5	this	this	DET
fcis-31421	118	6	solution	solution	NOUN
fcis-31421	118	7	significantly	significantly	ADV
fcis-31421	118	8	reduces	reduce	VERB
fcis-31421	118	9	the	the	DET
fcis-31421	118	10	number	number	NOUN
fcis-31421	118	11	of	of	ADP
fcis-31421	118	12	parameters	parameter	NOUN
fcis-31421	118	13	while	while	SCONJ
fcis-31421	118	14	maintaining	maintain	VERB
fcis-31421	118	15	its	its	PRON
fcis-31421	118	16	lightweight	lightweight	ADJ
fcis-31421	118	17	characteristics	characteristic	NOUN
fcis-31421	118	18	,	,	PUNCT
fcis-31421	118	19	making	make	VERB
fcis-31421	118	20	it	it	PRON
fcis-31421	118	21	particularly	particularly	ADV
fcis-31421	118	22	suitable	suitable	ADJ
fcis-31421	118	23	for	for	ADP
fcis-31421	118	24	deployment	deployment	NOUN
fcis-31421	118	25	in	in	ADP
fcis-31421	118	26	resource	resource	NOUN
fcis-31421	118	27	constrained	constrain	VERB
fcis-31421	118	28	environments	environment	NOUN
fcis-31421	118	29	.	.	PUNCT
fcis-31421	119	1	this	this	DET
fcis-31421	119	2	technology	technology	NOUN
fcis-31421	119	3	achieves	achieve	VERB
fcis-31421	119	4	an	an	DET
fcis-31421	119	5	excellent	excellent	ADJ
fcis-31421	119	6	balance	balance	NOUN
fcis-31421	119	7	between	between	ADP
fcis-31421	119	8	detection	detection	NOUN
fcis-31421	119	9	accuracy	accuracy	NOUN
fcis-31421	119	10	and	and	CCONJ
fcis-31421	119	11	computational	computational	ADJ
fcis-31421	119	12	efficiency	efficiency	NOUN
fcis-31421	119	13	.	.	PUNCT
fcis-31421	120	1	meanwhile	meanwhile	ADV
fcis-31421	120	2	,	,	PUNCT
fcis-31421	120	3	the	the	DET
fcis-31421	120	4	model	model	NOUN
fcis-31421	120	5	detects	detect	VERB
fcis-31421	120	6	fire	fire	NOUN
fcis-31421	120	7	and	and	CCONJ
fcis-31421	120	8	smoke	smoke	VERB
fcis-31421	120	9	with	with	ADP
fcis-31421	120	10	95.7	95.7	NUM
fcis-31421	120	11	%	%	NOUN
fcis-31421	120	12	and	and	CCONJ
fcis-31421	120	13	99.3	99.3	NUM
fcis-31421	120	14	%	%	NOUN
fcis-31421	120	15	accuracy	accuracy	NOUN
fcis-31421	120	16	,	,	PUNCT
fcis-31421	120	17	respectively	respectively	ADV
fcis-31421	120	18	,	,	PUNCT
fcis-31421	120	19	and	and	CCONJ
fcis-31421	120	20	the	the	DET
fcis-31421	120	21	mean	mean	ADJ
fcis-31421	120	22	accuracy	accuracy	NOUN
fcis-31421	120	23	of	of	ADP
fcis-31421	120	24	detection	detection	NOUN
fcis-31421	120	25	is	be	AUX
fcis-31421	120	26	97.5	97.5	NUM
fcis-31421	120	27	%	%	NOUN
fcis-31421	120	28	and	and	CCONJ
fcis-31421	120	29	97.5	97.5	NUM
fcis-31421	120	30	%	%	NOUN
fcis-31421	120	31	for	for	ADP
fcis-31421	120	32	both	both	DET
fcis-31421	120	33	categories	category	NOUN
fcis-31421	120	34	,	,	PUNCT
fcis-31421	120	35	with	with	ADP
fcis-31421	120	36	a	a	DET
fcis-31421	120	37	high	high	ADJ
fcis-31421	120	38	accuracy	accuracy	NOUN
fcis-31421	120	39	of	of	ADP
fcis-31421	120	40	97.1	97.1	NUM
fcis-31421	120	41	%	%	NOUN
fcis-31421	120	42	for	for	ADP
fcis-31421	120	43	all	all	DET
fcis-31421	120	44	types	type	NOUN
fcis-31421	120	45	of	of	ADP
fcis-31421	120	46	fires	fire	NOUN
fcis-31421	120	47	,	,	PUNCT
fcis-31421	120	48	which	which	PRON
fcis-31421	120	49	is	be	AUX
fcis-31421	120	50	a	a	DET
fcis-31421	120	51	very	very	ADV
fcis-31421	120	52	high	high	ADJ
fcis-31421	120	53	level	level	NOUN
fcis-31421	120	54	of	of	ADP
fcis-31421	120	55	accuracy	accuracy	NOUN
fcis-31421	120	56	.	.	PUNCT
fcis-31421	121	1	in	in	ADP
fcis-31421	121	2	their	their	PRON
fcis-31421	121	3	study	study	NOUN
fcis-31421	121	4	,	,	PUNCT
fcis-31421	121	5	goyal	goyal	PROPN
fcis-31421	121	6	s	s	PROPN
fcis-31421	121	7	et	et	PROPN
fcis-31421	121	8	al	al	PROPN
fcis-31421	121	9	.	.	PROPN
fcis-31421	121	10	deployed	deploy	VERB
fcis-31421	121	11	deep	deep	ADJ
fcis-31421	121	12	learning	learning	NOUN
fcis-31421	121	13	models	model	NOUN
fcis-31421	121	14	on	on	ADP
fcis-31421	121	15	a	a	DET
fcis-31421	121	16	raspberry	raspberry	NOUN
fcis-31421	121	17	-	-	PUNCT
fcis-31421	121	18	pi	pi	NOUN
fcis-31421	121	19	with	with	ADP
fcis-31421	121	20	neural	neural	ADJ
fcis-31421	121	21	rods	rod	NOUN
fcis-31421	121	22	on	on	ADP
fcis-31421	121	23	a	a	DET
fcis-31421	121	24	drone	drone	NOUN
fcis-31421	121	25	to	to	PART
fcis-31421	121	26	help	help	VERB
fcis-31421	121	27	detect	detect	VERB
fcis-31421	121	28	fires	fire	NOUN
fcis-31421	121	29	.	.	PUNCT
fcis-31421	122	1	they	they	PRON
fcis-31421	122	2	embedded	embed	VERB
fcis-31421	122	3	the	the	DET
fcis-31421	122	4	yolov3	yolov3	PROPN
fcis-31421	122	5	model	model	NOUN
fcis-31421	122	6	into	into	ADP
fcis-31421	122	7	a	a	DET
fcis-31421	122	8	neural	neural	ADJ
fcis-31421	122	9	bar	bar	NOUN
fcis-31421	122	10	,	,	PUNCT
fcis-31421	122	11	and	and	CCONJ
fcis-31421	122	12	once	once	ADV
fcis-31421	122	13	an	an	DET
fcis-31421	122	14	image	image	NOUN
fcis-31421	122	15	is	be	AUX
fcis-31421	122	16	captured	capture	VERB
fcis-31421	122	17	,	,	PUNCT
fcis-31421	122	18	it	it	PRON
fcis-31421	122	19	is	be	AUX
fcis-31421	122	20	processed	process	VERB
fcis-31421	122	21	and	and	CCONJ
fcis-31421	122	22	passed	pass	VERB
fcis-31421	122	23	to	to	ADP
fcis-31421	122	24	a	a	DET
fcis-31421	122	25	cnn	cnn	PROPN
fcis-31421	122	26	,	,	PUNCT
fcis-31421	122	27	which	which	PRON
fcis-31421	122	28	will	will	AUX
fcis-31421	122	29	check	check	VERB
fcis-31421	122	30	for	for	ADP
fcis-31421	122	31	forest	forest	NOUN
fcis-31421	122	32	fires	fire	NOUN
fcis-31421	122	33	.	.	PUNCT
fcis-31421	123	1	and	and	CCONJ
fcis-31421	123	2	if	if	SCONJ
fcis-31421	123	3	it	it	PRON
fcis-31421	123	4	detects	detect	VERB
fcis-31421	123	5	that	that	SCONJ
fcis-31421	123	6	the	the	DET
fcis-31421	123	7	confidence	confidence	NOUN
fcis-31421	123	8	level	level	NOUN
fcis-31421	123	9	is	be	AUX
fcis-31421	123	10	exceeded	exceed	VERB
fcis-31421	123	11	,	,	PUNCT
fcis-31421	123	12	then	then	ADV
fcis-31421	123	13	it	it	PRON
fcis-31421	123	14	will	will	AUX
fcis-31421	123	15	send	send	VERB
fcis-31421	123	16	a	a	DET
fcis-31421	123	17	notification	notification	NOUN
fcis-31421	123	18	[	[	X
fcis-31421	123	19	10	10	NUM
fcis-31421	123	20	]	]	PUNCT
fcis-31421	123	21	.	.	PUNCT
fcis-31421	124	1	goyal	goyal	PROPN
fcis-31421	124	2	s	s	PART
fcis-31421	124	3	et	et	PROPN
fcis-31421	124	4	al	al	PROPN
fcis-31421	124	5	.	.	PROPN
fcis-31421	124	6	fused	fuse	VERB
fcis-31421	124	7	the	the	DET
fcis-31421	124	8	yolo	yolo	ADJ
fcis-31421	124	9	model	model	NOUN
fcis-31421	124	10	with	with	ADP
fcis-31421	124	11	cnn	cnn	PROPN
fcis-31421	124	12	and	and	CCONJ
fcis-31421	124	13	achieved	achieve	VERB
fcis-31421	124	14	better	well	ADJ
fcis-31421	124	15	experimental	experimental	ADJ
fcis-31421	124	16	results	result	NOUN
fcis-31421	124	17	with	with	ADP
fcis-31421	124	18	90	90	NUM
fcis-31421	124	19	%	%	NOUN
fcis-31421	124	20	accuracy	accuracy	NOUN
fcis-31421	124	21	of	of	ADP
fcis-31421	124	22	yolo	yolo	PROPN
fcis-31421	124	23	based	base	VERB
fcis-31421	124	24	forest	forest	NOUN
fcis-31421	124	25	fire	fire	NOUN
fcis-31421	124	26	detection	detection	NOUN
fcis-31421	124	27	system	system	NOUN
fcis-31421	124	28	.	.	PUNCT
fcis-31421	125	1	its	its	PRON
fcis-31421	125	2	sensitivity	sensitivity	NOUN
fcis-31421	125	3	and	and	CCONJ
fcis-31421	125	4	specificity	specificity	NOUN
fcis-31421	125	5	are	be	AUX
fcis-31421	125	6	92	92	NUM
fcis-31421	125	7	%	%	NOUN
fcis-31421	125	8	and	and	CCONJ
fcis-31421	125	9	90	90	NUM
fcis-31421	125	10	%	%	NOUN
fcis-31421	125	11	respectively	respectively	ADV
fcis-31421	125	12	.	.	PUNCT
fcis-31421	126	1	this	this	PRON
fcis-31421	126	2	also	also	ADV
fcis-31421	126	3	reflects	reflect	VERB
fcis-31421	126	4	the	the	DET
fcis-31421	126	5	powerful	powerful	ADJ
fcis-31421	126	6	plasticity	plasticity	NOUN
fcis-31421	126	7	and	and	CCONJ
fcis-31421	126	8	compatibility	compatibility	NOUN
fcis-31421	126	9	of	of	ADP
fcis-31421	126	10	yolo	yolo	ADJ
fcis-31421	126	11	model	model	NOUN
fcis-31421	126	12	.	.	PUNCT
fcis-31421	127	1	wu	wu	PROPN
fcis-31421	127	2	h	h	PROPN
fcis-31421	127	3	et	et	PROPN
fcis-31421	127	4	al	al	PROPN
fcis-31421	127	5	.	.	PROPN
fcis-31421	127	6	proposed	propose	VERB
fcis-31421	127	7	a	a	DET
fcis-31421	127	8	ship	ship	NOUN
fcis-31421	127	9	fire	fire	NOUN
fcis-31421	127	10	detection	detection	NOUN
fcis-31421	127	11	model	model	NOUN
fcis-31421	127	12	based	base	VERB
fcis-31421	127	13	on	on	ADP
fcis-31421	127	14	enhanced	enhance	VERB
fcis-31421	127	15	yolov	yolov	ADJ
fcis-31421	127	16	4	4	NUM
fcis-31421	127	17	-	-	PUNCT
fcis-31421	127	18	tiny	tiny	ADJ
fcis-31421	127	19	network	network	NOUN
fcis-31421	127	20	to	to	PART
fcis-31421	127	21	address	address	VERB
fcis-31421	127	22	the	the	DET
fcis-31421	127	23	shortcomings	shortcoming	NOUN
fcis-31421	127	24	of	of	ADP
fcis-31421	127	25	sensor	sensor	NOUN
fcis-31421	127	26	detection	detection	NOUN
fcis-31421	127	27	and	and	CCONJ
fcis-31421	127	28	detection	detection	NOUN
fcis-31421	127	29	methods	method	NOUN
fcis-31421	127	30	of	of	ADP
fcis-31421	127	31	traditional	traditional	ADJ
fcis-31421	127	32	image	image	NOUN
fcis-31421	127	33	processing	processing	NOUN
fcis-31421	127	34	techniques	technique	NOUN
fcis-31421	127	35	,	,	PUNCT
fcis-31421	127	36	adding	add	VERB
fcis-31421	127	37	the	the	DET
fcis-31421	127	38	se	se	PROPN
fcis-31421	127	39	attention	attention	NOUN
fcis-31421	127	40	mechanism	mechanism	NOUN
fcis-31421	127	41	to	to	ADP
fcis-31421	127	42	the	the	DET
fcis-31421	127	43	enhanced	enhance	VERB
fcis-31421	127	44	feature	feature	NOUN
fcis-31421	127	45	extraction	extraction	NOUN
fcis-31421	127	46	network	network	NOUN
fcis-31421	127	47	partially	partially	ADV
fcis-31421	127	48	generating	generate	VERB
fcis-31421	127	49	the	the	DET
fcis-31421	127	50	i	i	NOUN
fcis-31421	127	51	-	-	PUNCT
fcis-31421	127	52	yolov	yolov	ADJ
fcis-31421	127	53	4	4	NUM
fcis-31421	127	54	-	-	PUNCT
fcis-31421	127	55	tiny	tiny	ADJ
fcis-31421	127	56	+	+	CCONJ
fcis-31421	127	57	senet	senet	NOUN
fcis-31421	127	58	model	model	NOUN
fcis-31421	127	59	[	[	X
fcis-31421	127	60	11	11	NUM
fcis-31421	127	61	]	]	PUNCT
fcis-31421	127	62	.	.	PUNCT
fcis-31421	128	1	the	the	DET
fcis-31421	128	2	addition	addition	NOUN
fcis-31421	128	3	of	of	ADP
fcis-31421	128	4	the	the	DET
fcis-31421	128	5	se	se	PROPN
fcis-31421	128	6	attention	attention	NOUN
fcis-31421	128	7	mechanism	mechanism	NOUN
fcis-31421	128	8	reduces	reduce	VERB
fcis-31421	128	9	the	the	DET
fcis-31421	128	10	dependence	dependence	NOUN
fcis-31421	128	11	on	on	ADP
fcis-31421	128	12	the	the	DET
fcis-31421	128	13	ship	ship	NOUN
fcis-31421	128	14	fire	fire	NOUN
fcis-31421	128	15	dataset	dataset	NOUN
fcis-31421	128	16	,	,	PUNCT
fcis-31421	128	17	speeds	speed	VERB
fcis-31421	128	18	up	up	ADP
fcis-31421	128	19	convergence	convergence	NOUN
fcis-31421	128	20	,	,	PUNCT
fcis-31421	128	21	and	and	CCONJ
fcis-31421	128	22	achieves	achieve	VERB
fcis-31421	128	23	the	the	DET
fcis-31421	128	24	goal	goal	NOUN
fcis-31421	128	25	of	of	ADP
fcis-31421	128	26	accurate	accurate	ADJ
fcis-31421	128	27	ship	ship	NOUN
fcis-31421	128	28	fire	fire	NOUN
fcis-31421	128	29	detection	detection	NOUN
fcis-31421	128	30	.	.	PUNCT
fcis-31421	129	1	the	the	DET
fcis-31421	129	2	experimental	experimental	ADJ
fcis-31421	129	3	results	result	NOUN
fcis-31421	129	4	show	show	VERB
fcis-31421	129	5	that	that	SCONJ
fcis-31421	129	6	the	the	DET
fcis-31421	129	7	deep	deep	ADJ
fcis-31421	129	8	network	network	NOUN
fcis-31421	129	9	learning	learning	NOUN
fcis-31421	129	10	model	model	NOUN
fcis-31421	129	11	iyolov4	iyolov4	PROPN
fcis-31421	129	12	tiny+se	tiny+se	ADV
fcis-31421	129	13	55	55	NUM
fcis-31421	129	14	significantly	significantly	ADV
fcis-31421	129	15	outperforms	outperform	VERB
fcis-31421	129	16	other	other	ADJ
fcis-31421	129	17	compared	compare	VERB
fcis-31421	129	18	models	model	NOUN
fcis-31421	129	19	in	in	ADP
fcis-31421	129	20	terms	term	NOUN
fcis-31421	129	21	of	of	ADP
fcis-31421	129	22	performance	performance	NOUN
fcis-31421	129	23	.	.	PUNCT
fcis-31421	130	1	compared	compare	VERB
fcis-31421	130	2	to	to	ADP
fcis-31421	130	3	yolov3	yolov3	PROPN
fcis-31421	130	4	tiny	tiny	ADJ
fcis-31421	130	5	,	,	PUNCT
fcis-31421	130	6	ssd	ssd	PROPN
fcis-31421	130	7	,	,	PUNCT
fcis-31421	130	8	yolov4	yolov4	PROPN
fcis-31421	130	9	tiny	tiny	ADJ
fcis-31421	130	10	,	,	PUNCT
fcis-31421	130	11	and	and	CCONJ
fcis-31421	130	12	i	i	PROPN
fcis-31421	130	13	-	-	PUNCT
fcis-31421	130	14	yolov4	yolov4	PROPN
fcis-31421	130	15	tiny	tiny	ADJ
fcis-31421	130	16	,	,	PUNCT
fcis-31421	130	17	this	this	DET
fcis-31421	130	18	model	model	NOUN
fcis-31421	130	19	has	have	AUX
fcis-31421	130	20	map@.5	map@.5	VERB
fcis-31421	130	21	the	the	DET
fcis-31421	130	22	indicators	indicator	NOUN
fcis-31421	130	23	have	have	AUX
fcis-31421	130	24	increased	increase	VERB
fcis-31421	130	25	by	by	ADP
fcis-31421	130	26	19.5	19.5	NUM
fcis-31421	130	27	%	%	NOUN
fcis-31421	130	28	,	,	PUNCT
fcis-31421	130	29	10.9	10.9	NUM
fcis-31421	130	30	%	%	NOUN
fcis-31421	130	31	,	,	PUNCT
fcis-31421	130	32	8.5	8.5	NUM
fcis-31421	130	33	%	%	NOUN
fcis-31421	130	34	,	,	PUNCT
fcis-31421	130	35	and	and	CCONJ
fcis-31421	130	36	2.1	2.1	NUM
fcis-31421	130	37	%	%	NOUN
fcis-31421	130	38	respectively	respectively	ADV
fcis-31421	130	39	.	.	PUNCT
fcis-31421	131	1	in	in	ADP
fcis-31421	131	2	terms	term	NOUN
fcis-31421	131	3	of	of	ADP
fcis-31421	131	4	accuracy	accuracy	NOUN
fcis-31421	131	5	,	,	PUNCT
fcis-31421	131	6	its	its	PRON
fcis-31421	131	7	advantages	advantage	NOUN
fcis-31421	131	8	reach	reach	VERB
fcis-31421	131	9	16.3	16.3	NUM
fcis-31421	131	10	%	%	NOUN
fcis-31421	131	11	,	,	PUNCT
fcis-31421	131	12	10.2	10.2	NUM
fcis-31421	131	13	%	%	NOUN
fcis-31421	131	14	,	,	PUNCT
fcis-31421	131	15	7.7	7.7	NUM
fcis-31421	131	16	%	%	NOUN
fcis-31421	131	17	,	,	PUNCT
fcis-31421	131	18	and	and	CCONJ
fcis-31421	131	19	1.9	1.9	NUM
fcis-31421	131	20	%	%	NOUN
fcis-31421	131	21	,	,	PUNCT
fcis-31421	131	22	while	while	SCONJ
fcis-31421	131	23	the	the	DET
fcis-31421	131	24	improvement	improvement	NOUN
fcis-31421	131	25	in	in	ADP
fcis-31421	131	26	recall	recall	NOUN
fcis-31421	131	27	rate	rate	NOUN
fcis-31421	131	28	is	be	AUX
fcis-31421	131	29	more	more	ADV
fcis-31421	131	30	significant	significant	ADJ
fcis-31421	131	31	,	,	PUNCT
fcis-31421	131	32	reaching	reach	VERB
fcis-31421	131	33	22.1	22.1	NUM
fcis-31421	131	34	%	%	NOUN
fcis-31421	131	35	,	,	PUNCT
fcis-31421	131	36	18.1	18.1	NUM
fcis-31421	131	37	%	%	NOUN
fcis-31421	131	38	,	,	PUNCT
fcis-31421	131	39	9.2	9.2	NUM
fcis-31421	131	40	%	%	NOUN
fcis-31421	131	41	,	,	PUNCT
fcis-31421	131	42	and	and	CCONJ
fcis-31421	131	43	3.9	3.9	NUM
fcis-31421	131	44	%	%	NOUN
fcis-31421	131	45	,	,	PUNCT
fcis-31421	131	46	respectively	respectively	ADV
fcis-31421	131	47	.	.	PUNCT
fcis-31421	132	1	although	although	SCONJ
fcis-31421	132	2	these	these	DET
fcis-31421	132	3	improvements	improvement	NOUN
fcis-31421	132	4	have	have	AUX
fcis-31421	132	5	brought	bring	VERB
fcis-31421	132	6	significant	significant	ADJ
