id sid tid token lemma pos brj-24033 1 1 peer peer NOUN brj-24033 1 2 - - PUNCT brj-24033 1 3 review review NOUN brj-24033 1 4 article article NOUN brj-24033 1 5 peer peer NOUN brj-24033 1 6 - - PUNCT brj-24033 1 7 reviewed review VERB brj-24033 1 8 article article NOUN brj-24033 1 9 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 1 10 zhang zhang PROPN brj-24033 1 11 & & CCONJ brj-24033 1 12 zhao zhao PROPN brj-24033 1 13 ( ( PUNCT brj-24033 1 14 2025 2025 NUM brj-24033 1 15 ) ) PUNCT brj-24033 1 16 . . PUNCT brj-24033 2 1 β€œ " PUNCT brj-24033 2 2 wood wood NOUN brj-24033 2 3 image image NOUN brj-24033 2 4 classification classification NOUN brj-24033 2 5 , , PUNCT brj-24033 2 6 ” " PUNCT brj-24033 2 7 bioresources bioresource NOUN brj-24033 2 8 20(1 20(1 NUM brj-24033 2 9 ) ) PUNCT brj-24033 2 10 , , PUNCT brj-24033 2 11 944 944 NUM brj-24033 2 12 - - SYM brj-24033 2 13 955 955 NUM brj-24033 2 14 . . PUNCT brj-24033 3 1 944 944 NUM brj-24033 3 2 wood wood NOUN brj-24033 3 3 species specie NOUN brj-24033 3 4 classification classification NOUN brj-24033 3 5 in in ADP brj-24033 3 6 open open ADJ brj-24033 3 7 set set NOUN brj-24033 3 8 using use VERB brj-24033 3 9 an an DET brj-24033 3 10 improved improved ADJ brj-24033 3 11 nno nno NOUN brj-24033 3 12 classifier classifier PROPN brj-24033 3 13 ke ke PROPN brj-24033 3 14 - - PUNCT brj-24033 3 15 xin xin PROPN brj-24033 3 16 zhang zhang PROPN brj-24033 3 17 and and CCONJ brj-24033 3 18 peng peng PROPN brj-24033 3 19 zhao zhao PROPN brj-24033 3 20 * * PUNCT brj-24033 3 21 a a DET brj-24033 3 22 wood wood NOUN brj-24033 3 23 species species NOUN brj-24033 3 24 classification classification NOUN brj-24033 3 25 scheme scheme NOUN brj-24033 3 26 was be AUX brj-24033 3 27 developed develop VERB brj-24033 3 28 based base VERB brj-24033 3 29 on on ADP brj-24033 3 30 open open ADJ brj-24033 3 31 set set NOUN brj-24033 3 32 using use VERB brj-24033 3 33 an an DET brj-24033 3 34 improved improved ADJ brj-24033 3 35 nearest near ADJ brj-24033 3 36 non non ADJ brj-24033 3 37 - - ADJ brj-24033 3 38 outlier outlier ADJ brj-24033 3 39 ( ( PUNCT brj-24033 3 40 nno nno NOUN brj-24033 3 41 ) ) PUNCT brj-24033 3 42 classifier classifier NOUN brj-24033 3 43 . . PUNCT brj-24033 4 1 near near ADP brj-24033 4 2 infrared infrared PROPN brj-24033 4 3 ( ( PUNCT brj-24033 4 4 nir nir ADJ brj-24033 4 5 ) ) PUNCT brj-24033 4 6 spectral spectral ADJ brj-24033 4 7 curves curve NOUN brj-24033 4 8 were be AUX brj-24033 4 9 collected collect VERB brj-24033 4 10 in in ADP brj-24033 4 11 spectral spectral ADJ brj-24033 4 12 band band NOUN brj-24033 4 13 950 950 NUM brj-24033 4 14 to to ADP brj-24033 4 15 1650 1650 NUM brj-24033 4 16 nm nm NOUN brj-24033 4 17 by by ADP brj-24033 4 18 a a DET brj-24033 4 19 micro micro ADJ brj-24033 4 20 spectrometer spectrometer NOUN brj-24033 4 21 . . PUNCT brj-24033 5 1 the the DET brj-24033 5 2 spectral spectral ADJ brj-24033 5 3 dimension dimension NOUN brj-24033 5 4 reduction reduction NOUN brj-24033 5 5 was be AUX brj-24033 5 6 performed perform VERB brj-24033 5 7 with with ADP brj-24033 5 8 a a DET brj-24033 5 9 metric metric ADJ brj-24033 5 10 learning learning NOUN brj-24033 5 11 ( ( PUNCT brj-24033 5 12 ml ml NOUN brj-24033 5 13 ) ) PUNCT brj-24033 5 14 algorithm algorithm NOUN brj-24033 5 15 . . PUNCT brj-24033 6 1 two two NUM brj-24033 6 2 improvements improvement NOUN brj-24033 6 3 were be