fcis-31421	132	7	improvements	improvement	NOUN
fcis-31421	132	8	in	in	ADP
fcis-31421	132	9	accuracy	accuracy	NOUN
fcis-31421	132	10	,	,	PUNCT
fcis-31421	132	11	the	the	DET
fcis-31421	132	12	model	model	NOUN
fcis-31421	132	13	still	still	ADV
fcis-31421	132	14	has	have	VERB
fcis-31421	132	15	several	several	ADJ
fcis-31421	132	16	limitations	limitation	NOUN
fcis-31421	132	17	.	.	PUNCT
fcis-31421	133	1	firstly	firstly	ADV
fcis-31421	133	2	,	,	PUNCT
fcis-31421	133	3	its	its	PRON
fcis-31421	133	4	training	training	NOUN
fcis-31421	133	5	process	process	NOUN
fcis-31421	133	6	relies	rely	VERB
fcis-31421	133	7	on	on	ADP
fcis-31421	133	8	a	a	DET
fcis-31421	133	9	large	large	ADJ
fcis-31421	133	10	amount	amount	NOUN
fcis-31421	133	11	of	of	ADP
fcis-31421	133	12	sample	sample	NOUN
fcis-31421	133	13	data	datum	NOUN
fcis-31421	133	14	,	,	PUNCT
fcis-31421	133	15	and	and	CCONJ
fcis-31421	133	16	insufficient	insufficient	ADJ
fcis-31421	133	17	samples	sample	NOUN
fcis-31421	133	18	can	can	AUX
fcis-31421	133	19	lead	lead	VERB
fcis-31421	133	20	to	to	ADP
fcis-31421	133	21	a	a	DET
fcis-31421	133	22	decrease	decrease	NOUN
fcis-31421	133	23	in	in	ADP
fcis-31421	133	24	recognition	recognition	NOUN
fcis-31421	133	25	performance	performance	NOUN
fcis-31421	133	26	.	.	PUNCT
fcis-31421	134	1	secondly	secondly	ADV
fcis-31421	134	2	,	,	PUNCT
fcis-31421	134	3	training	train	VERB
fcis-31421	134	4	deep	deep	ADJ
fcis-31421	134	5	convolutional	convolutional	ADJ
fcis-31421	134	6	neural	neural	ADJ
fcis-31421	134	7	networks	network	NOUN
fcis-31421	134	8	from	from	ADP
fcis-31421	134	9	scratch	scratch	NOUN
fcis-31421	134	10	takes	take	VERB
fcis-31421	134	11	a	a	DET
fcis-31421	134	12	long	long	ADJ
fcis-31421	134	13	time	time	NOUN
fcis-31421	134	14	and	and	CCONJ
fcis-31421	134	15	requires	require	VERB
fcis-31421	134	16	high	high	ADJ
fcis-31421	134	17	computational	computational	ADJ
fcis-31421	134	18	resources	resource	NOUN
fcis-31421	134	19	.	.	PUNCT
fcis-31421	135	1	in	in	ADP
fcis-31421	135	2	addition	addition	NOUN
fcis-31421	135	3	,	,	PUNCT
fcis-31421	135	4	high	high	ADJ
fcis-31421	135	5	-	-	PUNCT
fcis-31421	135	6	performance	performance	NOUN
fcis-31421	135	7	hardware	hardware	NOUN
fcis-31421	135	8	platforms	platform	NOUN
fcis-31421	135	9	are	be	AUX
fcis-31421	135	10	expensive	expensive	ADJ
fcis-31421	135	11	and	and	CCONJ
fcis-31421	135	12	require	require	VERB
fcis-31421	135	13	strong	strong	ADJ
fcis-31421	135	14	computing	computing	NOUN
fcis-31421	135	15	power	power	NOUN
fcis-31421	135	16	and	and	CCONJ
fcis-31421	135	17	storage	storage	NOUN
fcis-31421	135	18	capacity	capacity	NOUN
fcis-31421	135	19	support	support	NOUN
fcis-31421	135	20	,	,	PUNCT
fcis-31421	135	21	which	which	PRON
fcis-31421	135	22	poses	pose	VERB
fcis-31421	135	23	challenges	challenge	NOUN
fcis-31421	135	24	for	for	ADP
fcis-31421	135	25	the	the	DET
fcis-31421	135	26	practical	practical	ADJ
fcis-31421	135	27	application	application	NOUN
fcis-31421	135	28	of	of	ADP
fcis-31421	135	29	this	this	DET
fcis-31421	135	30	model	model	NOUN
fcis-31421	135	31	in	in	ADP
fcis-31421	135	32	large	large	ADJ
fcis-31421	135	33	-	-	PUNCT
fcis-31421	135	34	scale	scale	NOUN
fcis-31421	135	35	detection	detection	NOUN
fcis-31421	135	36	tasks	task	NOUN
fcis-31421	135	37	.	.	PUNCT
fcis-31421	136	1	the	the	DET
fcis-31421	136	2	cao	cao	PROPN
fcis-31421	136	3	l	l	PROPN
fcis-31421	136	4	research	research	NOUN
fcis-31421	136	5	team	team	NOUN
fcis-31421	136	6	has	have	AUX
fcis-31421	136	7	made	make	VERB
fcis-31421	136	8	multiple	multiple	ADJ
fcis-31421	136	9	innovative	innovative	ADJ
fcis-31421	136	10	improvements	improvement	NOUN
fcis-31421	136	11	to	to	ADP
fcis-31421	136	12	yolov5	yolov5	NOUN
fcis-31421	136	13	architecture	architecture	NOUN
fcis-31421	136	14	.	.	PUNCT
fcis-31421	137	1	firstly	firstly	ADV
fcis-31421	137	2	,	,	PUNCT
fcis-31421	137	3	they	they	PRON
fcis-31421	137	4	integrated	integrate	VERB
fcis-31421	137	5	the	the	DET
fcis-31421	137	6	c3tr	c3tr	ADJ
fcis-31421	137	7	module	module	NOUN
fcis-31421	137	8	and	and	CCONJ
fcis-31421	137	9	global	global	ADJ
fcis-31421	137	10	attention	attention	NOUN
fcis-31421	137	11	mechanism	mechanism	NOUN
fcis-31421	137	12	into	into	ADP
fcis-31421	137	13	the	the	DET
fcis-31421	137	14	neck	neck	NOUN
fcis-31421	137	15	trunk	trunk	NOUN
fcis-31421	137	16	network	network	NOUN
fcis-31421	137	17	,	,	PUNCT
fcis-31421	137	18	significantly	significantly	ADV
fcis-31421	137	19	improving	improve	VERB
fcis-31421	137	20	the	the	DET
fcis-31421	137	21	visual	visual	ADJ
fcis-31421	137	22	feature	feature	NOUN
fcis-31421	137	23	extraction	extraction	NOUN
fcis-31421	137	24	ability	ability	NOUN
fcis-31421	137	25	of	of	ADP
fcis-31421	137	26	the	the	DET
fcis-31421	137	27	network	network	NOUN
fcis-31421	137	28	receptive	receptive	ADJ
fcis-31421	137	29	field	field	NOUN
fcis-31421	137	30	.	.	PUNCT
fcis-31421	138	1	secondly	secondly	ADV
fcis-31421	138	2	,	,	PUNCT
fcis-31421	138	3	by	by	ADP
fcis-31421	138	4	optimizing	optimize	VERB
fcis-31421	138	5	the	the	DET
fcis-31421	138	6	reparameterized	reparameterize	VERB
fcis-31421	138	7	convolution	convolution	NOUN
fcis-31421	138	8	module	module	NOUN
fcis-31421	138	9	and	and	CCONJ
fcis-31421	138	10	reconstructing	reconstruct	VERB
fcis-31421	138	11	the	the	DET
fcis-31421	138	12	decoupled	decoupled	ADJ
fcis-31421	138	13	head	head	NOUN
fcis-31421	138	14	and	and	CCONJ
fcis-31421	138	15	neck	neck	NOUN
fcis-31421	138	16	network	network	NOUN
fcis-31421	138	17	structure	structure	NOUN
fcis-31421	138	18	,	,	PUNCT
fcis-31421	138	19	the	the	DET
fcis-31421	138	20	convergence	convergence	NOUN
fcis-31421	138	21	speed	speed	NOUN
fcis-31421	138	22	of	of	ADP
fcis-31421	138	23	model	model	NOUN
fcis-31421	138	24	training	training	NOUN
fcis-31421	138	25	has	have	AUX
fcis-31421	138	26	been	be	AUX
fcis-31421	138	27	effectively	effectively	ADV
fcis-31421	138	28	accelerated	accelerate	VERB
fcis-31421	138	29	.	.	PUNCT
fcis-31421	139	1	in	in	ADP
fcis-31421	139	2	terms	term	NOUN
fcis-31421	139	3	of	of	ADP
fcis-31421	139	4	loss	loss	NOUN
fcis-31421	139	5	function	function	NOUN
fcis-31421	139	6	,	,	PUNCT
fcis-31421	139	7	fully	fully	ADV
fcis-31421	139	8	concurrent	concurrent	ADJ
fcis-31421	139	9	intersection	intersection	NOUN
fcis-31421	139	10	is	be	AUX
fcis-31421	139	11	used	use	VERB
fcis-31421	139	12	as	as	ADP
fcis-31421	139	13	the	the	DET
fcis-31421	139	14	bounding	bounding	NOUN
fcis-31421	139	15	box	box	NOUN
fcis-31421	139	16	regression	regression	NOUN
fcis-31421	139	17	loss	loss	NOUN
fcis-31421	139	18	to	to	PART
fcis-31421	139	19	achieve	achieve	VERB
fcis-31421	139	20	optimized	optimize	VERB
fcis-31421	139	21	configuration	configuration	NOUN
fcis-31421	139	22	of	of	ADP
fcis-31421	139	23	multitask	multitask	ADJ
fcis-31421	139	24	loss	loss	NOUN
fcis-31421	139	25	function	function	NOUN
fcis-31421	139	26	.	.	PUNCT
fcis-31421	140	1	of	of	ADP
fcis-31421	140	2	particular	particular	ADJ
fcis-31421	140	3	note	note	NOUN
fcis-31421	140	4	,	,	PUNCT
fcis-31421	140	5	this	this	DET
fcis-31421	140	6	study	study	NOUN
fcis-31421	140	7	employs	employ	VERB
fcis-31421	140	8	a	a	DET
fcis-31421	140	9	weighted	weighted	ADJ
fcis-31421	140	10	bidirectional	bidirectional	ADJ
fcis-31421	140	11	feature	feature	NOUN
fcis-31421	140	12	pyramid	pyramid	NOUN
fcis-31421	140	13	network	network	PROPN
fcis-31421	140	14	(	(	PUNCT
fcis-31421	140	15	bifpn	bifpn	PROPN
fcis-31421	140	16	)	)	PUNCT
fcis-31421	140	17	instead	instead	ADV
fcis-31421	140	18	of	of	ADP
fcis-31421	140	19	the	the	DET
fcis-31421	140	20	traditional	traditional	ADJ
fcis-31421	140	21	feature	feature	NOUN
fcis-31421	140	22	pyramid	pyramid	NOUN
fcis-31421	140	23	structure	structure	NOUN
fcis-31421	140	24	,	,	PUNCT
fcis-31421	140	25	enhancing	enhance	VERB
fcis-31421	140	26	the	the	DET
fcis-31421	140	27	robustness	robustness	NOUN
fcis-31421	140	28	of	of	ADP
fcis-31421	140	29	multi	multi	ADJ
fcis-31421	140	30	-	-	ADJ
fcis-31421	140	31	scale	scale	ADJ
fcis-31421	140	32	feature	feature	NOUN
fcis-31421	140	33	fusion	fusion	NOUN
fcis-31421	141	1	[	[	X
fcis-31421	141	2	12	12	NUM
fcis-31421	141	3	]	]	PUNCT
fcis-31421	141	4	.	.	PUNCT
fcis-31421	142	1	the	the	DET
fcis-31421	142	2	experimental	experimental	ADJ
fcis-31421	142	3	results	result	NOUN
fcis-31421	142	4	show	show	VERB
fcis-31421	142	5	that	that	SCONJ
fcis-31421	142	6	the	the	DET
fcis-31421	142	7	improved	improved	ADJ
fcis-31421	142	8	model	model	NOUN
fcis-31421	142	9	achieves	achieve	VERB
fcis-31421	142	10	high	high	ADJ
fcis-31421	142	11	accuracy	accuracy	NOUN
fcis-31421	142	12	,	,	PUNCT
fcis-31421	142	13	recall	recall	NOUN
fcis-31421	142	14	,	,	PUNCT
fcis-31421	142	15	map	map	NOUN
fcis-31421	142	16	,	,	PUNCT
fcis-31421	142	17	and	and	CCONJ
fcis-31421	142	18	map@0.50.95	map@0.50.95	PROPN
fcis-31421	142	19	significant	significant	ADJ
fcis-31421	142	20	improvements	improvement	NOUN
fcis-31421	142	21	of	of	ADP
fcis-31421	142	22	4.2	4.2	NUM
fcis-31421	142	23	%	%	NOUN
fcis-31421	142	24	,	,	PUNCT
fcis-31421	142	25	3.8	3.8	NUM
fcis-31421	142	26	%	%	NOUN
fcis-31421	142	27	,	,	PUNCT
fcis-31421	142	28	4.6	4.6	NUM
fcis-31421	142	29	%	%	NOUN
fcis-31421	142	30	,	,	PUNCT
fcis-31421	142	31	and	and	CCONJ
fcis-31421	142	32	2.2	2.2	NUM
fcis-31421	142	33	%	%	NOUN
fcis-31421	142	34	were	be	AUX
fcis-31421	142	35	achieved	achieve	VERB
fcis-31421	142	36	in	in	ADP
fcis-31421	142	37	key	key	ADJ
fcis-31421	142	38	indicators	indicator	NOUN
fcis-31421	142	39	.	.	PUNCT
fcis-31421	143	1	however	however	ADV
fcis-31421	143	2	,	,	PUNCT
fcis-31421	143	3	these	these	DET
fcis-31421	143	4	performance	performance	NOUN
fcis-31421	143	5	improvements	improvement	NOUN
fcis-31421	143	6	are	be	AUX
fcis-31421	143	7	accompanied	accompany	VERB
fcis-31421	143	8	by	by	ADP
fcis-31421	143	9	a	a	DET
fcis-31421	143	10	decrease	decrease	NOUN
fcis-31421	143	11	of	of	ADP
fcis-31421	143	12	about	about	ADV
fcis-31421	143	13	5	5	NUM
fcis-31421	143	14	%	%	NOUN
fcis-31421	143	15	in	in	ADP
fcis-31421	143	16	fps	fps	PROPN
fcis-31421	143	17	,	,	PUNCT
fcis-31421	143	18	which	which	PRON
fcis-31421	143	19	may	may	AUX
fcis-31421	143	20	have	have	VERB
fcis-31421	143	21	a	a	DET
fcis-31421	143	22	certain	certain	ADJ
fcis-31421	143	23	impact	impact	NOUN
fcis-31421	143	24	on	on	ADP
fcis-31421	143	25	the	the	DET
fcis-31421	143	26	response	response	NOUN
fcis-31421	143	27	speed	speed	NOUN
fcis-31421	143	28	,	,	PUNCT
fcis-31421	143	29	missed	miss	VERB
fcis-31421	143	30	detection	detection	NOUN
fcis-31421	143	31	rate	rate	NOUN
fcis-31421	143	32	,	,	PUNCT
fcis-31421	143	33	and	and	CCONJ
fcis-31421	143	34	operational	operational	ADJ
fcis-31421	143	35	stability	stability	NOUN
fcis-31421	143	36	of	of	ADP
fcis-31421	143	37	real	real	ADJ
fcis-31421	143	38	-	-	PUNCT
fcis-31421	143	39	time	time	NOUN
fcis-31421	143	40	detection	detection	NOUN
fcis-31421	143	41	systems	system	NOUN
fcis-31421	143	42	.	.	PUNCT
fcis-31421	144	1	the	the	DET
fcis-31421	144	2	bensheng	bensheng	PROPN
fcis-31421	144	3	yun	yun	PROPN
fcis-31421	144	4	research	research	NOUN
fcis-31421	144	5	team	team	NOUN
fcis-31421	144	6	has	have	AUX
fcis-31421	144	7	developed	develop	VERB
fcis-31421	144	8	a	a	DET
fcis-31421	144	9	lightweight	lightweight	ADJ
fcis-31421	144	10	forest	forest	NOUN
fcis-31421	144	11	fire	fire	NOUN
fcis-31421	144	12	detection	detection	NOUN
fcis-31421	144	13	model	model	NOUN
fcis-31421	144	14	ffyolo	ffyolo	PROPN
fcis-31421	144	15	,	,	PUNCT
fcis-31421	144	16	which	which	PRON
fcis-31421	144	17	has	have	AUX
fcis-31421	144	18	undergone	undergo	VERB
fcis-31421	144	19	multiple	multiple	ADJ
fcis-31421	144	20	innovative	innovative	ADJ
fcis-31421	144	21	improvements	improvement	NOUN
fcis-31421	144	22	based	base	VERB
fcis-31421	144	23	on	on	ADP
fcis-31421	144	24	the	the	DET
fcis-31421	144	25	yolov8	yolov8	NOUN
fcis-31421	144	26	architecture	architecture	NOUN
fcis-31421	144	27	.	.	PUNCT
fcis-31421	145	1	in	in	ADP
fcis-31421	145	2	terms	term	NOUN
fcis-31421	145	3	of	of	ADP
fcis-31421	145	4	feature	feature	NOUN
fcis-31421	145	5	extraction	extraction	NOUN
fcis-31421	145	6	,	,	PUNCT
fcis-31421	145	7	researchers	researcher	NOUN
fcis-31421	145	8	have	have	AUX
fcis-31421	145	9	designed	design	VERB
fcis-31421	145	10	a	a	DET
fcis-31421	145	11	channel	channel	NOUN
fcis-31421	145	12	prior	prior	ADV
fcis-31421	145	13	extended	extend	VERB
fcis-31421	145	14	attention	attention	NOUN
fcis-31421	145	15	module	module	NOUN
fcis-31421	145	16	(	(	PUNCT
fcis-31421	145	17	cpda	cpda	PROPN
fcis-31421	145	18	)	)	PUNCT
fcis-31421	145	19	to	to	PART
fcis-31421	145	20	enhance	enhance	VERB
fcis-31421	145	21	the	the	DET
fcis-31421	145	22	ability	ability	NOUN
fcis-31421	145	23	to	to	PART
fcis-31421	145	24	capture	capture	VERB
fcis-31421	145	25	fire	fire	NOUN
fcis-31421	145	26	and	and	CCONJ
fcis-31421	145	27	smoke	smoke	NOUN
fcis-31421	145	28	features	feature	NOUN
fcis-31421	145	29	.	.	PUNCT
fcis-31421	146	1	the	the	DET
fcis-31421	146	2	detection	detection	NOUN
fcis-31421	146	3	head	head	NOUN
fcis-31421	146	4	adopts	adopt	VERB
fcis-31421	146	5	a	a	DET
fcis-31421	146	6	new	new	ADJ
fcis-31421	146	7	hybrid	hybrid	ADJ
fcis-31421	146	8	layered	layered	ADJ
fcis-31421	146	9	structure	structure	NOUN
fcis-31421	146	10	(	(	PUNCT
fcis-31421	146	11	mcdh	mcdh	NOUN
fcis-31421	146	12	)	)	PUNCT
fcis-31421	146	13	to	to	PART
fcis-31421	146	14	improve	improve	VERB
fcis-31421	146	15	detection	detection	NOUN
fcis-31421	146	16	performance	performance	NOUN
fcis-31421	146	17	.	.	PUNCT
fcis-31421	147	1	to	to	PART
fcis-31421	147	2	improve	improve	VERB
fcis-31421	147	3	model	model	NOUN
fcis-31421	147	4	accuracy	accuracy	NOUN
fcis-31421	147	5	,	,	PUNCT
fcis-31421	147	6	the	the	DET
fcis-31421	147	7	mpdiou	mpdiou	NOUN
fcis-31421	147	8	method	method	NOUN
fcis-31421	147	9	was	be	AUX
fcis-31421	147	10	innovatively	innovatively	ADV
fcis-31421	147	11	introduced	introduce	VERB
fcis-31421	147	12	to	to	PART
fcis-31421	147	13	optimize	optimize	VERB
fcis-31421	147	14	bounding	bounding	NOUN
fcis-31421	147	15	box	box	NOUN
fcis-31421	147	16	regression	regression	NOUN
fcis-31421	147	17	and	and	CCONJ
fcis-31421	147	18	classification	classification	NOUN
fcis-31421	147	19	performance	performance	NOUN
fcis-31421	147	20	.	.	PUNCT
fcis-31421	148	1	the	the	DET
fcis-31421	148	2	network	network	NOUN
fcis-31421	148	3	neck	neck	NOUN
fcis-31421	148	4	adopts	adopt	VERB
fcis-31421	148	5	a	a	DET
fcis-31421	148	6	lightweight	lightweight	ADJ
fcis-31421	148	7	gsconv	gsconv	NOUN
fcis-31421	148	8	module	module	NOUN
fcis-31421	148	9	,	,	PUNCT
fcis-31421	148	10	which	which	PRON
fcis-31421	148	11	significantly	significantly	ADV
fcis-31421	148	12	reduces	reduce	VERB
fcis-31421	148	13	the	the	DET
fcis-31421	148	14	number	number	NOUN
fcis-31421	148	15	of	of	ADP
fcis-31421	148	16	parameters	parameter	NOUN
fcis-31421	148	17	while	while	SCONJ
fcis-31421	148	18	ensuring	ensure	VERB
fcis-31421	148	19	accuracy	accuracy	NOUN
fcis-31421	148	20	.	.	PUNCT
fcis-31421	149	1	the	the	DET
fcis-31421	149	2	use	use	NOUN
fcis-31421	149	3	of	of	ADP
fcis-31421	149	4	knowledge	knowledge	NOUN
fcis-31421	149	5	refinement	refinement	NOUN
fcis-31421	149	6	strategies	strategy	NOUN
fcis-31421	149	7	during	during	ADP
fcis-31421	149	8	the	the	DET
fcis-31421	149	9	training	training	NOUN
fcis-31421	149	10	process	process	NOUN
fcis-31421	149	11	effectively	effectively	ADV
fcis-31421	149	12	improved	improve	VERB
fcis-31421	149	13	the	the	DET
fcis-31421	149	14	model	model	NOUN
fcis-31421	149	15	's	's	PART
fcis-31421	149	16	generalization	generalization	NOUN
fcis-31421	149	17	ability	ability	NOUN
fcis-31421	149	18	and	and	CCONJ
fcis-31421	149	19	reduced	reduce	VERB
fcis-31421	149	20	false	false	ADJ
fcis-31421	149	21	positive	positive	ADJ
fcis-31421	149	22	rates	rate	NOUN
fcis-31421	149	23	[	[	X
fcis-31421	149	24	13	13	NUM
fcis-31421	149	25	]	]	PUNCT
fcis-31421	149	26	.	.	PUNCT
fcis-31421	150	1	the	the	DET
fcis-31421	150	2	experimental	experimental	ADJ
fcis-31421	150	3	data	datum	NOUN
fcis-31421	150	4	shows	show	VERB
fcis-31421	150	5	that	that	SCONJ
fcis-31421	150	6	the	the	DET
fcis-31421	150	7	improved	improved	ADJ
fcis-31421	150	8	model	model	NOUN
fcis-31421	150	9	achieves	achieve	VERB
fcis-31421	150	10	88.8	88.8	NUM
fcis-31421	150	11	%	%	NOUN
fcis-31421	150	12	in	in	ADP
fcis-31421	150	13	map0.5	map0.5	PROPN
fcis-31421	150	14	index	index	NOUN
fcis-31421	150	15	,	,	PUNCT
fcis-31421	150	16	which	which	PRON
fcis-31421	150	17	is	be	AUX
fcis-31421	150	18	3.4	3.4	NUM
fcis-31421	150	19	%	%	NOUN
fcis-31421	150	20	higher	high	ADJ
fcis-31421	150	21	than	than	ADP
fcis-31421	150	22	the	the	DET
fcis-31421	150	23	original	original	ADJ
fcis-31421	150	24	model	model	NOUN
fcis-31421	150	25	.	.	PUNCT
fcis-31421	151	1	the	the	DET
fcis-31421	151	2	number	number	NOUN
fcis-31421	151	3	of	of	ADP
fcis-31421	151	4	model	model	NOUN
fcis-31421	151	5	parameters	parameter	NOUN
fcis-31421	151	6	is	be	AUX
fcis-31421	151	7	reduced	reduce	VERB
fcis-31421	151	8	by	by	ADP
fcis-31421	151	9	25.3	25.3	NUM
fcis-31421	151	10	%	%	NOUN
fcis-31421	151	11	,	,	PUNCT
fcis-31421	151	12	and	and	CCONJ
fcis-31421	151	13	the	the	DET
fcis-31421	151	14	inference	inference	NOUN
fcis-31421	151	15	speed	speed	NOUN
fcis-31421	151	16	is	be	AUX
fcis-31421	151	17	improved	improve	VERB
fcis-31421	151	18	by	by	ADP
fcis-31421	151	19	9.3	9.3	NUM
fcis-31421	151	20	%	%	NOUN
fcis-31421	151	21	.	.	PUNCT
fcis-31421	152	1	thanks	thank	NOUN
fcis-31421	152	2	to	to	ADP
fcis-31421	152	3	its	its	PRON
fcis-31421	152	4	lower	low	ADJ
fcis-31421	152	5	parameter	parameter	NOUN
fcis-31421	152	6	count	count	NOUN
fcis-31421	152	7	and	and	CCONJ
fcis-31421	152	8	computational	computational	ADJ
fcis-31421	152	9	complexity	complexity	NOUN
fcis-31421	152	10	,	,	PUNCT
fcis-31421	152	11	this	this	DET
fcis-31421	152	12	model	model	NOUN
fcis-31421	152	13	is	be	AUX
fcis-31421	152	14	particularly	particularly	ADV
fcis-31421	152	15	suitable	suitable	ADJ
fcis-31421	152	16	for	for	ADP
fcis-31421	152	17	deployment	deployment	NOUN
fcis-31421	152	18	on	on	ADP
fcis-31421	152	19	edge	edge	NOUN
fcis-31421	152	20	devices	device	NOUN
fcis-31421	152	21	with	with	ADP
fcis-31421	152	22	limited	limited	ADJ
fcis-31421	152	23	computing	computing	NOUN
fcis-31421	152	24	resources	resource	NOUN
fcis-31421	152	25	.	.	PUNCT
fcis-31421	153	1	3	3	X
fcis-31421	153	2	.	.	X
fcis-31421	153	3	method	method	NOUN
fcis-31421	153	4	if	if	SCONJ
fcis-31421	153	5	wanting	want	VERB
fcis-31421	153	6	to	to	PART
fcis-31421	153	7	realize	realize	VERB
fcis-31421	153	8	large	large	ADJ
fcis-31421	153	9	-	-	PUNCT
fcis-31421	153	10	scale	scale	NOUN
fcis-31421	153	11	fire	fire	NOUN
fcis-31421	153	12	prevention	prevention	NOUN
fcis-31421	153	13	detection	detection	NOUN
fcis-31421	153	14	,	,	PUNCT
fcis-31421	153	15	yolov8	yolov8	PROPN
fcis-31421	153	16	is	be	AUX
fcis-31421	153	17	a	a	DET
fcis-31421	153	18	good	good	ADJ
fcis-31421	153	19	experimental	experimental	ADJ
fcis-31421	153	20	object	object	NOUN
fcis-31421	153	21	,	,	PUNCT
fcis-31421	153	22	because	because	SCONJ
fcis-31421	153	23	yolov8	yolov8	NOUN
fcis-31421	153	24	is	be	AUX
fcis-31421	153	25	able	able	ADJ
fcis-31421	153	26	to	to	PART
fcis-31421	153	27	integrate	integrate	VERB
fcis-31421	153	28	the	the	DET
fcis-31421	153	29	multi	multi	ADJ
fcis-31421	153	30	-	-	ADJ
fcis-31421	153	31	scale	scale	ADJ
fcis-31421	153	32	attention	attention	NOUN
fcis-31421	153	33	mechanism	mechanism	NOUN
fcis-31421	153	34	and	and	CCONJ
fcis-31421	153	35	cross	cross	ADJ
fcis-31421	153	36	-	-	ADJ
fcis-31421	153	37	modal	modal	ADJ
fcis-31421	153	38	lightweight	lightweight	NOUN
fcis-31421	153	39	design	design	NOUN
fcis-31421	153	40	,	,	PUNCT
fcis-31421	153	41	considering	consider	VERB
fcis-31421	153	42	the	the	DET
fcis-31421	153	43	high	high	ADJ
fcis-31421	153	44	accuracy	accuracy	NOUN
fcis-31421	153	45	and	and	CCONJ
fcis-31421	153	46	real	real	ADJ
fcis-31421	153	47	-	-	PUNCT
fcis-31421	153	48	time	time	NOUN
fcis-31421	153	49	,	,	PUNCT
fcis-31421	153	50	especially	especially	ADV
fcis-31421	153	51	suitable	suitable	ADJ
fcis-31421	153	52	for	for	ADP
fcis-31421	153	53	uavs	uavs	NOUN
fcis-31421	153	54	,	,	PUNCT
fcis-31421	153	55	satellite	satellite	NOUN
fcis-31421	153	56	remote	remote	ADJ
fcis-31421	153	57	sensing	sensing	NOUN
fcis-31421	153	58	and	and	CCONJ
fcis-31421	153	59	other	other	ADJ
fcis-31421	153	60	large	large	ADJ
fcis-31421	153	61	-	-	PUNCT
fcis-31421	153	62	scale	scale	NOUN
fcis-31421	153	63	inspection	inspection	NOUN
fcis-31421	153	64	needs	need	NOUN
fcis-31421	153	65	.	.	PUNCT
fcis-31421	154	1	compared	compare	VERB
fcis-31421	154	2	to	to	ADP
fcis-31421	154	3	cnn	cnn	PROPN
fcis-31421	154	4	,	,	PUNCT
fcis-31421	154	5	yolov8	yolov8	PROPN
fcis-31421	154	6	supports	support	VERB
fcis-31421	154	7	continuous	continuous	ADJ
fcis-31421	154	8	frame	frame	NOUN
fcis-31421	154	9	detection	detection	NOUN
fcis-31421	154	10	of	of	ADP
fcis-31421	154	11	video	video	NOUN
fcis-31421	154	12	streams	stream	NOUN
fcis-31421	154	13	,	,	PUNCT
fcis-31421	154	14	combined	combine	VERB
fcis-31421	154	15	with	with	ADP
fcis-31421	154	16	dynamic	dynamic	ADJ
fcis-31421	154	17	label	label	NOUN
fcis-31421	154	18	assignment	assignment	NOUN
fcis-31421	154	19	to	to	PART
fcis-31421	154	20	improve	improve	VERB
fcis-31421	154	21	flame	flame	NOUN
fcis-31421	154	22	trajectory	trajectory	NOUN
fcis-31421	154	23	stability	stability	NOUN
fcis-31421	154	24	,	,	PUNCT
fcis-31421	154	25	while	while	SCONJ
fcis-31421	154	26	cnn	cnn	PROPN
fcis-31421	154	27	is	be	AUX
fcis-31421	154	28	mostly	mostly	ADV
fcis-31421	154	29	used	use	VERB
fcis-31421	154	30	for	for	ADP
fcis-31421	154	31	static	static	ADJ
fcis-31421	154	32	image	image	NOUN
fcis-31421	154	33	processing	processing	NOUN
fcis-31421	154	34	,	,	PUNCT
fcis-31421	154	35	and	and	CCONJ
fcis-31421	154	36	can	can	AUX
fcis-31421	154	37	not	not	PART
fcis-31421	154	38	capture	capture	VERB
fcis-31421	154	39	temporal	temporal	ADJ
fcis-31421	154	40	features	feature	NOUN
fcis-31421	154	41	such	such	ADJ
fcis-31421	154	42	as	as	ADP
fcis-31421	154	43	flame	flame	NOUN
fcis-31421	154	44	diffusion	diffusion	NOUN
fcis-31421	154	45	and	and	CCONJ
fcis-31421	154	46	smoke	smoke	VERB
fcis-31421	154	47	fluttering	fluttering	NOUN
fcis-31421	154	48	.	.	PUNCT
fcis-31421	155	1	moreover	moreover	ADV
fcis-31421	155	2	,	,	PUNCT
fcis-31421	155	3	traditional	traditional	ADJ
fcis-31421	155	4	cnn	cnn	PROPN
fcis-31421	155	5	requires	require	VERB
fcis-31421	155	6	multiple	multiple	ADJ
fcis-31421	155	7	pooling	pooling	NOUN
fcis-31421	155	8	and	and	CCONJ
fcis-31421	155	9	fully	fully	ADV
fcis-31421	155	10	connected	connected	ADJ
fcis-31421	155	11	layers	layer	NOUN
fcis-31421	155	12	,	,	PUNCT
fcis-31421	155	13	and	and	CCONJ
fcis-31421	155	14	processing	processing	NOUN
fcis-31421	155	15	singleframe	singleframe	NOUN
fcis-31421	155	16	images	image	NOUN
fcis-31421	155	17	is	be	AUX
fcis-31421	155	18	time	time	NOUN
fcis-31421	155	19	-	-	PUNCT
fcis-31421	155	20	consuming	consume	VERB
fcis-31421	155	21	,	,	PUNCT
fcis-31421	155	22	making	make	VERB
fcis-31421	155	23	it	it	PRON
fcis-31421	155	24	difficult	difficult	ADJ
fcis-31421	155	25	to	to	PART
fcis-31421	155	26	meet	meet	VERB
fcis-31421	155	27	the	the	DET
fcis-31421	155	28	demand	demand	NOUN
fcis-31421	155	29	for	for	ADP
fcis-31421	155	30	real	real	ADJ
fcis-31421	155	31	-	-	PUNCT
fcis-31421	155	32	time	time	NOUN
fcis-31421	155	33	video	video	NOUN
fcis-31421	155	34	stream	stream	NOUN
fcis-31421	155	35	detection	detection	NOUN
fcis-31421	155	36	.	.	PUNCT
fcis-31421	156	1	however	however	ADV
fcis-31421	156	2	,	,	PUNCT
fcis-31421	156	3	yolov8	yolov8	PROPN
fcis-31421	156	4	can	can	AUX
fcis-31421	156	5	increase	increase	VERB
fcis-31421	156	6	the	the	DET
fcis-31421	156	7	inference	inference	NOUN
fcis-31421	156	8	speed	speed	NOUN
fcis-31421	156	9	to	to	ADP
fcis-31421	156	10	90	90	NUM
fcis-31421	156	11	-	-	SYM
fcis-31421	156	12	120	120	NUM
fcis-31421	156	13	fps	fps	NOUN
fcis-31421	156	14	to	to	PART
fcis-31421	156	15	meet	meet	VERB
fcis-31421	156	16	the	the	DET
fcis-31421	156	17	demand	demand	NOUN
fcis-31421	156	18	for	for	ADP
fcis-31421	156	19	fire	fire	NOUN
fcis-31421	156	20	warning	warning	NOUN
fcis-31421	156	21	in	in	ADP
fcis-31421	156	22	a	a	DET
fcis-31421	156	23	very	very	ADV
fcis-31421	156	24	short	short	ADJ
fcis-31421	156	25	time	time	NOUN
fcis-31421	156	26	.	.	PUNCT
fcis-31421	157	1	multimodal	multimodal	NOUN
fcis-31421	157	2	need	need	VERB
fcis-31421	157	3	to	to	PART
fcis-31421	157	4	integrate	integrate	VERB
fcis-31421	157	5	visible	visible	ADJ
fcis-31421	157	6	light	light	NOUN
fcis-31421	157	7	,	,	PUNCT
fcis-31421	157	8	infrared	infrare	VERB
fcis-31421	157	9	,	,	PUNCT
fcis-31421	157	10	sensor	sensor	NOUN
fcis-31421	157	11	multi	multi	ADJ
fcis-31421	157	12	-	-	ADJ
fcis-31421	157	13	modal	modal	ADJ
fcis-31421	157	14	model	model	NOUN
fcis-31421	157	15	,	,	PUNCT
fcis-31421	157	16	the	the	DET
fcis-31421	157	17	need	need	NOUN
fcis-31421	157	18	for	for	ADP
fcis-31421	157	19	gpu	gpu	NOUN
fcis-31421	157	20	cluster	cluster	NOUN
fcis-31421	157	21	support	support	NOUN
fcis-31421	157	22	,	,	PUNCT
fcis-31421	157	23	difficult	difficult	ADJ
fcis-31421	157	24	to	to	PART
fcis-31421	157	25	run	run	VERB
fcis-31421	157	26	in	in	ADP
fcis-31421	157	27	real	real	ADJ
fcis-31421	157	28	time	time	NOUN
fcis-31421	157	29	,	,	PUNCT
fcis-31421	157	30	and	and	CCONJ
fcis-31421	157	31	need	need	VERB
fcis-31421	157	32	to	to	PART
fcis-31421	157	33	align	align	VERB
fcis-31421	157	34	the	the	DET
fcis-31421	157	35	multi	multi	ADJ
fcis-31421	157	36	-	-	ADJ
fcis-31421	157	37	source	source	NOUN
fcis-31421	157	38	data	datum	NOUN
fcis-31421	157	39	,	,	PUNCT
fcis-31421	157	40	labeling	labeling	NOUN
fcis-31421	157	41	costs	cost	NOUN
fcis-31421	157	42	for	for	ADP
fcis-31421	157	43	the	the	DET
fcis-31421	157	44	single	single	ADJ
fcis-31421	157	45	-	-	PUNCT
fcis-31421	157	46	modal	modal	NOUN
fcis-31421	157	47	3	3	NUM
fcis-31421	157	48	-	-	SYM
fcis-31421	157	49	5	5	NUM
fcis-31421	157	50	times	time	NOUN
fcis-31421	157	51	,	,	PUNCT
fcis-31421	157	52	want	want	VERB
fcis-31421	157	53	to	to	PART
fcis-31421	157	54	achieve	achieve	VERB
fcis-31421	157	55	better	well	ADJ
fcis-31421	157	56	detection	detection	NOUN
fcis-31421	157	57	results	result	NOUN
fcis-31421	157	58	also	also	ADV
fcis-31421	157	59	need	need	VERB
fcis-31421	157	60	high	high	ADJ
fcis-31421	157	61	-	-	PUNCT
fcis-31421	157	62	cost	cost	NOUN
fcis-31421	157	63	hardware	hardware	NOUN
fcis-31421	157	64	support	support	NOUN
fcis-31421	157	65	,	,	PUNCT
fcis-31421	157	66	it	it	PRON
fcis-31421	157	67	is	be	AUX
fcis-31421	157	68	difficult	difficult	ADJ
fcis-31421	157	69	to	to	PART
fcis-31421	157	70	use	use	VERB
fcis-31421	157	71	on	on	ADP
fcis-31421	157	72	a	a	DET
fcis-31421	157	73	wide	wide	ADJ
fcis-31421	157	74	scale	scale	NOUN
fcis-31421	157	75	.	.	PUNCT
fcis-31421	158	1	yolov8	yolov8	NOUN
fcis-31421	158	2	is	be	AUX
fcis-31421	158	3	much	much	ADV
fcis-31421	158	4	more	more	ADV
fcis-31421	158	5	lightweight	lightweight	ADJ
fcis-31421	158	6	,	,	PUNCT
fcis-31421	158	7	does	do	AUX
fcis-31421	158	8	not	not	PART
fcis-31421	158	9	need	need	VERB
fcis-31421	158	10	so	so	ADV
fcis-31421	158	11	much	much	ADJ
fcis-31421	158	12	arithmetic	arithmetic	ADJ
fcis-31421	158	13	support	support	NOUN
fcis-31421	158	14	,	,	PUNCT
fcis-31421	158	15	and	and	CCONJ
fcis-31421	158	16	by	by	ADP
fcis-31421	158	17	adding	add	VERB
fcis-31421	158	18	modifications	modification	NOUN
fcis-31421	158	19	to	to	ADP
fcis-31421	158	20	the	the	DET
fcis-31421	158	21	module	module	NOUN
fcis-31421	158	22	can	can	AUX
fcis-31421	158	23	increase	increase	VERB
fcis-31421	158	24	the	the	DET
fcis-31421	158	25	application	application	NOUN
fcis-31421	158	26	of	of	ADP
fcis-31421	158	27	each	each	DET
fcis-31421	158	28	scenario	scenario	NOUN
fcis-31421	158	29	.	.	PUNCT
fcis-31421	159	1	yolov8	yolov8	NOUN
fcis-31421	159	2	is	be	AUX
fcis-31421	159	3	more	more	ADV
fcis-31421	159	4	lightweight	lightweight	ADJ
fcis-31421	159	5	,	,	PUNCT
fcis-31421	159	6	does	do	AUX
fcis-31421	159	7	not	not	PART
fcis-31421	159	8	require	require	VERB
fcis-31421	159	9	so	so	ADV
fcis-31421	159	10	much	much	ADJ
fcis-31421	159	11	arithmetic	arithmetic	ADJ
fcis-31421	159	12	support	support	NOUN
fcis-31421	159	13	,	,	PUNCT
fcis-31421	159	14	and	and	CCONJ
fcis-31421	159	15	by	by	ADP
fcis-31421	159	16	adding	add	VERB
fcis-31421	159	17	and	and	CCONJ
fcis-31421	159	18	modifying	modify	VERB
fcis-31421	159	19	the	the	DET
fcis-31421	159	20	module	module	NOUN
fcis-31421	159	21	to	to	PART
fcis-31421	159	22	increase	increase	VERB
fcis-31421	159	23	the	the	DET
fcis-31421	159	24	application	application	NOUN
fcis-31421	159	25	of	of	ADP
fcis-31421	159	26	each	each	DET
fcis-31421	159	27	scene	scene	NOUN
fcis-31421	159	28	,	,	PUNCT
fcis-31421	159	29	ensuring	ensure	VERB
fcis-31421	159	30	that	that	SCONJ
fcis-31421	159	31	the	the	DET
fcis-31421	159	32	cost	cost	NOUN
fcis-31421	159	33	is	be	AUX
fcis-31421	159	34	low	low	ADJ
fcis-31421	159	35	,	,	PUNCT
fcis-31421	159	36	to	to	PART
fcis-31421	159	37	meet	meet	VERB
fcis-31421	159	38	the	the	DET
fcis-31421	159	39	wide	wide	ADJ
fcis-31421	159	40	range	range	NOUN
fcis-31421	159	41	of	of	ADP
fcis-31421	159	42	use	use	NOUN
fcis-31421	159	43	.	.	PUNCT
fcis-31421	160	1	so	so	ADV
fcis-31421	160	2	why	why	SCONJ
fcis-31421	160	3	choose	choose	VERB
fcis-31421	160	4	the	the	DET
fcis-31421	160	5	model	model	NOUN