AUX brj-24033 6 4 proposed propose VERB brj-24033 6 5 in in ADP brj-24033 6 6 the the DET brj-24033 6 7 following follow VERB brj-24033 6 8 nno nno PROPN brj-24033 6 9 classifier classifier PROPN brj-24033 6 10 . . PUNCT brj-24033 7 1 first first ADV brj-24033 7 2 , , PUNCT brj-24033 7 3 a a DET brj-24033 7 4 cluster cluster NOUN brj-24033 7 5 analysis analysis NOUN brj-24033 7 6 was be AUX brj-24033 7 7 performed perform VERB brj-24033 7 8 in in ADP brj-24033 7 9 each each DET brj-24033 7 10 wood wood NOUN brj-24033 7 11 class class NOUN brj-24033 7 12 by by ADP brj-24033 7 13 using use VERB brj-24033 7 14 a a DET brj-24033 7 15 density density NOUN brj-24033 7 16 peak peak NOUN brj-24033 7 17 clustering clustering NOUN brj-24033 7 18 ( ( PUNCT brj-24033 7 19 dpc dpc PROPN brj-24033 7 20 ) ) PUNCT brj-24033 7 21 algorithm algorithm NOUN brj-24033 7 22 to to PART brj-24033 7 23 get get VERB brj-24033 7 24 1 1 NUM brj-24033 7 25 to to PART brj-24033 7 26 3 3 NUM brj-24033 7 27 clusters cluster NOUN brj-24033 7 28 . . PUNCT brj-24033 8 1 a a DET brj-24033 8 2 fixed fix VERB brj-24033 8 3 threshold threshold NOUN brj-24033 8 4 𝜏 𝜏 NOUN brj-24033 8 5 for for ADP brj-24033 8 6 all all DET brj-24033 8 7 wood wood NOUN brj-24033 8 8 classes class NOUN brj-24033 8 9 was be AUX brj-24033 8 10 replaced replace VERB brj-24033 8 11 by by ADP brj-24033 8 12 a a DET brj-24033 8 13 variable variable ADJ brj-24033 8 14 𝜏 𝜏 NOUN brj-24033 8 15 for for ADP brj-24033 8 16 all all DET brj-24033 8 17 clusters cluster NOUN brj-24033 8 18 . . PUNCT brj-24033 9 1 this this DET brj-24033 9 2 threshold threshold NOUN brj-24033 9 3 𝜏 𝜏 NOUN brj-24033 9 4 defines define VERB brj-24033 9 5 an an DET brj-24033 9 6 internal internal ADJ brj-24033 9 7 boundary boundary NOUN brj-24033 9 8 for for ADP brj-24033 9 9 one one NUM brj-24033 9 10 wood wood NOUN brj-24033 9 11 species specie NOUN brj-24033 9 12 to to PART brj-24033 9 13 further far ADV brj-24033 9 14 compute compute VERB brj-24033 9 15 a a DET brj-24033 9 16 class class NOUN brj-24033 9 17 membership membership NOUN brj-24033 9 18 score score NOUN brj-24033 9 19 for for ADP brj-24033 9 20 all all DET brj-24033 9 21 wood wood NOUN brj-24033 9 22 species specie NOUN brj-24033 9 23 . . PUNCT brj-24033 10 1 the the DET brj-24033 10 2 classification classification NOUN brj-24033 10 3 accuracy accuracy NOUN brj-24033 10 4 based base VERB brj-24033 10 5 on on ADP brj-24033 10 6 these these DET brj-24033 10 7 clusters cluster NOUN brj-24033 10 8 of of ADP brj-24033 10 9 each each DET brj-24033 10 10 wood wood NOUN brj-24033 10 11 class class NOUN brj-24033 10 12 was be AUX brj-24033 10 13 better well ADJ brj-24033 10 14 than than ADP brj-24033 10 15 that that PRON brj-24033 10 16 based base VERB brj-24033 10 17 on on ADP brj-24033 10 18 each each DET brj-24033 10 19 class class NOUN brj-24033 10 20 . . PUNCT brj-24033 11 1 the the DET brj-24033 11 2 experimental experimental ADJ brj-24033 11 3 results result NOUN brj-24033 11 4 in in ADP brj-24033 11 5 different different ADJ brj-24033 11 6 open open ADJ brj-24033 11 7 set set VERB brj-24033 11 8 scenarios scenario NOUN brj-24033 11 9 demonstrate demonstrate VERB brj-24033 11 10 that that SCONJ brj-24033 11 11 the the DET brj-24033 11 12 improved improved ADJ brj-24033 11 13 nno nno PROPN brj-24033 11 14 classifier classifier NOUN brj-24033 11 15 outperformed outperform VERB brj-24033 