fcis-31421	160	6	yolov8	yolov8	NOUN
fcis-31421	160	7	in	in	ADP
fcis-31421	160	8	the	the	DET
fcis-31421	160	9	yolo	yolo	ADJ
fcis-31421	160	10	series	series	NOUN
fcis-31421	160	11	?	?	PUNCT
fcis-31421	161	1	compared	compare	VERB
fcis-31421	161	2	with	with	ADP
fcis-31421	161	3	previous	previous	ADJ
fcis-31421	161	4	yolo	yolo	ADJ
fcis-31421	161	5	versions	version	NOUN
fcis-31421	161	6	(	(	PUNCT
fcis-31421	161	7	yolov3	yolov3	PROPN
fcis-31421	161	8	/	/	SYM
fcis-31421	161	9	yolov5	yolov5	PROPN
fcis-31421	161	10	)	)	PUNCT
fcis-31421	161	11	,	,	PUNCT
fcis-31421	161	12	yolov8	yolov8	NOUN
fcis-31421	161	13	adds	add	VERB
fcis-31421	161	14	p2	p2	PROPN
fcis-31421	161	15	detection	detection	NOUN
fcis-31421	161	16	layer	layer	NOUN
fcis-31421	161	17	,	,	PUNCT
fcis-31421	161	18	supports	support	VERB
fcis-31421	161	19	detection	detection	NOUN
fcis-31421	161	20	of	of	ADP
fcis-31421	161	21	tiny	tiny	ADJ
fcis-31421	161	22	targets	target	NOUN
fcis-31421	161	23	with	with	ADP
fcis-31421	161	24	resolution	resolution	NOUN
fcis-31421	161	25	≥	≥	NUM
fcis-31421	161	26	4×4,the	4×4,the	DET
fcis-31421	161	27	small	small	ADJ
fcis-31421	161	28	target	target	NOUN
fcis-31421	161	29	map@0.5	map@0.5	NOUN
fcis-31421	161	30	increases	increase	NOUN
fcis-31421	161	31	to	to	ADP
fcis-31421	161	32	65	65	NUM
fcis-31421	161	33	%	%	NOUN
fcis-31421	161	34	and	and	CCONJ
fcis-31421	161	35	classification	classification	NOUN
fcis-31421	161	36	and	and	CCONJ
fcis-31421	161	37	regression	regression	NOUN
fcis-31421	161	38	tasks	task	NOUN
fcis-31421	161	39	are	be	AUX
fcis-31421	161	40	optimized	optimize	VERB
fcis-31421	161	41	independently	independently	ADV
fcis-31421	161	42	.	.	PUNCT
fcis-31421	162	1	yolov3	yolov3	PROPN
fcis-31421	162	2	/	/	SYM
fcis-31421	162	3	yolov5	yolov5	NOUN
fcis-31421	162	4	has	have	VERB
fcis-31421	162	5	weak	weak	ADJ
fcis-31421	162	6	detection	detection	NOUN
fcis-31421	162	7	of	of	ADP
fcis-31421	162	8	small	small	ADJ
fcis-31421	162	9	flames	flame	NOUN
fcis-31421	162	10	,	,	PUNCT
fcis-31421	162	11	and	and	CCONJ
fcis-31421	162	12	high	high	ADJ
fcis-31421	162	13	-	-	PUNCT
fcis-31421	162	14	level	level	NOUN
fcis-31421	162	15	feature	feature	NOUN
fcis-31421	162	16	resolution	resolution	NOUN
fcis-31421	162	17	is	be	AUX
fcis-31421	162	18	insufficient	insufficient	ADJ
fcis-31421	162	19	.	.	PUNCT
fcis-31421	163	1	yolov8	yolov8	PROPN
fcis-31421	163	2	adopts	adopt	VERB
fcis-31421	163	3	the	the	DET
fcis-31421	163	4	ghostnet	ghostnet	NOUN
fcis-31421	163	5	lightweight	lightweight	NOUN
fcis-31421	163	6	module	module	NOUN
fcis-31421	163	7	,	,	PUNCT
fcis-31421	163	8	the	the	DET
fcis-31421	163	9	model	model	NOUN
fcis-31421	163	10	volume	volume	NOUN
fcis-31421	163	11	is	be	AUX
fcis-31421	163	12	compressed	compress	VERB
fcis-31421	163	13	to	to	ADP
fcis-31421	163	14	3.8	3.8	NUM
fcis-31421	163	15	mb	mb	NOUN
fcis-31421	163	16	,	,	PUNCT
fcis-31421	163	17	and	and	CCONJ
fcis-31421	163	18	it	it	PRON
fcis-31421	163	19	supports	support	VERB
fcis-31421	163	20	the	the	DET
fcis-31421	163	21	deployment	deployment	NOUN
fcis-31421	163	22	of	of	ADP
fcis-31421	163	23	edge	edge	NOUN
fcis-31421	163	24	devices	device	NOUN
fcis-31421	163	25	,	,	PUNCT
fcis-31421	163	26	which	which	PRON
fcis-31421	163	27	is	be	AUX
fcis-31421	163	28	convenient	convenient	ADJ
fcis-31421	163	29	for	for	ADP
fcis-31421	163	30	large	large	ADJ
fcis-31421	163	31	-	-	PUNCT
fcis-31421	163	32	scale	scale	NOUN
fcis-31421	163	33	deployment	deployment	NOUN
fcis-31421	163	34	.	.	PUNCT
fcis-31421	164	1	yolov8	yolov8	NOUN
fcis-31421	164	2	can	can	AUX
fcis-31421	164	3	also	also	ADV
fcis-31421	164	4	improve	improve	VERB
fcis-31421	164	5	the	the	DET
fcis-31421	164	6	robustness	robustness	NOUN
fcis-31421	164	7	of	of	ADP
fcis-31421	164	8	complex	complex	ADJ
fcis-31421	164	9	lighting	lighting	NOUN
fcis-31421	164	10	and	and	CCONJ
fcis-31421	164	11	smoke	smoke	NOUN
fcis-31421	164	12	occlusion	occlusion	NOUN
fcis-31421	164	13	scenes	scene	NOUN
fcis-31421	164	14	after	after	ADP
fcis-31421	164	15	introducing	introduce	VERB
fcis-31421	164	16	multi	multi	ADJ
fcis-31421	164	17	-	-	ADJ
fcis-31421	164	18	scale	scale	ADJ
fcis-31421	164	19	feature	feature	NOUN
fcis-31421	164	20	fusion	fusion	NOUN
fcis-31421	164	21	and	and	CCONJ
fcis-31421	164	22	senet	senet	NOUN
fcis-31421	164	23	/	/	SYM
fcis-31421	164	24	cbam	cbam	NOUN
fcis-31421	164	25	attention	attention	NOUN
fcis-31421	164	26	mechanism	mechanism	NOUN
fcis-31421	164	27	.	.	PUNCT
fcis-31421	165	1	compared	compare	VERB
fcis-31421	165	2	with	with	ADP
fcis-31421	165	3	the	the	DET
fcis-31421	165	4	new	new	ADJ
fcis-31421	165	5	yolo	yolo	ADJ
fcis-31421	165	6	model	model	NOUN
fcis-31421	165	7	,	,	PUNCT
fcis-31421	165	8	although	although	SCONJ
fcis-31421	165	9	the	the	DET
fcis-31421	165	10	new	new	ADJ
fcis-31421	165	11	model	model	NOUN
fcis-31421	165	12	has	have	AUX
fcis-31421	165	13	improved	improve	VERB
fcis-31421	165	14	in	in	ADP
fcis-31421	165	15	accuracy	accuracy	NOUN
fcis-31421	165	16	,	,	PUNCT
fcis-31421	165	17	the	the	DET
fcis-31421	165	18	complexity	complexity	NOUN
fcis-31421	165	19	of	of	ADP
fcis-31421	165	20	the	the	DET
fcis-31421	165	21	model	model	NOUN
fcis-31421	165	22	is	be	AUX
fcis-31421	165	23	high	high	ADJ
fcis-31421	165	24	and	and	CCONJ
fcis-31421	165	25	it	it	PRON
fcis-31421	165	26	requires	require	VERB
fcis-31421	165	27	high	high	ADJ
fcis-31421	165	28	-	-	PUNCT
fcis-31421	165	29	performance	performance	NOUN
fcis-31421	165	30	hardware	hardware	NOUN
fcis-31421	165	31	support	support	NOUN
fcis-31421	165	32	,	,	PUNCT
fcis-31421	165	33	so	so	CCONJ
fcis-31421	165	34	the	the	DET
fcis-31421	165	35	cost	cost	NOUN
fcis-31421	165	36	of	of	ADP
fcis-31421	165	37	wide	wide	ADJ
fcis-31421	165	38	-	-	PUNCT
fcis-31421	165	39	scale	scale	NOUN
fcis-31421	165	40	use	use	NOUN
fcis-31421	165	41	will	will	AUX
fcis-31421	165	42	be	be	AUX
fcis-31421	165	43	much	much	ADV
fcis-31421	165	44	higher	high	ADJ
fcis-31421	165	45	than	than	ADP
fcis-31421	165	46	yolov8	yolov8	NOUN
fcis-31421	165	47	.	.	PUNCT
fcis-31421	166	1	yolov8	yolov8	PROPN
fcis-31421	166	2	also	also	ADV
fcis-31421	166	3	has	have	VERB
fcis-31421	166	4	significant	significant	ADJ
fcis-31421	166	5	improvements	improvement	NOUN
fcis-31421	166	6	in	in	ADP
fcis-31421	166	7	model	model	NOUN
fcis-31421	166	8	architecture	architecture	NOUN
fcis-31421	166	9	,	,	PUNCT
fcis-31421	166	10	using	use	VERB
fcis-31421	166	11	an	an	DET
fcis-31421	166	12	enhanced	enhanced	ADJ
fcis-31421	166	13	version	version	NOUN
fcis-31421	166	14	of	of	ADP
fcis-31421	166	15	cspdarknet53	cspdarknet53	NOUN
fcis-31421	166	16	on	on	ADP
fcis-31421	166	17	the	the	DET
fcis-31421	166	18	backbone	backbone	NOUN
fcis-31421	166	19	network	network	NOUN
fcis-31421	166	20	to	to	PART
fcis-31421	166	21	reduce	reduce	VERB
fcis-31421	166	22	computational	computational	ADJ
fcis-31421	166	23	redundancy	redundancy	NOUN
fcis-31421	166	24	by	by	ADP
fcis-31421	166	25	connecting	connect	VERB
fcis-31421	166	26	partially	partially	ADV
fcis-31421	166	27	across	across	ADP
fcis-31421	166	28	stages	stage	NOUN
fcis-31421	166	29	and	and	CCONJ
fcis-31421	166	30	introducing	introduce	VERB
fcis-31421	166	31	a	a	DET
fcis-31421	166	32	deeper	deep	ADJ
fcis-31421	166	33	network	network	NOUN
fcis-31421	166	34	structure	structure	NOUN
fcis-31421	166	35	to	to	PART
fcis-31421	166	36	improve	improve	VERB
fcis-31421	166	37	feature	feature	NOUN
fcis-31421	166	38	extraction	extraction	NOUN
fcis-31421	166	39	.	.	PUNCT
fcis-31421	167	1	the	the	DET
fcis-31421	167	2	multiscale	multiscale	ADJ
fcis-31421	167	3	feature	feature	NOUN
fcis-31421	167	4	fusion	fusion	NOUN
fcis-31421	167	5	structure	structure	NOUN
fcis-31421	167	6	of	of	ADP
fcis-31421	167	7	pan	pan	NOUN
fcis-31421	167	8	-	-	ADJ
fcis-31421	167	9	fpn	fpn	ADJ
fcis-31421	167	10	(	(	PUNCT
fcis-31421	167	11	path	path	NOUN
fcis-31421	167	12	aggregation	aggregation	NOUN
fcis-31421	167	13	network	network	NOUN
fcis-31421	167	14	+	+	CCONJ
fcis-31421	167	15	feature	feature	NOUN
fcis-31421	167	16	pyramid	pyramid	NOUN
fcis-31421	167	17	network	network	NOUN
fcis-31421	167	18	)	)	PUNCT
fcis-31421	167	19	is	be	AUX
fcis-31421	167	20	used	use	VERB
fcis-31421	167	21	on	on	ADP
fcis-31421	167	22	the	the	DET
fcis-31421	167	23	neck	neck	NOUN
fcis-31421	167	24	to	to	PART
fcis-31421	167	25	enhance	enhance	VERB
fcis-31421	167	26	the	the	DET
fcis-31421	167	27	combination	combination	NOUN
fcis-31421	167	28	of	of	ADP
fcis-31421	167	29	shallow	shallow	ADJ
fcis-31421	167	30	details	detail	NOUN
fcis-31421	167	31	and	and	CCONJ
fcis-31421	167	32	deep	deep	ADJ
fcis-31421	167	33	semantic	semantic	ADJ
fcis-31421	167	34	information	information	NOUN
fcis-31421	167	35	,	,	PUNCT
fcis-31421	167	36	and	and	CCONJ
fcis-31421	167	37	to	to	PART
fcis-31421	167	38	improve	improve	VERB
fcis-31421	167	39	the	the	DET
fcis-31421	167	40	small	small	ADJ
fcis-31421	167	41	target	target	NOUN
fcis-31421	167	42	detection	detection	NOUN
fcis-31421	167	43	effect	effect	NOUN
fcis-31421	167	44	.	.	PUNCT
fcis-31421	168	1	the	the	DET
fcis-31421	168	2	detection	detection	NOUN
fcis-31421	168	3	head	head	NOUN
fcis-31421	168	4	is	be	AUX
fcis-31421	168	5	changed	change	VERB
fcis-31421	168	6	to	to	ADP
fcis-31421	168	7	an	an	DET
fcis-31421	168	8	anchorless	anchorless	ADJ
fcis-31421	168	9	design	design	NOUN
fcis-31421	168	10	,	,	PUNCT
fcis-31421	168	11	which	which	PRON
fcis-31421	168	12	directly	directly	ADV
fcis-31421	168	13	predicts	predict	VERB
fcis-31421	168	14	the	the	DET
fcis-31421	168	15	center	center	NOUN
fcis-31421	168	16	point	point	NOUN
fcis-31421	168	17	and	and	CCONJ
fcis-31421	168	18	width	width	VERB
fcis-31421	168	19	and	and	CCONJ
fcis-31421	168	20	height	height	NOUN
fcis-31421	168	21	of	of	ADP
fcis-31421	168	22	the	the	DET
fcis-31421	168	23	target	target	NOUN
fcis-31421	168	24	,	,	PUNCT
fcis-31421	168	25	simplifying	simplify	VERB
fcis-31421	168	26	the	the	DET
fcis-31421	168	27	model	model	NOUN
fcis-31421	168	28	and	and	CCONJ
fcis-31421	168	29	reducing	reduce	VERB
fcis-31421	168	30	the	the	DET
fcis-31421	168	31	need	need	NOUN
fcis-31421	168	32	for	for	ADP
fcis-31421	168	33	hyperparameter	hyperparameter	NOUN
fcis-31421	168	34	tuning	tuning	NOUN
fcis-31421	168	35	.	.	PUNCT
fcis-31421	169	1	based	base	VERB
fcis-31421	169	2	on	on	ADP
fcis-31421	169	3	the	the	DET
fcis-31421	169	4	yolov8	yolov8	PROPN
fcis-31421	169	5	model	model	NOUN
fcis-31421	169	6	,	,	PUNCT
fcis-31421	169	7	this	this	DET
fcis-31421	169	8	paper	paper	NOUN
fcis-31421	169	9	designs	design	VERB
fcis-31421	169	10	a	a	DET
fcis-31421	169	11	method	method	NOUN
fcis-31421	169	12	to	to	PART
fcis-31421	169	13	achieve	achieve	VERB
fcis-31421	169	14	lightweighting	lightweighting	NOUN
fcis-31421	169	15	by	by	ADP
fcis-31421	169	16	modifying	modify	VERB
fcis-31421	169	17	the	the	DET
fcis-31421	169	18	56	56	NUM
fcis-31421	169	19	module	module	NOUN
fcis-31421	169	20	and	and	CCONJ
fcis-31421	169	21	network	network	NOUN
fcis-31421	169	22	structure	structure	NOUN
fcis-31421	169	23	and	and	CCONJ
fcis-31421	169	24	other	other	ADJ
fcis-31421	169	25	methods	method	NOUN
fcis-31421	169	26	,	,	PUNCT
fcis-31421	169	27	such	such	ADJ
fcis-31421	169	28	as	as	ADP
fcis-31421	169	29	replacing	replace	VERB
fcis-31421	169	30	conv	conv	ADJ
fcis-31421	169	31	with	with	ADP
fcis-31421	169	32	the	the	DET
fcis-31421	169	33	ghostconv	ghostconv	NOUN
fcis-31421	169	34	module	module	NOUN
fcis-31421	169	35	.	.	PUNCT
fcis-31421	170	1	overall	overall	ADJ
fcis-31421	170	2	,	,	PUNCT
fcis-31421	170	3	yolov8	yolov8	NOUN
fcis-31421	170	4	rebalances	rebalance	NOUN
fcis-31421	170	5	accuracy	accuracy	NOUN
fcis-31421	170	6	and	and	CCONJ
fcis-31421	170	7	efficiency	efficiency	NOUN
fcis-31421	170	8	while	while	SCONJ
fcis-31421	170	9	maintaining	maintain	VERB
fcis-31421	170	10	the	the	DET
fcis-31421	170	11	real	real	ADJ
fcis-31421	170	12	-	-	PUNCT
fcis-31421	170	13	time	time	NOUN
fcis-31421	170	14	advantages	advantage	NOUN
fcis-31421	170	15	of	of	ADP
fcis-31421	170	16	the	the	DET
fcis-31421	170	17	yolo	yolo	ADJ
fcis-31421	170	18	series	series	NOUN
fcis-31421	170	19	through	through	ADP
fcis-31421	170	20	architectural	architectural	ADJ
fcis-31421	170	21	lightweighting	lightweighting	ADJ
fcis-31421	170	22	and	and	CCONJ
fcis-31421	170	23	dynamic	dynamic	ADJ
fcis-31421	170	24	training	training	NOUN
fcis-31421	170	25	strategies	strategy	NOUN
fcis-31421	170	26	.	.	PUNCT
fcis-31421	171	1	its	its	PRON
fcis-31421	171	2	performance	performance	NOUN
fcis-31421	171	3	in	in	ADP
fcis-31421	171	4	the	the	DET
fcis-31421	171	5	field	field	NOUN
fcis-31421	171	6	of	of	ADP
fcis-31421	171	7	fire	fire	NOUN
fcis-31421	171	8	detection	detection	NOUN
fcis-31421	171	9	is	be	AUX
fcis-31421	171	10	particularly	particularly	ADV
fcis-31421	171	11	outstanding	outstanding	ADJ
fcis-31421	171	12	,	,	PUNCT
fcis-31421	171	13	considering	consider	VERB
fcis-31421	171	14	the	the	DET
fcis-31421	171	15	needs	need	NOUN
fcis-31421	171	16	of	of	ADP
fcis-31421	171	17	tiny	tiny	ADJ
fcis-31421	171	18	target	target	NOUN
fcis-31421	171	19	capture	capture	NOUN
fcis-31421	171	20	,	,	PUNCT
fcis-31421	171	21	complex	complex	ADJ
fcis-31421	171	22	environment	environment	NOUN
fcis-31421	171	23	adaptability	adaptability	NOUN
fcis-31421	171	24	and	and	CCONJ
fcis-31421	171	25	edge	edge	NOUN
fcis-31421	171	26	device	device	NOUN
fcis-31421	171	27	deployment	deployment	NOUN
fcis-31421	171	28	,	,	PUNCT
fcis-31421	171	29	and	and	CCONJ
fcis-31421	171	30	has	have	VERB
fcis-31421	171	31	a	a	DET
fcis-31421	171	32	greater	great	ADJ
fcis-31421	171	33	advantage	advantage	NOUN
fcis-31421	171	34	in	in	ADP
fcis-31421	171	35	fire	fire	NOUN
fcis-31421	171	36	detection	detection	NOUN
fcis-31421	171	37	.	.	PUNCT
fcis-31421	172	1	table	table	NOUN
fcis-31421	172	2	2	2	NUM
fcis-31421	172	3	.	.	PUNCT
fcis-31421	172	4	performance	performance	NOUN
fcis-31421	172	5	comparison	comparison	NOUN
fcis-31421	172	6	of	of	ADP
fcis-31421	172	7	yolov8	yolov8	NOUN
fcis-31421	172	8	yolov8	yolov8	PROPN
fcis-31421	173	1	yolov5	yolov5	PROPN
fcis-31421	173	2	backbone	backbone	PROPN
fcis-31421	173	3	cspdarknet53	cspdarknet53	PROPN
fcis-31421	173	4	enhanced	enhance	VERB
fcis-31421	173	5	deeper	deep	ADJ
fcis-31421	173	6	structure	structure	NOUN
fcis-31421	173	7	cspdarknet53	cspdarknet53	NOUN
fcis-31421	173	8	head	head	NOUN
fcis-31421	173	9	anchor	anchor	NOUN
fcis-31421	173	10	-	-	PUNCT
fcis-31421	173	11	free	free	ADJ
fcis-31421	173	12	no	no	DET
fcis-31421	173	13	anchor	anchor	NOUN
fcis-31421	173	14	anchor	anchor	NOUN
fcis-31421	173	15	-	-	PUNCT
fcis-31421	173	16	basedbased	basedbase	VERB
fcis-31421	173	17	on	on	ADP
fcis-31421	173	18	predefined	predefine	VERB
fcis-31421	173	19	anchor	anchor	NOUN
fcis-31421	173	20	frames	frame	NOUN
fcis-31421	173	21	neck	neck	VERB
fcis-31421	173	22	pan	pan	ADJ
fcis-31421	173	23	-	-	ADJ
fcis-31421	173	24	fpnmultiscale	fpnmultiscale	ADJ
fcis-31421	173	25	enhancement	enhancement	NOUN
fcis-31421	173	26	pan	pan	ADJ
fcis-31421	173	27	-	-	ADJ
fcis-31421	173	28	fpn	fpn	VERB
fcis-31421	173	29	loss	loss	NOUN
fcis-31421	173	30	function	function	VERB
fcis-31421	173	31	varifocal	varifocal	ADJ
fcis-31421	173	32	loss	loss	NOUN
fcis-31421	173	33	+	+	CCONJ
fcis-31421	173	34	ciou	ciou	NOUN
fcis-31421	173	35	+	+	CCONJ
fcis-31421	173	36	dfl	dfl	PROPN
fcis-31421	173	37	focal	focal	ADJ
fcis-31421	173	38	loss	loss	NOUN
fcis-31421	173	39	+	+	CCONJ
fcis-31421	173	40	ciou	ciou	NOUN
fcis-31421	173	41	coco	coco	PROPN
fcis-31421	173	42	map	map	NOUN
fcis-31421	173	43	53.9	53.9	NUM
fcis-31421	173	44	%	%	NOUN
fcis-31421	173	45	50.7	50.7	NUM
fcis-31421	173	46	%	%	NOUN
fcis-31421	173	47	4	4	NUM
fcis-31421	173	48	.	.	PUNCT
fcis-31421	173	49	experimental	experimental	ADJ
fcis-31421	173	50	design	design	NOUN
fcis-31421	173	51	because	because	SCONJ
fcis-31421	173	52	yolov8	yolov8	NOUN
fcis-31421	173	53	has	have	VERB
fcis-31421	173	54	a	a	DET
fcis-31421	173	55	powerful	powerful	ADJ
fcis-31421	173	56	performance	performance	NOUN
fcis-31421	173	57	,	,	PUNCT
fcis-31421	173	58	scholars	scholar	NOUN
fcis-31421	173	59	can	can	AUX
fcis-31421	173	60	improve	improve	VERB
fcis-31421	173	61	and	and	CCONJ
fcis-31421	173	62	upgrade	upgrade	VERB
fcis-31421	173	63	it	it	PRON
fcis-31421	173	64	on	on	ADP
fcis-31421	173	65	this	this	DET
fcis-31421	173	66	basis	basis	NOUN
fcis-31421	173	67	.	.	PUNCT
fcis-31421	174	1	before	before	ADP
fcis-31421	174	2	that	that	PRON
fcis-31421	174	3	,	,	PUNCT
fcis-31421	174	4	i	i	PRON
fcis-31421	174	5	need	need	VERB
fcis-31421	174	6	to	to	PART
fcis-31421	174	7	understand	understand	VERB
fcis-31421	174	8	the	the	DET
fcis-31421	174	9	architecture	architecture	NOUN
fcis-31421	174	10	of	of	ADP
fcis-31421	174	11	yolov8	yolov8	NOUN
fcis-31421	174	12	(	(	PUNCT
fcis-31421	174	13	figure	figure	NOUN
fcis-31421	174	14	2	2	NUM
fcis-31421	174	15	)	)	PUNCT
fcis-31421	174	16	.	.	PUNCT
fcis-31421	175	1	fig	fig	NOUN
fcis-31421	176	1	2	2	NUM
fcis-31421	176	2	.	.	PUNCT
fcis-31421	176	3	yolo	yolo	ADJ
fcis-31421	176	4	v8	v8	PROPN
fcis-31421	176	5	architecture	architecture	NOUN
fcis-31421	176	6	as	as	SCONJ
fcis-31421	176	7	can	can	AUX
fcis-31421	176	8	be	be	AUX
fcis-31421	176	9	seen	see	VERB
fcis-31421	176	10	from	from	ADP
fcis-31421	176	11	fig	fig	NOUN
fcis-31421	176	12	.	.	PUNCT
fcis-31421	177	1	2	2	NUM
fcis-31421	177	2	,	,	PUNCT
fcis-31421	177	3	yolov8	yolov8	NOUN
fcis-31421	177	4	is	be	AUX
fcis-31421	177	5	mainly	mainly	ADV
fcis-31421	177	6	divided	divide	VERB
fcis-31421	177	7	into	into	ADP
fcis-31421	177	8	two	two	NUM
fcis-31421	177	9	big	big	ADJ
fcis-31421	177	10	parts	part	NOUN
fcis-31421	177	11	,	,	PUNCT
fcis-31421	177	12	namely	namely	ADV
fcis-31421	177	13	backbone	backbone	NOUN
fcis-31421	177	14	and	and	CCONJ
fcis-31421	177	15	head	head	NOUN
fcis-31421	177	16	.	.	PUNCT
fcis-31421	178	1	in	in	ADP
fcis-31421	178	2	backbone	backbone	NOUN
fcis-31421	178	3	,	,	PUNCT
fcis-31421	178	4	the	the	DET
fcis-31421	178	5	core	core	NOUN
fcis-31421	178	6	modules	module	NOUN
fcis-31421	178	7	are	be	AUX
fcis-31421	178	8	mainly	mainly	ADV
fcis-31421	178	9	c2f	c2f	VERB
fcis-31421	178	10	,	,	PUNCT
fcis-31421	178	11	conv	conv	ADJ
fcis-31421	178	12	,	,	PUNCT
fcis-31421	178	13	and	and	CCONJ
fcis-31421	178	14	sppf	sppf	ADJ
fcis-31421	178	15	.	.	PUNCT
fcis-31421	179	1	the	the	DET
fcis-31421	179	2	conv	conv	ADJ
fcis-31421	179	3	module	module	NOUN
fcis-31421	179	4	extracts	extract	NOUN
fcis-31421	179	5	features	feature	NOUN
fcis-31421	179	6	by	by	ADP
fcis-31421	179	7	convolution	convolution	NOUN
fcis-31421	179	8	operation	operation	NOUN
fcis-31421	179	9	and	and	CCONJ
fcis-31421	179	10	integrates	integrate	NOUN
fcis-31421	179	11	the	the	DET
fcis-31421	179	12	batch	batch	NOUN
fcis-31421	179	13	normalization	normalization	NOUN
fcis-31421	179	14	and	and	CCONJ
fcis-31421	179	15	activation	activation	NOUN
fcis-31421	179	16	functions	function	NOUN
fcis-31421	179	17	at	at	ADP
fcis-31421	179	18	the	the	DET
fcis-31421	179	19	same	same	ADJ
fcis-31421	179	20	time	time	NOUN
fcis-31421	179	21	,	,	PUNCT
fcis-31421	179	22	to	to	PART
fcis-31421	179	23	realize	realize	VERB
fcis-31421	179	24	efficient	efficient	ADJ
fcis-31421	179	25	spatial	spatial	ADJ
fcis-31421	179	26	information	information	NOUN
fcis-31421	179	27	encoding	encode	VERB
fcis-31421	179	28	with	with	ADP
fcis-31421	179	29	the	the	DET
fcis-31421	179	30	nonlinear	nonlinear	ADJ
fcis-31421	179	31	feature	feature	NOUN
fcis-31421	179	32	transformation	transformation	NOUN
fcis-31421	179	33	,	,	PUNCT
fcis-31421	179	34	which	which	PRON
fcis-31421	179	35	is	be	AUX
fcis-31421	179	36	the	the	DET
fcis-31421	179	37	basic	basic	ADJ
fcis-31421	179	38	computational	computational	ADJ
fcis-31421	179	39	unit	unit	NOUN
fcis-31421	179	40	of	of	ADP
fcis-31421	179	41	the	the	DET
fcis-31421	179	42	model	model	NOUN
fcis-31421	179	43	.	.	PUNCT
fcis-31421	180	1	the	the	DET
fcis-31421	180	2	c2f	c2f	NOUN
fcis-31421	180	3	module	module	NOUN
fcis-31421	180	4	enhances	enhance	VERB
fcis-31421	180	5	the	the	DET
fcis-31421	180	6	feature	feature	NOUN
fcis-31421	180	7	fusion	fusion	NOUN
fcis-31421	180	8	capability	capability	NOUN
fcis-31421	180	9	while	while	SCONJ
fcis-31421	180	10	reducing	reduce	VERB
fcis-31421	180	11	the	the	DET
fcis-31421	180	12	computation	computation	NOUN
fcis-31421	180	13	volume	volume	NOUN
fcis-31421	180	14	through	through	ADP
fcis-31421	180	15	multi	multi	ADJ
fcis-31421	180	16	-	-	ADJ
fcis-31421	180	17	branch	branch	ADJ
fcis-31421	180	18	gradient	gradient	NOUN
fcis-31421	180	19	flow	flow	NOUN
fcis-31421	180	20	and	and	CCONJ
fcis-31421	180	21	cross	cross	ADJ
fcis-31421	180	22	-	-	ADJ
fcis-31421	180	23	stage	stage	ADJ
fcis-31421	180	24	partial	partial	ADJ
fcis-31421	180	25	connection	connection	NOUN
fcis-31421	180	26	,	,	PUNCT
fcis-31421	180	27	which	which	PRON
fcis-31421	180	28	is	be	AUX
fcis-31421	180	29	the	the	DET
fcis-31421	180	30	core	core	NOUN
fcis-31421	180	31	design	design	NOUN
fcis-31421	180	32	for	for	ADP
fcis-31421	180	33	balancing	balance	VERB
fcis-31421	180	34	lightweight	lightweight	NOUN
fcis-31421	180	35	and	and	CCONJ
fcis-31421	180	36	performance	performance	NOUN
fcis-31421	180	37	in	in	ADP
fcis-31421	180	38	yolov8.the	yolov8.the	DET
fcis-31421	180	39	sppf	sppf	ADJ
fcis-31421	180	40	module	module	NOUN
fcis-31421	180	41	fuses	fuse	VERB
fcis-31421	180	42	the	the	DET
fcis-31421	180	43	features	feature	NOUN
fcis-31421	180	44	of	of	ADP
fcis-31421	180	45	different	different	ADJ
fcis-31421	180	46	sensory	sensory	ADJ
fcis-31421	180	47	fields	field	NOUN
fcis-31421	180	48	through	through	ADP
fcis-31421	180	49	multi	multi	ADJ
fcis-31421	180	50	-	-	ADJ
fcis-31421	180	51	scale	scale	ADJ
fcis-31421	180	52	maximal	maximal	ADJ
fcis-31421	180	53	pooling	pooling	NOUN
fcis-31421	180	54	to	to	PART
fcis-31421	180	55	enhance	enhance	VERB
fcis-31421	180	56	the	the	DET
fcis-31421	180	57	robustness	robustness	NOUN
fcis-31421	180	58	of	of	ADP
fcis-31421	180	59	the	the	DET
fcis-31421	180	60	model	model	NOUN
fcis-31421	180	61	to	to	ADP
fcis-31421	180	62	the	the	DET
fcis-31421	180	63	change	change	NOUN
fcis-31421	180	64	of	of	ADP
fcis-31421	180	65	the	the	DET
fcis-31421	180	66	target	target	NOUN
fcis-31421	180	67	size	size	NOUN
fcis-31421	180	68	while	while	SCONJ
fcis-31421	180	69	maintaining	maintain	VERB
fcis-31421	180	70	a	a	DET
fcis-31421	180	71	high	high	ADJ
fcis-31421	180	72	computational	computational	ADJ
fcis-31421	180	73	efficiency	efficiency	NOUN
fcis-31421	180	74	.	.	PUNCT
fcis-31421	181	1	bottleneck	bottleneck	NOUN
fcis-31421	181	2	.	.	PUNCT
fcis-31421	182	1	detect	detect	NOUN
fcis-31421	182	2	and	and	CCONJ
fcis-31421	182	3	other	other	ADJ
fcis-31421	182	4	modules	module	NOUN
fcis-31421	182	5	such	such	ADJ
fcis-31421	182	6	as	as	ADP
fcis-31421	182	7	detect	detect	NOUN
fcis-31421	182	8	are	be	AUX
fcis-31421	182	9	used	use	VERB
fcis-31421	182	10	for	for	ADP
fcis-31421	182	11	feature	feature	NOUN
fcis-31421	182	12	compression	compression	NOUN
fcis-31421	182	13	/	/	SYM
fcis-31421	182	14	expansion	expansion	NOUN
fcis-31421	182	15	,	,	PUNCT
fcis-31421	182	16	up	up	ADV
fcis-31421	182	17	-	-	PUNCT
fcis-31421	182	18	sampling	sample	VERB
fcis-31421	182	19	to	to	PART
fcis-31421	182	20	recover	recover	VERB
fcis-31421	182	21	resolution	resolution	NOUN
fcis-31421	182	22	,	,	PUNCT
fcis-31421	182	23	and	and	CCONJ
fcis-31421	182	24	multi	multi	ADJ
fcis-31421	182	25	-	-	ADJ
fcis-31421	182	26	scale	scale	ADJ
fcis-31421	182	27	target	target	NOUN
fcis-31421	182	28	detection	detection	NOUN
fcis-31421	182	29	output	output	NOUN
fcis-31421	182	30	,	,	PUNCT
fcis-31421	182	31	respectively	respectively	ADV
fcis-31421	182	32	,	,	PUNCT
fcis-31421	182	33	which	which	PRON
fcis-31421	182	34	work	work	VERB
fcis-31421	182	35	together	together	ADV
fcis-31421	182	36	to	to	PART
fcis-31421	182	37	build	build	VERB
fcis-31421	182	38	the	the	DET
fcis-31421	182	39	complete	complete	ADJ
fcis-31421	182	40	inference	inference	NOUN
fcis-31421	182	41	process	process	NOUN
fcis-31421	182	42	of	of	ADP
fcis-31421	182	43	yolov8	yolov8	PROPN
fcis-31421	182	44	.	.	PUNCT
fcis-31421	183	1	the	the	DET
fcis-31421	183	2	following	follow	VERB
fcis-31421	183	3	image	image	NOUN
fcis-31421	183	4	shows	show	VERB
fcis-31421	183	5	the	the	DET
fcis-31421	183	6	result	result	NOUN
fcis-31421	183	7	graph	graph	NOUN
fcis-31421	183	8	of	of	ADP
fcis-31421	183	9	yolov8	yolov8	NOUN
fcis-31421	183	10	after	after	ADP
fcis-31421	183	11	thirty	thirty	NUM
fcis-31421	183	12	rounds	round	NOUN
fcis-31421	183	13	of	of	ADP
fcis-31421	183	14	training	training	NOUN
fcis-31421	183	15	with	with	ADP
fcis-31421	183	16	9441	9441	NUM
fcis-31421	183	17	fire	fire	NOUN
fcis-31421	183	18	and	and	CCONJ
fcis-31421	183	19	smoke	smoke	VERB
fcis-31421	183	20	related	related	ADJ
fcis-31421	183	21	datasets	dataset	NOUN
fcis-31421	183	22	.	.	PUNCT
fcis-31421	184	1	57	57	NUM
fcis-31421	184	2	fig	fig	NOUN
fcis-31421	184	3	3	3	NUM
fcis-31421	184	4	.	.	PUNCT
fcis-31421	184	5	confusion	confusion	NOUN
fcis-31421	184	6	matrix	matrix	NOUN
fcis-31421	184	7	fig	fig	NOUN
fcis-31421	184	8	4	4	NUM
fcis-31421	184	9	.	.	PUNCT
fcis-31421	184	10	f1	f1	NOUN
fcis-31421	184	11	-	-	PUNCT
fcis-31421	184	12	confidence	confidence	NOUN
fcis-31421	184	13	curve	curve	NOUN
fcis-31421	184	14	fig	fig	NOUN
fcis-31421	184	15	5	5	NUM
fcis-31421	184	16	.	.	PUNCT
fcis-31421	184	17	precision	precision	NOUN
fcis-31421	184	18	-	-	PUNCT
fcis-31421	184	19	confidence	confidence	NOUN
fcis-31421	184	20	curve	curve	NOUN
fcis-31421	184	21	fig	fig	NOUN
fcis-31421	184	22	6	6	NUM
fcis-31421	184	23	.	.	PUNCT
fcis-31421	185	1	recall	recall	NOUN
fcis-31421	185	2	-	-	PUNCT
fcis-31421	185	3	confidence	confidence	NOUN
fcis-31421	185	4	curve	curve	NOUN
fcis-31421	185	5	fig	fig	NOUN
fcis-31421	185	6	7	7	NUM
fcis-31421	185	7	.	.	PUNCT
fcis-31421	185	8	precision	precision	NOUN
fcis-31421	185	9	-	-	PUNCT
fcis-31421	185	10	recall	recall	NOUN
fcis-31421	185	11	curve	curve	NOUN
fcis-31421	185	12	fig	fig	NOUN
fcis-31421	185	13	8	8	NUM
fcis-31421	185	14	.	.	PUNCT
fcis-31421	186	1	result	result	VERB
fcis-31421	186	2	58	58	NUM
fcis-31421	186	3	as	as	SCONJ
fcis-31421	186	4	can	can	AUX
fcis-31421	186	5	be	be	AUX
fcis-31421	186	6	seen	see	VERB
fcis-31421	186	7	from	from	ADP
fcis-31421	186	8	the	the	DET
fcis-31421	186	9	data	datum	NOUN
fcis-31421	186	10	presented	present	VERB
fcis-31421	186	11	in	in	ADP
fcis-31421	186	12	figure	figure	NOUN
fcis-31421	186	13	3	3	NUM
fcis-31421	186	14	,	,	PUNCT
fcis-31421	186	15	the	the	DET
fcis-31421	186	16	precision	precision	NOUN
fcis-31421	186	17	rate	rate	NOUN
fcis-31421	186	18	for	for	ADP
fcis-31421	186	19	detection	detection	NOUN
fcis-31421	186	20	of	of	ADP
fcis-31421	186	21	smoke	smoke	NOUN
fcis-31421	186	22	reaches	reach	VERB
fcis-31421	186	23	78.7	78.7	NUM
fcis-31421	186	24	%	%	NOUN
fcis-31421	186	25	(	(	PUNCT
fcis-31421	186	26	1383/1756	1383/1756	NUM
fcis-31421	186	27	)	)	PUNCT
fcis-31421	186	28	and	and	CCONJ
fcis-31421	186	29	the	the	DET
fcis-31421	186	30	recall	recall	NOUN
fcis-31421	186	31	rate	rate	NOUN
fcis-31421	186	32	is	be	AUX
fcis-31421	186	33	74.7	74.7	NUM
fcis-31421	186	34	%	%	NOUN
fcis-31421	186	35	(	(	PUNCT
fcis-31421	186	36	1383/1851	1383/1851	NUM
fcis-31421	186	37	)	)	PUNCT
fcis-31421	186	38	.	.	PUNCT
fcis-31421	187	1	precision	precision	NOUN
fcis-31421	187	2	for	for	ADP
fcis-31421	187	3	fire	fire	NOUN
fcis-31421	187	4	reaches	reach	VERB
fcis-31421	187	5	67.4	67.4	NUM
fcis-31421	187	6	%	%	NOUN
fcis-31421	187	7	(	(	PUNCT
fcis-31421	187	8	1467/2176	1467/2176	NUM
fcis-31421	187	9	)	)	PUNCT
fcis-31421	187	10	and	and	CCONJ
fcis-31421	187	11	recall	recall	NOUN
fcis-31421	187	12	is	be	AUX
fcis-31421	187	13	66.2	66.2	NUM
fcis-31421	187	14	%	%	NOUN
fcis-31421	187	15	(	(	PUNCT
fcis-31421	187	16	1467/2215	1467/2215	NUM
fcis-31421	187	17	)	)	PUNCT
fcis-31421	187	18	.	.	PUNCT
fcis-31421	188	1	figure	figure	NOUN
fcis-31421	188	2	4	4	NUM
fcis-31421	188	3	shows	show	VERB
fcis-31421	188	4	the	the	DET
fcis-31421	188	5	f1	f1	NOUN
fcis-31421	188	6	-	-	PUNCT
fcis-31421	188	7	confidence	confidence	NOUN
fcis-31421	188	8	curve	curve	NOUN
fcis-31421	188	9	with	with	ADP
fcis-31421	188	10	an	an	DET
fcis-31421	188	11	f1	f1	ADJ
fcis-31421	188	12	value	value	NOUN
fcis-31421	188	13	of	of	ADP
fcis-31421	188	14	0.70	0.70	NUM
fcis-31421	188	15	at	at	ADP
fcis-31421	188	16	a	a	DET
fcis-31421	188	17	confidence	confidence	NOUN
fcis-31421	188	18	threshold	threshold	NOUN
fcis-31421	188	19	of	of	ADP
fcis-31421	188	20	0.288	0.288	NUM
fcis-31421	188	21	.	.	PUNCT
fcis-31421	189	1	figure	figure	NOUN
fcis-31421	189	2	5	5	NUM
fcis-31421	189	3	shows	show	VERB
fcis-31421	189	4	a	a	DET
fcis-31421	189	5	precision	precision	NOUN
fcis-31421	189	6	of	of	ADP
fcis-31421	189	7	1.0	1.0	NUM
fcis-31421	189	8	at	at	ADP
fcis-31421	189	9	a	a	DET
fcis-31421	189	10	confidence	confidence	NOUN
fcis-31421	189	11	level	level	NOUN
fcis-31421	189	12	of	of	ADP
fcis-31421	189	13	0.935	0.935	NUM
fcis-31421	189	14	,	,	PUNCT
fcis-31421	189	15	and	and	CCONJ
fcis-31421	189	16	figure	figure	VERB
fcis-31421	189	17	6	6	NUM
fcis-31421	189	18	shows	show	VERB
fcis-31421	189	19	a	a	DET
fcis-31421	189	20	recall	recall	NOUN
fcis-31421	189	21	of	of	ADP
fcis-31421	189	22	0.93	0.93	NUM
fcis-31421	189	23	at	at	ADP
fcis-31421	189	24	a	a	DET
fcis-31421	189	25	confidence	confidence	NOUN
fcis-31421	189	26	level	level	NOUN
fcis-31421	189	27	of	of	ADP
fcis-31421	189	28	0	0	NUM
fcis-31421	189	29	.	.	PUNCT
fcis-31421	190	1	the	the	DET
fcis-31421	190	2	pr	pr	NOUN
fcis-31421	190	3	curve	curve	NOUN
fcis-31421	190	4	’s	’s	PART
fcis-31421	190	5	map@0.5=0.747	map@0.5=0.747	PROPN
fcis-31421	190	6	indicates	indicate	VERB
fcis-31421	190	7	that	that	SCONJ
fcis-31421	190	8	the	the	DET
fcis-31421	190	9	model	model	NOUN
fcis-31421	190	10	performs	perform	VERB
fcis-31421	190	11	well	well	ADV
fcis-31421	190	12	in	in	ADP
fcis-31421	190	13	fire	fire	NOUN
fcis-31421	190	14	detection	detection	NOUN
fcis-31421	190	15	.	.	PUNCT
fcis-31421	191	1	fig	fig	NOUN
fcis-31421	191	2	.	.	PUNCT
fcis-31421	192	1	7	7	NUM
fcis-31421	192	2	and	and	CCONJ
fcis-31421	192	3	fig	fig	NOUN
fcis-31421	192	4	.	.	PUNCT
fcis-31421	193	1	8	8	NUM
fcis-31421	193	2	also	also	ADV
fcis-31421	193	3	show	show	VERB
fcis-31421	193	4	comprehensive	comprehensive	ADJ
fcis-31421	193	5	experimental	experimental	ADJ
fcis-31421	193	6	results	result	NOUN
fcis-31421	193	7	.	.	PUNCT
fcis-31421	194	1	fig	fig	NOUN
fcis-31421	194	2	9	9	NUM
fcis-31421	194	3	.	.	PUNCT
fcis-31421	195	1	yolov8	yolov8	NOUN
fcis-31421	195	2	improved	improve	VERB
fcis-31421	195	3	model	model	NOUN
fcis-31421	195	4	precision	precision	NOUN
fcis-31421	195	5	recall	recall	NOUN
fcis-31421	195	6	curves	curve	NOUN
fcis-31421	195	7	for	for	ADP
fcis-31421	195	8	smoke	smoke	NOUN
fcis-31421	195	9	and	and	CCONJ
fcis-31421	195	10	flame	flame	NOUN
fcis-31421	195	11	detection	detection	NOUN
fcis-31421	195	12	fig	fig	NOUN
fcis-31421	195	13	10	10	NUM
fcis-31421	195	14	.	.	PUNCT
fcis-31421	196	1	the	the	DET
fcis-31421	196	2	results	result	NOUN
fcis-31421	196	3	of	of	ADP
fcis-31421	196	4	the	the	DET
fcis-31421	196	5	proposed	propose	VERB
fcis-31421	196	6	model	model	NOUN
fcis-31421	196	7	fig	fig	PROPN
fcis-31421	196	8	.	.	PUNCT
fcis-31421	197	1	9	9	NUM
fcis-31421	197	2	shows	show	VERB
fcis-31421	197	3	the	the	DET
fcis-31421	197	4	precision	precision	NOUN
fcis-31421	197	5	recall	recall	NOUN
fcis-31421	197	6	curves	curve	NOUN
fcis-31421	197	7	of	of	ADP
fcis-31421	197	8	the	the	DET
fcis-31421	197	9	improved	improved	ADJ
fcis-31421	197	10	model	model	NOUN
fcis-31421	197	11	for	for	ADP
fcis-31421	197	12	flame	flame	NOUN
fcis-31421	197	13	and	and	CCONJ
fcis-31421	197	14	smoke	smoke	NOUN
fcis-31421	197	15	detection	detection	NOUN
fcis-31421	197	16	,	,	PUNCT
fcis-31421	197	17	fig	fig	NOUN
fcis-31421	197	18	.	.	PUNCT
fcis-31421	198	1	10	10	NUM
fcis-31421	198	2	shows	show	VERB
fcis-31421	198	3	the	the	DET
fcis-31421	198	4	overall	overall	ADJ
fcis-31421	198	5	results	result	NOUN
fcis-31421	198	6	of	of	ADP
fcis-31421	198	7	the	the	DET
fcis-31421	198	8	proposed	propose	VERB
fcis-31421	198	9	model	model	NOUN
fcis-31421	198	10	in	in	ADP
fcis-31421	198	11	terms	term	NOUN
fcis-31421	198	12	of	of	ADP
fcis-31421	198	13	loss	loss	NOUN
fcis-31421	198	14	,	,	PUNCT
fcis-31421	198	15	precision	precision	NOUN
fcis-31421	198	16	and	and	CCONJ
fcis-31421	198	17	recall	recall	NOUN
fcis-31421	198	18	,	,	PUNCT
fcis-31421	198	19	and	and	CCONJ
fcis-31421	198	20	fig	fig	NOUN
fcis-31421	198	21	.	.	PUNCT
fcis-31421	199	1	11	11	NUM
fcis-31421	199	2	shows	show	VERB
fcis-31421	199	3	the	the	DET
fcis-31421	199	4	different	different	ADJ
fcis-31421	199	5	weights	weight	NOUN
fcis-31421	199	6	of	of	ADP
fcis-31421	199	7	the	the	DET
fcis-31421	199	8	yolov8	yolov8	NOUN
fcis-31421	199	9	model	model	NOUN
fcis-31421	199	10	using	use	VERB
fcis-31421	199	11	mcdh	mcdh	NOUN
fcis-31421	199	12	.	.	PUNCT
fcis-31421	200	1	these	these	DET
fcis-31421	200	2	images	image	NOUN
fcis-31421	200	3	show	show	VERB
fcis-31421	200	4	that	that	SCONJ
fcis-31421	200	5	the	the	DET
fcis-31421	200	6	improvements	improvement	NOUN
fcis-31421	200	7	made	make	VERB
fcis-31421	200	8	to	to	ADP
fcis-31421	200	9	the	the	DET
fcis-31421	200	10	yolov8	yolov8	NOUN
fcis-31421	200	11	model	model	NOUN
fcis-31421	200	12	have	have	VERB
fcis-31421	200	13	a	a	DET
fcis-31421	200	14	high	high	ADJ
fcis-31421	200	15	degree	degree	NOUN
fcis-31421	200	16	of	of	ADP
fcis-31421	200	17	precision	precision	NOUN
fcis-31421	200	18	while	while	SCONJ
fcis-31421	200	19	keeping	keep	VERB
fcis-31421	200	20	it	it	PRON
fcis-31421	200	21	lightweight	lightweight	ADJ
fcis-31421	200	22	.	.	PUNCT
fcis-31421	201	1	yolov8	yolov8	NOUN
fcis-31421	201	2	has	have	AUX
fcis-31421	201	3	become	become	VERB
fcis-31421	201	4	the	the	DET
fcis-31421	201	5	model	model	NOUN
fcis-31421	201	6	of	of	ADP
fcis-31421	201	7	choice	choice	NOUN
fcis-31421	201	8	for	for	ADP
fcis-31421	201	9	real	real	ADJ
fcis-31421	201	10	-	-	PUNCT
fcis-31421	201	11	time	time	NOUN
fcis-31421	201	12	target	target	NOUN
fcis-31421	201	13	detection	detection	NOUN
fcis-31421	201	14	due	due	ADP
fcis-31421	201	15	to	to	ADP
fcis-31421	201	16	its	its	PRON
fcis-31421	201	17	efficient	efficient	ADJ
fcis-31421	201	18	inference	inference	NOUN
fcis-31421	201	19	speed	speed	NOUN
fcis-31421	201	20	,	,	PUNCT
fcis-31421	201	21	multi	multi	ADJ
fcis-31421	201	22	-	-	ADJ
fcis-31421	201	23	task	task	ADJ
fcis-31421	201	24	compatibility	compatibility	NOUN
fcis-31421	201	25	and	and	CCONJ
fcis-31421	201	26	lightweight	lightweight	ADJ
fcis-31421	201	27	design	design	NOUN
fcis-31421	201	28	.	.	PUNCT
fcis-31421	202	1	however	however	ADV
fcis-31421	202	2	,	,	PUNCT
fcis-31421	202	3	it	it	PRON
fcis-31421	202	4	still	still	ADV
fcis-31421	202	5	has	have	VERB
fcis-31421	202	6	room	room	NOUN
fcis-31421	202	7	for	for	ADP
fcis-31421	202	8	improvement	improvement	NOUN
fcis-31421	202	9	in	in	ADP
fcis-31421	202	10	small	small	ADJ
fcis-31421	202	11	target	target	NOUN
fcis-31421	202	12	detection	detection	NOUN
fcis-31421	202	13	,	,	PUNCT
fcis-31421	202	14	complex	complex	ADJ
fcis-31421	202	15	background	background	NOUN
fcis-31421	202	16	robustness	robustness	NOUN
fcis-31421	202	17	and	and	CCONJ
fcis-31421	202	18	resource	resource	NOUN
fcis-31421	202	19	consumption	consumption	NOUN
fcis-31421	202	20	.	.	PUNCT
fcis-31421	203	1	in	in	ADP
fcis-31421	203	2	the	the	DET
fcis-31421	203	3	59	59	NUM
fcis-31421	203	4	future	future	NOUN
fcis-31421	203	5	,	,	PUNCT
fcis-31421	203	6	the	the	DET
fcis-31421	203	7	performance	performance	NOUN
fcis-31421	203	8	can	can	AUX
fcis-31421	203	9	be	be	AUX
fcis-31421	203	10	further	far	ADV
fcis-31421	203	11	improved	improve	VERB
fcis-31421	203	12	by	by	ADP
fcis-31421	203	13	strategies	strategy	NOUN
fcis-31421	203	14	such	such	ADJ
fcis-31421	203	15	as	as	ADP
fcis-31421	203	16	attention	attention	NOUN
fcis-31421	203	17	mechanism	mechanism	NOUN
fcis-31421	203	18	enhancement	enhancement	NOUN
fcis-31421	203	19	,	,	PUNCT
fcis-31421	203	20	multimodal	multimodal	ADJ
fcis-31421	203	21	fusion	fusion	NOUN
fcis-31421	203	22	and	and	CCONJ
fcis-31421	203	23	quantization	quantization	NOUN
fcis-31421	203	24	compression	compression	NOUN
fcis-31421	203	25	,	,	PUNCT
fcis-31421	203	26	such	such	ADJ
fcis-31421	203	27	as	as	ADP
fcis-31421	203	28	improving	improve	VERB
fcis-31421	203	29	the	the	DET
fcis-31421	203	30	model	model	NOUN
fcis-31421	203	31	from	from	ADP
fcis-31421	203	32	the	the	DET
fcis-31421	203	33	following	follow	VERB
fcis-31421	203	34	aspects	aspect	NOUN
fcis-31421	203	35	:	:	PUNCT
fcis-31421	203	36	fig	fig	NOUN
fcis-31421	203	37	11	11	NUM
fcis-31421	203	38	.	.	PUNCT
fcis-31421	204	1	the	the	DET
fcis-31421	204	2	results	result	NOUN
fcis-31421	204	3	of	of	ADP
fcis-31421	204	4	yolov8	yolov8	NOUN
fcis-31421	204	5	with	with	ADP
fcis-31421	204	6	different	different	ADJ
fcis-31421	204	7	weight	weight	NOUN
fcis-31421	204	8	coefficients	coefficient	NOUN
fcis-31421	204	9	when	when	SCONJ
fcis-31421	204	10	using	use	VERB
fcis-31421	204	11	mcdh	mcdh	ADJ
fcis-31421	204	12	deab	deab	ADJ
fcis-31421	204	13	module	module	NOUN
fcis-31421	204	14	+	+	CCONJ
fcis-31421	204	15	super	super	ADJ
fcis-31421	204	16	-	-	ADJ
fcis-31421	204	17	resolution	resolution	ADJ
fcis-31421	204	18	preprocessing	preprocessing	NOUN
fcis-31421	204	19	can	can	AUX
fcis-31421	204	20	improve	improve	VERB
fcis-31421	204	21	in	in	ADP
fcis-31421	204	22	small	small	ADJ
fcis-31421	204	23	target	target	NOUN
fcis-31421	204	24	detection	detection	NOUN
fcis-31421	204	25	,	,	PUNCT
fcis-31421	204	26	the	the	DET
fcis-31421	204	27	expected	expect	VERB
fcis-31421	204	28	boost	boost	NOUN
fcis-31421	204	29	boosts	boost	NOUN
fcis-31421	204	30	map@0.5	map@0.5	VERB
fcis-31421	204	31	by	by	ADP
fcis-31421	204	32	10	10	NUM
fcis-31421	204	33	-	-	SYM
fcis-31421	204	34	15	15	NUM
fcis-31421	204	35	%	%	NOUN
fcis-31421	204	36	.	.	PUNCT
fcis-31421	205	1	improvement	improvement	NOUN
fcis-31421	205	2	of	of	ADP
fcis-31421	205	3	high	high	ADJ
fcis-31421	205	4	background	background	NOUN
fcis-31421	205	5	false	false	ADJ
fcis-31421	205	6	detection	detection	NOUN
fcis-31421	205	7	rate	rate	NOUN
fcis-31421	205	8	can	can	AUX
fcis-31421	205	9	be	be	AUX
fcis-31421	205	10	achieved	achieve	VERB
fcis-31421	205	11	by	by	ADP
fcis-31421	205	12	using	use	VERB
fcis-31421	205	13	cbam	cbam	NOUN
fcis-31421	205	14	attention	attention	NOUN
fcis-31421	205	15	+	+	CCONJ
fcis-31421	205	16	dynamic	dynamic	ADJ
fcis-31421	205	17	label	label	NOUN
fcis-31421	205	18	assignment	assignment	NOUN
fcis-31421	205	19	,	,	PUNCT
fcis-31421	205	20	which	which	PRON
fcis-31421	205	21	can	can	AUX
fcis-31421	205	22	reduce	reduce	VERB
fcis-31421	205	23	more	more	ADJ
fcis-31421	205	24	than	than	ADP
fcis-31421	205	25	20	20	NUM
fcis-31421	205	26	%	%	NOUN
fcis-31421	205	27	based	base	VERB
fcis-31421	205	28	on	on	ADP
fcis-31421	205	29	the	the	DET
fcis-31421	205	30	original	original	ADJ
fcis-31421	205	31	false	false	ADJ
fcis-31421	205	32	detection	detection	NOUN
fcis-31421	205	33	rate	rate	NOUN
fcis-31421	205	34	.	.	PUNCT
fcis-31421	206	1	to	to	PART
fcis-31421	206	2	be	be	AUX
fcis-31421	206	3	widely	widely	ADV
fcis-31421	206	4	used	use	VERB
fcis-31421	206	5	in	in	ADP
fcis-31421	206	6	fire	fire	NOUN
fcis-31421	206	7	detection	detection	NOUN
fcis-31421	206	8	,	,	PUNCT
fcis-31421	206	9	the	the	DET
fcis-31421	206	10	model	model	NOUN
fcis-31421	206	11	is	be	AUX
fcis-31421	206	12	very	very	ADV
fcis-31421	206	13	necessary	necessary	ADJ
fcis-31421	206	14	to	to	PART
fcis-31421	206	15	achieve	achieve	VERB
fcis-31421	206	16	lightweight	lightweight	NOUN
fcis-31421	206	17	,	,	PUNCT
fcis-31421	206	18	in	in	ADP
fcis-31421	206	19	yolov8	yolov8	NOUN
fcis-31421	206	20	can	can	AUX
fcis-31421	206	21	be	be	AUX
fcis-31421	206	22	used	use	VERB
fcis-31421	206	23	to	to	ADP
fcis-31421	206	24	the	the	DET
fcis-31421	206	25	ghost	ghost	NOUN
fcis-31421	206	26	convolution	convolution	NOUN
fcis-31421	206	27	,	,	PUNCT
fcis-31421	206	28	adding	add	VERB
fcis-31421	206	29	cbam	cbam	NOUN
fcis-31421	206	30	module	module	NOUN
fcis-31421	206	31	to	to	PART
fcis-31421	206	32	achieve	achieve	VERB
fcis-31421	206	33	a	a	DET
fcis-31421	206	34	reduction	reduction	NOUN
fcis-31421	206	35	in	in	ADP
fcis-31421	206	36	the	the	DET
fcis-31421	206	37	number	number	NOUN
fcis-31421	206	38	of	of	ADP
fcis-31421	206	39	parameters	parameter	NOUN
fcis-31421	206	40	and	and	CCONJ
fcis-31421	206	41	calculations	calculation	NOUN
fcis-31421	206	42	at	at	ADP
fcis-31421	206	43	the	the	DET
fcis-31421	206	44	same	same	ADJ
fcis-31421	206	45	time	time	NOUN
fcis-31421	206	46	to	to	PART
fcis-31421	206	47	reduce	reduce	VERB
fcis-31421	206	48	the	the	DET
fcis-31421	206	49	false	false	ADJ
fcis-31421	206	50	detection	detection	NOUN
fcis-31421	206	51	rate	rate	NOUN
fcis-31421	206	52	.	.	PUNCT
fcis-31421	207	1	5	5	X
fcis-31421	207	2	.	.	X
fcis-31421	207	3	conclusion	conclusion	NOUN
fcis-31421	207	4	in	in	ADP
fcis-31421	207	5	the	the	DET
fcis-31421	207	6	field	field	NOUN
fcis-31421	207	7	of	of	ADP
fcis-31421	207	8	fire	fire	NOUN
fcis-31421	207	9	detection	detection	NOUN
fcis-31421	207	10	,	,	PUNCT
fcis-31421	207	11	yolov8	yolov8	PROPN
fcis-31421	207	12	is	be	AUX
fcis-31421	207	13	ideal	ideal	ADJ
fcis-31421	207	14	for	for	ADP
fcis-31421	207	15	intelligent	intelligent	ADJ
fcis-31421	207	16	early	early	ADJ
fcis-31421	207	17	warning	warning	NOUN
fcis-31421	207	18	systems	system	NOUN
fcis-31421	207	19	due	due	ADP
fcis-31421	207	20	to	to	ADP
fcis-31421	207	21	its	its	PRON
fcis-31421	207	22	excellent	excellent	ADJ
fcis-31421	207	23	real	real	ADJ
fcis-31421	207	24	-	-	PUNCT
fcis-31421	207	25	time	time	NOUN
fcis-31421	207	26	performance	performance	NOUN
fcis-31421	207	27	and	and	CCONJ
fcis-31421	207	28	powerful	powerful	ADJ
fcis-31421	207	29	multi	multi	ADJ
fcis-31421	207	30	-	-	ADJ
fcis-31421	207	31	scale	scale	ADJ
fcis-31421	207	32	feature	feature	NOUN
fcis-31421	207	33	extraction	extraction	NOUN
fcis-31421	207	34	capabilities	capability	NOUN
fcis-31421	207	35	.	.	PUNCT
fcis-31421	208	1	the	the	DET
fcis-31421	208	2	model	model	NOUN
fcis-31421	208	3	can	can	AUX
fcis-31421	208	4	accurately	accurately	ADV
fcis-31421	208	5	capture	capture	VERB
fcis-31421	208	6	tiny	tiny	ADJ
fcis-31421	208	7	flame	flame	NOUN
fcis-31421	208	8	features	feature	NOUN
fcis-31421	208	9	and	and	CCONJ
fcis-31421	208	10	dynamic	dynamic	ADJ
fcis-31421	208	11	smoke	smoke	NOUN
fcis-31421	208	12	targets	target	NOUN
fcis-31421	208	13	in	in	ADP
fcis-31421	208	14	early	early	ADJ
fcis-31421	208	15	-	-	PUNCT
fcis-31421	208	16	stage	stage	NOUN
fcis-31421	208	17	fires	fire	NOUN
fcis-31421	208	18	,	,	PUNCT
fcis-31421	208	19	and	and	CCONJ
fcis-31421	208	20	its	its	PRON
fcis-31421	208	21	lightweight	lightweight	ADJ
fcis-31421	208	22	design	design	NOUN
fcis-31421	208	23	can	can	AUX
fcis-31421	208	24	be	be	AUX
fcis-31421	208	25	easily	easily	ADV
fcis-31421	208	26	deployed	deploy	VERB
fcis-31421	208	27	on	on	ADP
fcis-31421	208	28	edge	edge	NOUN
fcis-31421	208	29	devices	device	NOUN
fcis-31421	208	30	such	such	ADJ
fcis-31421	208	31	as	as	ADP
fcis-31421	208	32	drones	drone	NOUN
fcis-31421	208	33	and	and	CCONJ
fcis-31421	208	34	surveillance	surveillance	NOUN
fcis-31421	208	35	cameras	camera	NOUN
fcis-31421	208	36	to	to	PART
fcis-31421	208	37	achieve	achieve	VERB
fcis-31421	208	38	millisecond	millisecond	NOUN
fcis-31421	208	39	response	response	NOUN
fcis-31421	208	40	speed	speed	NOUN
fcis-31421	208	41	.	.	PUNCT
fcis-31421	209	1	compared	compare	VERB
fcis-31421	209	2	to	to	ADP
fcis-31421	209	3	traditional	traditional	ADJ
fcis-31421	209	4	detection	detection	NOUN
fcis-31421	209	5	methods	method	NOUN
fcis-31421	209	6	,	,	PUNCT
fcis-31421	209	7	yolov8	yolov8	NOUN
fcis-31421	209	8	performs	perform	VERB
fcis-31421	209	9	well	well	ADV
fcis-31421	209	10	in	in	ADP
fcis-31421	209	11	complex	complex	ADJ
fcis-31421	209	12	environments	environment	NOUN
fcis-31421	209	13	,	,	PUNCT
fcis-31421	209	14	effectively	effectively	ADV
fcis-31421	209	15	overcoming	overcome	VERB
fcis-31421	209	16	light	light	ADJ
fcis-31421	209	17	variations	variation	NOUN
fcis-31421	209	18	and	and	CCONJ
fcis-31421	209	19	dust	dust	NOUN
fcis-31421	209	20	interference	interference	NOUN
fcis-31421	209	21	,	,	PUNCT
fcis-31421	209	22	and	and	CCONJ
fcis-31421	209	23	significantly	significantly	ADV
fcis-31421	209	24	reducing	reduce	VERB
fcis-31421	209	25	the	the	DET
fcis-31421	209	26	false	false	ADJ
fcis-31421	209	27	alarm	alarm	NOUN
fcis-31421	209	28	rate	rate	NOUN
fcis-31421	209	29	while	while	SCONJ
fcis-31421	209	30	ensuring	ensure	VERB
fcis-31421	209	31	high	high	ADJ
fcis-31421	209	32	accuracy.yolov8	accuracy.yolov8	NOUN
fcis-31421	209	33	's	's	PART
fcis-31421	209	34	technological	technological	ADJ
fcis-31421	209	35	advantage	advantage	NOUN
fcis-31421	209	36	stems	stem	VERB
fcis-31421	209	37	from	from	ADP
fcis-31421	209	38	three	three	NUM
fcis-31421	209	39	major	major	ADJ
fcis-31421	209	40	innovations	innovation	NOUN
fcis-31421	209	41	:	:	PUNCT
fcis-31421	209	42	an	an	DET
fcis-31421	209	43	improved	improved	ADJ
fcis-31421	209	44	backbone	backbone	NOUN
fcis-31421	209	45	network	network	NOUN
fcis-31421	209	46	structure	structure	NOUN
fcis-31421	209	47	enhances	enhance	VERB
fcis-31421	209	48	feature	feature	NOUN
fcis-31421	209	49	extraction	extraction	NOUN
fcis-31421	209	50	,	,	PUNCT
fcis-31421	209	51	an	an	DET
fcis-31421	209	52	anchorless	anchorless	ADJ
fcis-31421	209	53	detection	detection	NOUN
fcis-31421	209	54	mechanism	mechanism	NOUN
fcis-31421	209	55	simplifies	simplifie	NOUN
fcis-31421	209	56	model	model	VERB
fcis-31421	209	57	complexity	complexity	NOUN
fcis-31421	209	58	,	,	PUNCT
fcis-31421	209	59	and	and	CCONJ
fcis-31421	209	60	a	a	DET
fcis-31421	209	61	dynamic	dynamic	ADJ
fcis-31421	209	62	label	label	NOUN
fcis-31421	209	63	assignment	assignment	NOUN
fcis-31421	209	64	strategy	strategy	NOUN
fcis-31421	209	65	improves	improve	VERB
fcis-31421	209	66	scenario	scenario	NOUN
fcis-31421	209	67	adaptability	adaptability	NOUN
fcis-31421	209	68	.	.	PUNCT
fcis-31421	210	1	these	these	DET
fcis-31421	210	2	features	feature	NOUN
fcis-31421	210	3	enable	enable	VERB
fcis-31421	210	4	it	it	PRON
fcis-31421	210	5	to	to	PART
fcis-31421	210	6	demonstrate	demonstrate	VERB
fcis-31421	210	7	performance	performance	NOUN
fcis-31421	210	8	beyond	beyond	ADP
fcis-31421	210	9	its	its	PRON
fcis-31421	210	10	predecessor	predecessor	NOUN
fcis-31421	210	11	model	model	NOUN
fcis-31421	210	12	in	in	ADP
fcis-31421	210	13	flame	flame	NOUN
fcis-31421	210	14	detection	detection	NOUN
fcis-31421	210	15	tasks	task	NOUN
fcis-31421	210	16	.	.	PUNCT
fcis-31421	211	1	looking	look	VERB
fcis-31421	211	2	towards	towards	ADP
fcis-31421	211	3	the	the	DET
fcis-31421	211	4	future	future	NOUN
fcis-31421	211	5	,	,	PUNCT
fcis-31421	211	6	there	there	PRON
fcis-31421	211	7	is	be	VERB
fcis-31421	211	8	still	still	ADV
fcis-31421	211	9	room	room	NOUN
fcis-31421	211	10	for	for	ADP
fcis-31421	211	11	optimization	optimization	NOUN
fcis-31421	211	12	of	of	ADP
fcis-31421	211	13	the	the	DET
fcis-31421	211	14	yolov8	yolov8	NOUN
fcis-31421	211	15	-	-	PUNCT
fcis-31421	211	16	based	base	VERB
fcis-31421	211	17	fire	fire	NOUN
fcis-31421	211	18	detection	detection	NOUN
fcis-31421	211	19	system	system	NOUN
fcis-31421	211	20	.	.	PUNCT
fcis-31421	212	1	multimodal	multimodal	NOUN
fcis-31421	212	2	data	datum	NOUN
fcis-31421	212	3	fusion	fusion	NOUN
fcis-31421	212	4	technology	technology	NOUN
fcis-31421	212	5	can	can	AUX
fcis-31421	212	6	combine	combine	VERB
fcis-31421	212	7	visible	visible	ADJ
fcis-31421	212	8	and	and	CCONJ
fcis-31421	212	9	infrared	infrared	ADJ
fcis-31421	212	10	images	image	NOUN
fcis-31421	212	11	to	to	PART
fcis-31421	212	12	build	build	VERB
fcis-31421	212	13	a	a	DET
fcis-31421	212	14	more	more	ADV
fcis-31421	212	15	comprehensive	comprehensive	ADJ
fcis-31421	212	16	fire	fire	NOUN
fcis-31421	212	17	characterization	characterization	NOUN
fcis-31421	212	18	system	system	NOUN
fcis-31421	212	19	.	.	PUNCT
fcis-31421	213	1	at	at	ADP
fcis-31421	213	2	the	the	DET
fcis-31421	213	3	same	same	ADJ
fcis-31421	213	4	time	time	NOUN
fcis-31421	213	5	,	,	PUNCT
fcis-31421	213	6	it	it	PRON
fcis-31421	213	7	is	be	AUX
fcis-31421	213	8	necessary	necessary	ADJ
fcis-31421	213	9	to	to	PART
fcis-31421	213	10	improve	improve	VERB
fcis-31421	213	11	the	the	DET
fcis-31421	213	12	generalization	generalization	NOUN
fcis-31421	213	13	ability	ability	NOUN
fcis-31421	213	14	of	of	ADP
fcis-31421	213	15	the	the	DET
fcis-31421	213	16	model	model	NOUN
fcis-31421	213	17	in	in	ADP
fcis-31421	213	18	extreme	extreme	ADJ
fcis-31421	213	19	scenarios	scenario	NOUN
fcis-31421	213	20	such	such	ADJ
fcis-31421	213	21	as	as	ADP
fcis-31421	213	22	smoke	smoke	NOUN
fcis-31421	213	23	obscuration	obscuration	NOUN
fcis-31421	213	24	and	and	CCONJ
fcis-31421	213	25	develop	develop	VERB
fcis-31421	213	26	adaptive	adaptive	ADJ
fcis-31421	213	27	optimization	optimization	NOUN
fcis-31421	213	28	algorithms	algorithm	NOUN
fcis-31421	213	29	to	to	PART
fcis-31421	213	30	adapt	adapt	VERB
fcis-31421	213	31	to	to	ADP
fcis-31421	213	32	different	different	ADJ
fcis-31421	213	33	environments	environment	NOUN
fcis-31421	213	34	.	.	PUNCT
fcis-31421	214	1	from	from	ADP
fcis-31421	214	2	the	the	DET
fcis-31421	214	3	application	application	NOUN
fcis-31421	214	4	point	point	NOUN
fcis-31421	214	5	of	of	ADP
fcis-31421	214	6	view	view	NOUN
fcis-31421	214	7	,	,	PUNCT
fcis-31421	214	8	large	large	ADJ
fcis-31421	214	9	-	-	PUNCT
fcis-31421	214	10	scale	scale	NOUN
fcis-31421	214	11	deployment	deployment	NOUN
fcis-31421	214	12	of	of	ADP
fcis-31421	214	13	yolov8	yolov8	NOUN
fcis-31421	214	14	requires	require	VERB
fcis-31421	214	15	the	the	DET
fcis-31421	214	16	establishment	establishment	NOUN
fcis-31421	214	17	of	of	ADP
fcis-31421	214	18	a	a	DET
fcis-31421	214	19	complete	complete	ADJ
fcis-31421	214	20	solution	solution	NOUN
fcis-31421	214	21	,	,	PUNCT
fcis-31421	214	22	including	include	VERB
fcis-31421	214	23	model	model	NOUN
fcis-31421	214	24	compression	compression	NOUN
fcis-31421	214	25	technology	technology	NOUN
fcis-31421	214	26	,	,	PUNCT
fcis-31421	214	27	efficient	efficient	ADJ
fcis-31421	214	28	data	datum	NOUN
fcis-31421	214	29	preprocessing	preprocessing	NOUN
fcis-31421	214	30	process	process	NOUN
fcis-31421	214	31	,	,	PUNCT
fcis-31421	214	32	and	and	CCONJ
fcis-31421	214	33	intelligent	intelligent	ADJ
fcis-31421	214	34	warning	warning	NOUN
fcis-31421	214	35	and	and	CCONJ
fcis-31421	214	36	decisionmaking	decisionmake	VERB
fcis-31421	214	37	system	system	NOUN
fcis-31421	214	38	.	.	PUNCT
fcis-31421	215	1	its	its	PRON
fcis-31421	215	2	modular	modular	ADJ
fcis-31421	215	3	design	design	NOUN
fcis-31421	215	4	provides	provide	VERB
fcis-31421	215	5	a	a	DET
fcis-31421	215	6	good	good	ADJ
fcis-31421	215	7	foundation	foundation	NOUN
fcis-31421	215	8	for	for	ADP
fcis-31421	215	9	function	function	NOUN
fcis-31421	215	10	expansion	expansion	NOUN
fcis-31421	215	11	.	.	PUNCT
fcis-31421	216	1	with	with	ADP
fcis-31421	216	2	the	the	DET
fcis-31421	216	3	popularization	popularization	NOUN
fcis-31421	216	4	of	of	ADP
fcis-31421	216	5	edge	edge	NOUN
fcis-31421	216	6	computing	computing	NOUN
fcis-31421	216	7	devices	device	NOUN
fcis-31421	216	8	,	,	PUNCT
fcis-31421	216	9	technology	technology	NOUN
fcis-31421	216	10	is	be	AUX
fcis-31421	216	11	expected	expect	VERB
fcis-31421	216	12	to	to	PART
fcis-31421	216	13	achieve	achieve	VERB
fcis-31421	216	14	large	large	ADJ
fcis-31421	216	15	-	-	PUNCT
fcis-31421	216	16	scale	scale	NOUN
fcis-31421	216	17	applications	application	NOUN
fcis-31421	216	18	in	in	ADP
fcis-31421	216	19	industrial	industrial	ADJ
fcis-31421	216	20	safety	safety	NOUN
fcis-31421	216	21	,	,	PUNCT
fcis-31421	216	22	forest	forest	NOUN
fcis-31421	216	23	fire	fire	NOUN
fcis-31421	216	24	prevention	prevention	NOUN
fcis-31421	216	25	and	and	CCONJ
fcis-31421	216	26	other	other	ADJ
fcis-31421	216	27	fields	field	NOUN
fcis-31421	216	28	,	,	PUNCT
fcis-31421	216	29	making	make	VERB
fcis-31421	216	30	an	an	DET
fcis-31421	216	31	important	important	ADJ
fcis-31421	216	32	contribution	contribution	NOUN
fcis-31421	216	33	to	to	ADP
fcis-31421	216	34	improving	improve	VERB
fcis-31421	216	35	public	public	ADJ
fcis-31421	216	36	safety	safety	NOUN
fcis-31421	216	37	.	.	PUNCT
fcis-31421	217	1	the	the	DET
fcis-31421	217	2	technology	technology	NOUN
fcis-31421	217	3	iteration	iteration	NOUN
fcis-31421	217	4	based	base	VERB
fcis-31421	217	5	on	on	ADP
fcis-31421	217	6	deep	deep	ADJ
fcis-31421	217	7	learning	learning	NOUN
fcis-31421	217	8	will	will	AUX
fcis-31421	217	9	continue	continue	VERB
fcis-31421	217	10	to	to	PART
fcis-31421	217	11	promote	promote	VERB
fcis-31421	217	12	the	the	DET
fcis-31421	217	13	development	development	NOUN
fcis-31421	217	14	and	and	CCONJ
fcis-31421	217	15	progress	progress	NOUN
fcis-31421	217	16	in	in	ADP
fcis-31421	217	17	the	the	DET
fcis-31421	217	18	field	field	NOUN
fcis-31421	217	19	of	of	ADP
fcis-31421	217	20	fire	fire	NOUN
fcis-31421	217	21	detection	detection	NOUN
fcis-31421	217	22	.	.	PUNCT
fcis-31421	218	1	references	reference	NOUN
fcis-31421	218	2	[	[	X
fcis-31421	218	3	1	1	NUM
fcis-31421	218	4	]	]	X
fcis-31421	218	5	muhammad	muhammad	PROPN
fcis-31421	218	6	,	,	PUNCT
fcis-31421	218	7	k.	k.	PROPN
fcis-31421	218	8	,	,	PUNCT
fcis-31421	218	9	ahmad	ahmad	PROPN
fcis-31421	218	10	,	,	PUNCT
fcis-31421	218	11	j.	j.	PROPN
fcis-31421	218	12	,	,	PUNCT
fcis-31421	218	13	lv	lv	PROPN
fcis-31421	218	14	,	,	PUNCT
fcis-31421	218	15	z.	z.	PROPN
fcis-31421	218	16	,	,	PUNCT
fcis-31421	218	17	bellavista	bellavista	PROPN
fcis-31421	218	18	,	,	PUNCT
fcis-31421	218	19	p.	p.	PROPN
fcis-31421	218	20	,	,	PUNCT
fcis-31421	218	21	yang	yang	PROPN
fcis-31421	218	22	,	,	PUNCT
fcis-31421	218	23	p.	p.	PROPN
fcis-31421	218	24	,	,	PUNCT
fcis-31421	218	25	&	&	CCONJ
fcis-31421	218	26	baik	baik	PROPN
fcis-31421	218	27	,	,	PUNCT
fcis-31421	218	28	s.	s.	PROPN
fcis-31421	218	29	w.	w.	PROPN
fcis-31421	218	30	(	(	PUNCT
fcis-31421	218	31	2018	2018	NUM
fcis-31421	218	32	)	)	PUNCT
fcis-31421	218	33	.	.	PUNCT
fcis-31421	219	1	efficient	efficient	ADJ
fcis-31421	219	2	deep	deep	ADJ
fcis-31421	219	3	cnn	cnn	PROPN
fcis-31421	219	4	-	-	PUNCT
fcis-31421	219	5	based	base	VERB
fcis-31421	219	6	fire	fire	NOUN
fcis-31421	219	7	detection	detection	NOUN
fcis-31421	219	8	and	and	CCONJ
fcis-31421	219	9	localization	localization	NOUN
fcis-31421	219	10	in	in	ADP
fcis-31421	219	11	video	video	NOUN
fcis-31421	219	12	surveillance	surveillance	NOUN
fcis-31421	219	13	applications	application	NOUN
fcis-31421	219	14	.	.	PUNCT
fcis-31421	220	1	ieee	ieee	NOUN
fcis-31421	220	2	transactions	transaction	NOUN
fcis-31421	220	3	on	on	ADP
fcis-31421	220	4	systems	system	NOUN
fcis-31421	220	5	,	,	PUNCT
fcis-31421	220	6	man	man	NOUN
fcis-31421	220	7	,	,	PUNCT
fcis-31421	220	8	and	and	CCONJ
fcis-31421	220	9	cybernetics	cybernetic	NOUN
fcis-31421	220	10	:	:	PUNCT
fcis-31421	220	11	systems	system	NOUN
fcis-31421	220	12	,	,	PUNCT
fcis-31421	220	13	49(7	49(7	NUM
fcis-31421	220	14	)	)	PUNCT
fcis-31421	220	15	,	,	PUNCT
fcis-31421	220	16	1419	1419	NUM
fcis-31421	220	17	-	-	SYM
fcis-31421	220	18	1434	1434	NUM
fcis-31421	220	19	.	.	PUNCT
fcis-31421	221	1	[	[	X
fcis-31421	221	2	2	2	NUM
fcis-31421	221	3	]	]	X
fcis-31421	221	4	muhammad	muhammad	PROPN
fcis-31421	221	5	,	,	PUNCT
fcis-31421	221	6	k.	k.	PROPN
fcis-31421	221	7	,	,	PUNCT
fcis-31421	221	8	ahmad	ahmad	PROPN
fcis-31421	221	9	,	,	PUNCT
fcis-31421	221	10	j.	j.	PROPN
fcis-31421	221	11	,	,	PUNCT
fcis-31421	221	12	mehmood	mehmood	PROPN
fcis-31421	221	13	,	,	PUNCT
fcis-31421	221	14	i.	i.	PROPN
fcis-31421	221	15	,	,	PUNCT
fcis-31421	221	16	rho	rho	PROPN
fcis-31421	221	17	,	,	PUNCT
fcis-31421	221	18	s.	s.	PROPN
fcis-31421	221	19	,	,	PUNCT
fcis-31421	221	20	&	&	CCONJ
fcis-31421	221	21	baik	baik	PROPN
fcis-31421	221	22	,	,	PUNCT
fcis-31421	221	23	s.	s.	PROPN
fcis-31421	221	24	w.	w.	PROPN
fcis-31421	221	25	(	(	PUNCT
fcis-31421	221	26	2018	2018	NUM
fcis-31421	221	27	)	)	PUNCT
fcis-31421	221	28	.	.	PUNCT
fcis-31421	222	1	convolutional	convolutional	ADJ
fcis-31421	222	2	neural	neural	ADJ
fcis-31421	222	3	networks	network	NOUN
fcis-31421	222	4	based	base	VERB
fcis-31421	222	5	fire	fire	NOUN
fcis-31421	222	6	detection	detection	NOUN
fcis-31421	222	7	in	in	ADP
fcis-31421	222	8	surveillance	surveillance	NOUN
fcis-31421	222	9	videos	video	NOUN
fcis-31421	222	10	.	.	PUNCT
fcis-31421	223	1	ieee	ieee	NOUN
fcis-31421	223	2	access	access	NOUN
fcis-31421	223	3	,	,	PUNCT
fcis-31421	223	4	6	6	NUM
fcis-31421	223	5	,	,	PUNCT
fcis-31421	223	6	18174	18174	NUM
fcis-31421	223	7	-	-	SYM
fcis-31421	223	8	18183	18183	NUM
fcis-31421	223	9	.	.	PUNCT
fcis-31421	224	1	[	[	X
fcis-31421	224	2	3	3	NUM
fcis-31421	224	3	]	]	X
fcis-31421	224	4	barmpoutis	barmpoutis	NOUN
fcis-31421	224	5	,	,	PUNCT
fcis-31421	224	6	p.	p.	NOUN
fcis-31421	224	7	,	,	PUNCT
fcis-31421	224	8	dimitropoulos	dimitropoulos	PROPN
fcis-31421	224	9	,	,	PUNCT
fcis-31421	224	10	k.	k.	PROPN
fcis-31421	224	11	,	,	PUNCT
fcis-31421	224	12	kaza	kaza	PROPN
fcis-31421	224	13	,	,	PUNCT
fcis-31421	224	14	k.	k.	PROPN
fcis-31421	224	15	,	,	PUNCT
fcis-31421	224	16	&	&	CCONJ
fcis-31421	224	17	grammalidis	grammalidis	PROPN
fcis-31421	224	18	,	,	PUNCT
fcis-31421	224	19	n.	n.	NOUN
fcis-31421	224	20	(	(	PUNCT
fcis-31421	224	21	2019	2019	NUM
fcis-31421	224	22	,	,	PUNCT
fcis-31421	224	23	may	may	AUX
fcis-31421	224	24	)	)	PUNCT
fcis-31421	224	25	.	.	PUNCT
fcis-31421	225	1	fire	fire	NOUN
fcis-31421	225	2	detection	detection	NOUN
fcis-31421	225	3	from	from	ADP
fcis-31421	225	4	images	image	NOUN
fcis-31421	225	5	using	use	VERB
fcis-31421	225	6	faster	fast	ADJ
fcis-31421	225	7	rcnn	rcnn	NOUN
fcis-31421	225	8	and	and	CCONJ
fcis-31421	225	9	multidimensional	multidimensional	ADJ
fcis-31421	225	10	texture	texture	ADJ
fcis-31421	225	11	analysis	analysis	NOUN
fcis-31421	225	12	.	.	PUNCT
fcis-31421	226	1	in	in	ADP
fcis-31421	226	2	icassp	icassp	PROPN
fcis-31421	226	3	20192019	20192019	NUM
fcis-31421	226	4	ieee	ieee	PROPN
fcis-31421	226	5	international	international	ADJ
fcis-31421	226	6	conference	conference	NOUN
fcis-31421	226	7	on	on	ADP
fcis-31421	226	8	acoustics	acoustic	NOUN
fcis-31421	226	9	,	,	PUNCT
fcis-31421	226	10	speech	speech	NOUN
fcis-31421	226	11	and	and	CCONJ
fcis-31421	226	12	signal	signal	NOUN
fcis-31421	226	13	processing	processing	NOUN
fcis-31421	226	14	(	(	PUNCT
fcis-31421	226	15	icassp	icassp	PROPN
fcis-31421	226	16	)	)	PUNCT
fcis-31421	226	17	(	(	PUNCT
fcis-31421	226	18	pp	pp	X
fcis-31421	226	19	.	.	PUNCT
fcis-31421	226	20	8301	8301	NUM
fcis-31421	226	21	-	-	SYM
fcis-31421	226	22	8305	8305	NUM
fcis-31421	226	23	)	)	PUNCT
fcis-31421	226	24	.	.	PUNCT
fcis-31421	227	1	ieee	ieee	PROPN
fcis-31421	227	2	.	.	PUNCT
fcis-31421	228	1	[	[	X
fcis-31421	228	2	4	4	NUM
fcis-31421	228	3	]	]	X
fcis-31421	228	4	saeed	saeed	PROPN
fcis-31421	228	5	,	,	PUNCT
fcis-31421	228	6	f.	f.	PROPN
fcis-31421	228	7	,	,	PUNCT
fcis-31421	228	8	paul	paul	PROPN
fcis-31421	228	9	,	,	PUNCT
fcis-31421	228	10	a.	a.	PROPN
fcis-31421	228	11	,	,	PUNCT
fcis-31421	228	12	karthigaikumar	karthigaikumar	PROPN
fcis-31421	228	13	,	,	PUNCT
fcis-31421	228	14	p.	p.	PROPN
fcis-31421	228	15	,	,	PUNCT
fcis-31421	228	16	&	&	CCONJ
fcis-31421	228	17	nayyar	nayyar	PROPN
fcis-31421	228	18	,	,	PUNCT
fcis-31421	228	19	a.	a.	NOUN
fcis-31421	228	20	(	(	PUNCT
fcis-31421	228	21	2020	2020	NUM
fcis-31421	228	22	)	)	PUNCT
fcis-31421	228	23	.	.	PUNCT
fcis-31421	229	1	convolutional	convolutional	ADJ
fcis-31421	229	2	neural	neural	ADJ
fcis-31421	229	3	network	network	NOUN
fcis-31421	229	4	based	base	VERB
fcis-31421	229	5	early	early	ADJ
fcis-31421	229	6	fire	fire	NOUN
fcis-31421	229	7	detection	detection	NOUN
fcis-31421	229	8	.	.	PUNCT
fcis-31421	230	1	multimedia	multimedia	NOUN
fcis-31421	230	2	tools	tool	NOUN
fcis-31421	230	3	and	and	CCONJ
fcis-31421	230	4	applications	application	NOUN
fcis-31421	230	5	,	,	PUNCT
fcis-31421	230	6	79(13	79(13	NUM
fcis-31421	230	7	)	)	PUNCT
fcis-31421	230	8	,	,	PUNCT
fcis-31421	230	9	9083	9083	NUM
fcis-31421	230	10	-	-	SYM
fcis-31421	230	11	9099	9099	NUM
fcis-31421	230	12	.	.	PUNCT
fcis-31421	231	1	[	[	X
fcis-31421	231	2	5	5	NUM
fcis-31421	231	3	]	]	PUNCT
fcis-31421	231	4	sharma	sharma	PROPN
fcis-31421	231	5	,	,	PUNCT
fcis-31421	231	6	a.	a.	PROPN
fcis-31421	231	7	,	,	PUNCT
fcis-31421	231	8	kumar	kumar	PROPN
fcis-31421	231	9	,	,	PUNCT
fcis-31421	231	10	r.	r.	PROPN
fcis-31421	231	11	,	,	PUNCT
fcis-31421	231	12	kansal	kansal	NOUN
fcis-31421	231	13	,	,	PUNCT
fcis-31421	231	14	i.	i.	NOUN
fcis-31421	231	15	,	,	PUNCT
fcis-31421	231	16	popli	popli	PROPN
fcis-31421	231	17	,	,	PUNCT
fcis-31421	231	18	r.	r.	PROPN
fcis-31421	231	19	,	,	PUNCT
fcis-31421	231	20	khullar	khullar	ADJ
fcis-31421	231	21	,	,	PUNCT
fcis-31421	231	22	v.	v.	PROPN
fcis-31421	231	23	,	,	PUNCT
fcis-31421	231	24	verma	verma	PROPN
fcis-31421	231	25	,	,	PUNCT
fcis-31421	231	26	j.	j.	PROPN
fcis-31421	231	27	,	,	PUNCT
fcis-31421	231	28	&	&	CCONJ
fcis-31421	231	29	kumar	kumar	PROPN
fcis-31421	231	30	,	,	PUNCT
fcis-31421	231	31	s.	s.	PROPN
fcis-31421	231	32	(	(	PUNCT
fcis-31421	231	33	2024	2024	NUM
fcis-31421	231	34	)	)	PUNCT
fcis-31421	231	35	.	.	PUNCT
fcis-31421	232	1	fire	fire	NOUN
fcis-31421	232	2	detection	detection	NOUN
fcis-31421	232	3	in	in	ADP
fcis-31421	232	4	urban	urban	ADJ
fcis-31421	232	5	areas	area	NOUN
fcis-31421	232	6	using	use	VERB
fcis-31421	232	7	multimodal	multimodal	ADJ
fcis-31421	232	8	data	datum	NOUN
fcis-31421	232	9	and	and	CCONJ
fcis-31421	232	10	federated	federated	ADJ
fcis-31421	232	11	learning	learning	NOUN
fcis-31421	232	12	.	.	PUNCT
fcis-31421	233	1	fire	fire	NOUN
fcis-31421	233	2	,	,	PUNCT
fcis-31421	233	3	7(4	7(4	NUM
fcis-31421	233	4	)	)	PUNCT
fcis-31421	233	5	,	,	PUNCT
fcis-31421	233	6	104	104	NUM
fcis-31421	233	7	.	.	PUNCT
fcis-31421	234	1	[	[	X
fcis-31421	234	2	6	6	NUM
fcis-31421	234	3	]	]	PUNCT
fcis-31421	234	4	bhamra	bhamra	PROPN
fcis-31421	234	5	,	,	PUNCT
fcis-31421	234	6	j.	j.	PROPN
fcis-31421	234	7	k.	k.	PROPN
fcis-31421	234	8	,	,	PUNCT
fcis-31421	234	9	anantha	anantha	PROPN
fcis-31421	234	10	ramaprasad	ramaprasad	PROPN
fcis-31421	234	11	,	,	PUNCT
fcis-31421	234	12	s.	s.	PROPN
fcis-31421	234	13	,	,	PUNCT
fcis-31421	234	14	baldota	baldota	ADJ
fcis-31421	234	15	,	,	PUNCT
fcis-31421	234	16	s.	s.	PROPN
fcis-31421	234	17	,	,	PUNCT
fcis-31421	234	18	luna	luna	PROPN
fcis-31421	234	19	,	,	PUNCT
fcis-31421	234	20	s.	s.	PROPN
fcis-31421	234	21	,	,	PUNCT
fcis-31421	234	22	zen	zen	PROPN
fcis-31421	234	23	,	,	PUNCT
fcis-31421	234	24	e.	e.	PROPN
fcis-31421	234	25	,	,	PUNCT
fcis-31421	234	26	ramachandra	ramachandra	PROPN
fcis-31421	234	27	,	,	PUNCT
fcis-31421	234	28	r.	r.	PROPN
fcis-31421	234	29	,	,	PUNCT
fcis-31421	234	30	...	...	PUNCT
fcis-31421	234	31	&	&	CCONJ
fcis-31421	234	32	nguyen	nguyen	PROPN
fcis-31421	234	33	,	,	PUNCT
fcis-31421	234	34	m.	m.	NOUN
fcis-31421	234	35	h.	h.	PROPN
fcis-31421	234	36	(	(	PUNCT
fcis-31421	234	37	2023	2023	NUM
fcis-31421	234	38	)	)	PUNCT
fcis-31421	234	39	.	.	PUNCT
fcis-31421	235	1	multimodal	multimodal	NOUN
fcis-31421	235	2	wildland	wildland	NOUN
fcis-31421	235	3	fire	fire	NOUN
fcis-31421	235	4	smoke	smoke	NOUN
fcis-31421	235	5	detection	detection	NOUN
fcis-31421	235	6	.	.	PUNCT
fcis-31421	236	1	remote	remote	ADJ
fcis-31421	236	2	sensing	sensing	NOUN
fcis-31421	236	3	,	,	PUNCT
fcis-31421	236	4	15(11	15(11	NUM
fcis-31421	236	5	)	)	PUNCT
fcis-31421	236	6	,	,	PUNCT
fcis-31421	236	7	2790	2790	NUM
fcis-31421	236	8	.	.	PUNCT
fcis-31421	237	1	[	[	X
fcis-31421	237	2	7	7	NUM
fcis-31421	237	3	]	]	X
fcis-31421	237	4	toulouse	toulouse	NOUN
fcis-31421	237	5	,	,	PUNCT
fcis-31421	237	6	t.	t.	PROPN
fcis-31421	237	7	,	,	PUNCT
fcis-31421	237	8	rossi	rossi	PROPN
fcis-31421	237	9	,	,	PUNCT
fcis-31421	237	10	l.	l.	PROPN
fcis-31421	237	11	,	,	PUNCT
fcis-31421	237	12	akhloufi	akhloufi	NOUN
fcis-31421	237	13	,	,	PUNCT
fcis-31421	237	14	m.	m.	NOUN
fcis-31421	237	15	a.	a.	PROPN
fcis-31421	237	16	,	,	PUNCT
fcis-31421	237	17	pieri	pieri	PROPN
fcis-31421	237	18	,	,	PUNCT
fcis-31421	237	19	a.	a.	NOUN
fcis-31421	237	20	,	,	PUNCT
fcis-31421	237	21	&	&	CCONJ
fcis-31421	237	22	maldague	maldague	PROPN
fcis-31421	237	23	,	,	PUNCT
fcis-31421	237	24	x.	x.	NOUN
fcis-31421	237	25	(	(	PUNCT
fcis-31421	237	26	2018	2018	NUM
fcis-31421	237	27	)	)	PUNCT
fcis-31421	237	28	.	.	PUNCT
fcis-31421	238	1	a	a	DET
fcis-31421	238	2	multimodal	multimodal	ADJ
fcis-31421	238	3	3d	3d	NOUN
fcis-31421	238	4	framework	framework	NOUN
fcis-31421	238	5	for	for	ADP
fcis-31421	238	6	fire	fire	NOUN
fcis-31421	238	7	characteristics	characteristic	NOUN
fcis-31421	238	8	estimation	estimation	NOUN
fcis-31421	238	9	.	.	PUNCT
fcis-31421	239	1	measurement	measurement	NOUN
fcis-31421	239	2	science	science	NOUN
fcis-31421	239	3	and	and	CCONJ
fcis-31421	239	4	technology	technology	NOUN
fcis-31421	239	5	,	,	PUNCT
fcis-31421	239	6	29(2	29(2	NUM
fcis-31421	239	7	)	)	PUNCT
fcis-31421	239	8	,	,	PUNCT
fcis-31421	239	9	025404	025404	NUM
fcis-31421	239	10	.	.	PUNCT
fcis-31421	240	1	60	60	NUM
fcis-31421	241	1	[	[	SYM
fcis-31421	241	2	8	8	NUM
fcis-31421	241	3	]	]	SYM
fcis-31421	241	4	chaoxia	chaoxia	PROPN
fcis-31421	241	5	,	,	PUNCT
fcis-31421	241	6	c.	c.	PROPN
fcis-31421	241	7	,	,	PUNCT
fcis-31421	241	8	shang	shang	PROPN
fcis-31421	241	9	,	,	PUNCT
fcis-31421	241	10	w.	w.	PROPN
fcis-31421	241	11	,	,	PUNCT
fcis-31421	241	12	zhang	zhang	PROPN
fcis-31421	241	13	,	,	PUNCT
fcis-31421	241	14	f.	f.	PROPN
fcis-31421	241	15	,	,	PUNCT
fcis-31421	241	16	&	&	CCONJ
fcis-31421	241	17	cong	cong	PROPN
fcis-31421	241	18	,	,	PUNCT
fcis-31421	241	19	s.	s.	PROPN
fcis-31421	241	20	(	(	PUNCT
fcis-31421	241	21	2022	2022	NUM
fcis-31421	241	22	)	)	PUNCT
fcis-31421	241	23	.	.	PUNCT
fcis-31421	242	1	weakly	weakly	ADJ
fcis-31421	242	2	aligned	align	VERB
fcis-31421	242	3	multimodal	multimodal	ADJ
fcis-31421	242	4	flame	flame	NOUN
fcis-31421	242	5	detection	detection	NOUN
fcis-31421	242	6	for	for	ADP
fcis-31421	242	7	fire	fire	NOUN
fcis-31421	242	8	-	-	PUNCT
fcis-31421	242	9	fighting	fight	VERB
fcis-31421	242	10	robots	robot	NOUN
fcis-31421	242	11	.	.	PUNCT
fcis-31421	243	1	ieee	ieee	NOUN
fcis-31421	243	2	transactions	transaction	NOUN
fcis-31421	243	3	on	on	ADP
fcis-31421	243	4	industrial	industrial	ADJ
fcis-31421	243	5	informatics	informatic	NOUN
fcis-31421	243	6	,	,	PUNCT
fcis-31421	243	7	19(3	19(3	NUM
fcis-31421	243	8	)	)	PUNCT
fcis-31421	243	9	,	,	PUNCT
fcis-31421	243	10	2866	2866	NUM
fcis-31421	243	11	-	-	SYM
fcis-31421	243	12	2875	2875	NUM
fcis-31421	243	13	.	.	PUNCT
fcis-31421	244	1	[	[	X
fcis-31421	244	2	9	9	NUM
fcis-31421	244	3	]	]	X
fcis-31421	244	4	talaat	talaat	PROPN
fcis-31421	244	5	,	,	PUNCT
fcis-31421	244	6	f.	f.	PROPN
fcis-31421	244	7	m.	m.	PROPN
fcis-31421	244	8	,	,	PUNCT
fcis-31421	244	9	&	&	CCONJ
fcis-31421	244	10	zaineldin	zaineldin	PROPN
fcis-31421	244	11	,	,	PUNCT
fcis-31421	244	12	h.	h.	PROPN
fcis-31421	244	13	(	(	PUNCT
fcis-31421	244	14	2023	2023	NUM
fcis-31421	244	15	)	)	PUNCT
fcis-31421	244	16	.	.	PUNCT
fcis-31421	245	1	an	an	DET
fcis-31421	245	2	improved	improve	VERB
fcis-31421	245	3	fire	fire	NOUN
fcis-31421	245	4	detection	detection	NOUN
fcis-31421	245	5	approach	approach	NOUN
fcis-31421	245	6	based	base	VERB
fcis-31421	245	7	on	on	ADP
fcis-31421	245	8	yolo	yolo	PROPN
fcis-31421	245	9	-	-	PUNCT
fcis-31421	245	10	v8	v8	PROPN
fcis-31421	245	11	for	for	ADP
fcis-31421	245	12	smart	smart	ADJ
fcis-31421	245	13	cities	city	NOUN
fcis-31421	245	14	.	.	PUNCT
fcis-31421	246	1	neural	neural	ADJ
fcis-31421	246	2	computing	computing	NOUN
fcis-31421	246	3	and	and	CCONJ
fcis-31421	246	4	applications	application	NOUN
fcis-31421	246	5	,	,	PUNCT
fcis-31421	246	6	35(28	35(28	NUM
fcis-31421	246	7	)	)	PUNCT
fcis-31421	246	8	,	,	PUNCT
fcis-31421	246	9	20939	20939	NUM
fcis-31421	246	10	-	-	SYM
fcis-31421	246	11	20954	20954	NUM
fcis-31421	246	12	.	.	PUNCT
fcis-31421	247	1	[	[	X
fcis-31421	247	2	10	10	NUM
fcis-31421	247	3	]	]	X
fcis-31421	247	4	goyal	goyal	PROPN
fcis-31421	247	5	,	,	PUNCT
fcis-31421	247	6	s.	s.	PROPN
fcis-31421	247	7	,	,	PUNCT
fcis-31421	247	8	shagill	shagill	NOUN
fcis-31421	247	9	,	,	PUNCT
fcis-31421	247	10	m.	m.	NOUN
fcis-31421	247	11	,	,	PUNCT
fcis-31421	247	12	kaur	kaur	PROPN
fcis-31421	247	13	,	,	PUNCT
fcis-31421	247	14	a.	a.	NOUN
fcis-31421	247	15	,	,	PUNCT
fcis-31421	247	16	vohra	vohra	NOUN
fcis-31421	247	17	,	,	PUNCT
fcis-31421	247	18	h.	h.	PROPN
fcis-31421	247	19	,	,	PUNCT
fcis-31421	247	20	&	&	CCONJ
fcis-31421	247	21	singh	singh	PROPN
fcis-31421	247	22	,	,	PUNCT
fcis-31421	247	23	a.	a.	NOUN
fcis-31421	247	24	(	(	PUNCT
fcis-31421	247	25	2020	2020	NUM
fcis-31421	247	26	)	)	PUNCT
fcis-31421	247	27	.	.	PUNCT
fcis-31421	248	1	a	a	DET
fcis-31421	248	2	yolo	yolo	ADJ
fcis-31421	248	3	based	base	VERB
fcis-31421	248	4	technique	technique	NOUN
fcis-31421	248	5	for	for	ADP
fcis-31421	248	6	early	early	ADJ
fcis-31421	248	7	forest	forest	NOUN
fcis-31421	248	8	fire	fire	NOUN
fcis-31421	248	9	detection	detection	NOUN
fcis-31421	248	10	.	.	PUNCT
fcis-31421	249	1	int	int	NOUN
fcis-31421	249	2	.	.	PUNCT
fcis-31421	250	1	j.	j.	PROPN
fcis-31421	250	2	innov	innov	PROPN
fcis-31421	250	3	.	.	PUNCT
fcis-31421	251	1	technol	technol	PROPN
fcis-31421	251	2	.	.	PUNCT
fcis-31421	251	3	explor	explor	PROPN
fcis-31421	251	4	.	.	PUNCT
fcis-31421	252	1	eng	eng	PROPN
fcis-31421	252	2	,	,	PUNCT
fcis-31421	252	3	9	9	NUM
fcis-31421	252	4	,	,	PUNCT
fcis-31421	252	5	1357	1357	NUM
fcis-31421	252	6	-	-	SYM
fcis-31421	252	7	1362	1362	NUM
fcis-31421	252	8	.	.	PUNCT
fcis-31421	253	1	[	[	X
fcis-31421	253	2	11	11	NUM
fcis-31421	253	3	]	]	SYM
fcis-31421	253	4	wu	wu	PROPN
fcis-31421	253	5	,	,	PUNCT
fcis-31421	253	6	h.	h.	PROPN
fcis-31421	253	7	,	,	PUNCT
fcis-31421	253	8	hu	hu	PROPN
fcis-31421	253	9	,	,	PUNCT
fcis-31421	253	10	y.	y.	PROPN
fcis-31421	253	11	,	,	PUNCT
fcis-31421	253	12	wang	wang	PROPN
fcis-31421	253	13	,	,	PUNCT
fcis-31421	253	14	w.	w.	PROPN
fcis-31421	253	15	,	,	PUNCT
fcis-31421	253	16	mei	mei	PROPN
fcis-31421	253	17	,	,	PUNCT
fcis-31421	253	18	x.	x.	PROPN
fcis-31421	253	19	,	,	PUNCT
fcis-31421	253	20	&	&	CCONJ
fcis-31421	253	21	xian	xian	PROPN
fcis-31421	253	22	,	,	PUNCT
fcis-31421	253	23	j.	j.	PROPN
fcis-31421	253	24	(	(	PUNCT
fcis-31421	253	25	2022	2022	NUM
fcis-31421	253	26	)	)	PUNCT
fcis-31421	253	27	.	.	PUNCT
fcis-31421	254	1	ship	ship	NOUN
fcis-31421	254	2	fire	fire	NOUN
fcis-31421	254	3	detection	detection	NOUN
fcis-31421	254	4	based	base	VERB
fcis-31421	254	5	on	on	ADP
fcis-31421	254	6	an	an	DET
fcis-31421	254	7	improved	improve	VERB
fcis-31421	254	8	yolo	yolo	ADJ
fcis-31421	254	9	algorithm	algorithm	NOUN
fcis-31421	254	10	with	with	ADP
fcis-31421	254	11	a	a	DET
fcis-31421	254	12	lightweight	lightweight	ADJ
fcis-31421	254	13	convolutional	convolutional	ADJ
fcis-31421	254	14	neural	neural	ADJ
fcis-31421	254	15	network	network	NOUN
fcis-31421	254	16	model	model	NOUN
fcis-31421	254	17	.	.	PUNCT
fcis-31421	255	1	sensors	sensor	NOUN
fcis-31421	255	2	,	,	PUNCT
fcis-31421	255	3	22(19	22(19	NUM
fcis-31421	255	4	)	)	PUNCT
fcis-31421	255	5	,	,	PUNCT
fcis-31421	255	6	7420	7420	NUM
fcis-31421	255	7	.	.	PUNCT
fcis-31421	256	1	[	[	X
fcis-31421	256	2	12	12	NUM
fcis-31421	256	3	]	]	X
fcis-31421	256	4	cao	cao	PROPN
fcis-31421	256	5	,	,	PUNCT
fcis-31421	256	6	l.	l.	PROPN
fcis-31421	256	7	,	,	PUNCT
fcis-31421	256	8	shen	shen	PROPN
fcis-31421	256	9	,	,	PUNCT
fcis-31421	256	10	z.	z.	PROPN
fcis-31421	256	11	,	,	PUNCT
fcis-31421	256	12	&	&	CCONJ
fcis-31421	256	13	xu	xu	PROPN
fcis-31421	256	14	,	,	PUNCT
fcis-31421	256	15	s.	s.	PROPN
fcis-31421	256	16	(	(	PUNCT
fcis-31421	256	17	2024	2024	NUM
fcis-31421	256	18	)	)	PUNCT
fcis-31421	256	19	.	.	PUNCT
fcis-31421	257	1	efficient	efficient	ADJ
fcis-31421	257	2	forest	forest	NOUN
fcis-31421	257	3	fire	fire	NOUN
fcis-31421	257	4	detection	detection	NOUN
fcis-31421	257	5	based	base	VERB
fcis-31421	257	6	on	on	ADP
fcis-31421	257	7	an	an	DET
fcis-31421	257	8	improved	improve	VERB
fcis-31421	257	9	yolo	yolo	ADJ
fcis-31421	257	10	model	model	NOUN
fcis-31421	257	11	.	.	PUNCT
fcis-31421	258	1	visual	visual	ADJ
fcis-31421	258	2	intelligence	intelligence	NOUN
fcis-31421	258	3	,	,	PUNCT
fcis-31421	258	4	2(1	2(1	NUM
fcis-31421	258	5	)	)	PUNCT
fcis-31421	258	6	,	,	PUNCT
fcis-31421	258	7	20	20	NUM
fcis-31421	258	8	.	.	PUNCT
fcis-31421	259	1	[	[	X
fcis-31421	259	2	13	13	NUM
fcis-31421	259	3	]	]	SYM
fcis-31421	259	4	yun	yun	PROPN
fcis-31421	259	5	,	,	PUNCT
fcis-31421	259	6	b.	b.	PROPN
fcis-31421	259	7	,	,	PUNCT
fcis-31421	259	8	zheng	zheng	PROPN
fcis-31421	259	9	,	,	PUNCT
fcis-31421	259	10	y.	y.	PROPN
fcis-31421	259	11	,	,	PUNCT
fcis-31421	259	12	lin	lin	PROPN
fcis-31421	259	13	,	,	PUNCT
fcis-31421	259	14	z.	z.	PROPN
fcis-31421	259	15	,	,	PUNCT
fcis-31421	259	16	&	&	CCONJ
fcis-31421	259	17	li	li	PROPN
fcis-31421	259	18	,	,	PUNCT
fcis-31421	259	19	t.	t.	PROPN
fcis-31421	259	20	(	(	PUNCT
fcis-31421	259	21	2024	2024	NUM
fcis-31421	259	22	)	)	PUNCT
fcis-31421	259	23	.	.	PUNCT
fcis-31421	260	1	ffyolo	ffyolo	NOUN
fcis-31421	260	2	:	:	PUNCT
fcis-31421	260	3	a	a	DET
fcis-31421	260	4	lightweight	lightweight	ADJ
fcis-31421	260	5	forest	forest	NOUN
fcis-31421	260	6	fire	fire	NOUN
fcis-31421	260	7	detection	detection	NOUN
fcis-31421	260	8	model	model	NOUN
fcis-31421	260	9	based	base	VERB
fcis-31421	260	10	on	on	ADP
fcis-31421	260	11	yolov8	yolov8	PROPN
fcis-31421	260	12	.	.	PUNCT
fcis-31421	261	1	fire	fire	PROPN
fcis-31421	261	2	,	,	PUNCT
fcis-31421	261	3	7(3	7(3	NUM
fcis-31421	261	4	)	)	PUNCT
fcis-31421	261	5	,	,	PUNCT
fcis-31421	261	6	93	93	NUM
fcis-31421	261	7	.	.	PUNCT