11 16 the the DET brj-24033 11 17 original original ADJ brj-24033 11 18 nno nno NOUN brj-24033 11 19 classifier classifier NOUN brj-24033 11 20 and and CCONJ brj-24033 11 21 some some DET brj-24033 11 22 other other ADJ brj-24033 11 23 state state NOUN brj-24033 11 24 - - PUNCT brj-24033 11 25 of of ADP brj-24033 11 26 - - PUNCT brj-24033 11 27 the the DET brj-24033 11 28 - - PUNCT brj-24033 11 29 art art NOUN brj-24033 11 30 open open ADJ brj-24033 11 31 set set NOUN brj-24033 11 32 recognition recognition NOUN brj-24033 11 33 ( ( PUNCT brj-24033 11 34 osr osr NOUN brj-24033 11 35 ) ) PUNCT brj-24033 11 36 algorithms algorithm NOUN brj-24033 11 37 . . PUNCT brj-24033 12 1 doi doi NOUN brj-24033 12 2 : : PUNCT brj-24033 12 3 10.15376 10.15376 NUM brj-24033 12 4 / / SYM brj-24033 12 5 biores.20.1.944 biores.20.1.944 NOUN brj-24033 12 6 - - PUNCT brj-24033 12 7 955 955 NUM brj-24033 12 8 keywords keyword NOUN brj-24033 12 9 : : PUNCT brj-24033 12 10 wood wood NOUN brj-24033 12 11 species specie NOUN brj-24033 12 12 classification classification NOUN brj-24033 12 13 ; ; PUNCT brj-24033 12 14 open open ADJ brj-24033 12 15 set set NOUN brj-24033 12 16 recognition recognition NOUN brj-24033 12 17 ; ; PUNCT brj-24033 12 18 spectral spectral ADJ brj-24033 12 19 analysis analysis NOUN brj-24033 12 20 ; ; PUNCT brj-24033 12 21 nno nno PROPN brj-24033 12 22 algorithm algorithm PROPN brj-24033 12 23 contact contact NOUN brj-24033 12 24 information information NOUN brj-24033 12 25 : : PUNCT brj-24033 12 26 school school NOUN brj-24033 12 27 of of ADP brj-24033 12 28 computer computer NOUN brj-24033 12 29 science science NOUN brj-24033 12 30 and and CCONJ brj-24033 12 31 technology technology NOUN brj-24033 12 32 , , PUNCT brj-24033 12 33 guangxi guangxi PROPN brj-24033 12 34 university university PROPN brj-24033 12 35 of of ADP brj-24033 12 36 science science NOUN brj-24033 12 37 and and CCONJ brj-24033 12 38 technology technology NOUN brj-24033 12 39 , , PUNCT brj-24033 12 40 liuzhou liuzhou PROPN brj-24033 12 41 545006 545006 NUM brj-24033 12 42 , , PUNCT brj-24033 12 43 china china PROPN brj-24033 12 44 ; ; PUNCT brj-24033 12 45 * * PUNCT brj-24033 12 46 corresponding correspond VERB brj-24033 12 47 author author NOUN brj-24033 12 48 : : PUNCT brj-24033 12 49 bit_zhao@aliyun.com bit_zhao@aliyun.com X brj-24033 12 50 introduction introduction NOUN brj-24033 12 51 there there PRON brj-24033 12 52 are be VERB brj-24033 12 53 approximately approximately ADV brj-24033 12 54 60,000 60,000 NUM brj-24033 12 55 tree tree NOUN brj-24033 12 56 species specie NOUN brj-24033 12 57 around around ADP brj-24033 12 58 the the DET brj-24033 12 59 world world NOUN brj-24033 12 60 according accord VERB brj-24033 12 61 to to ADP brj-24033 12 62 statistical statistical ADJ brj-24033 12 63 data datum NOUN brj-24033 12 64 . . PUNCT brj-24033 13 1 it it PRON brj-24033 13 2 is be AUX brj-24033 13 3 very very ADV brj-24033 13 4 hard hard ADJ brj-24033 13 5 to to PART brj-24033 13 6 develop develop VERB brj-24033 13 7 a a DET brj-24033 13 8 wood wood NOUN brj-24033 13 9 species species NOUN brj-24033 13 10 classification classification NOUN brj-24033 13 11 system system NOUN brj-24033 13 12 to to PART brj-24033 13 13 classify classify VERB brj-24033 13 14 all all DET brj-24033 13 15 these these DET brj-24033 13 16 tree tree NOUN brj-24033 13 17 species specie NOUN brj-24033 13 18 . . PUNCT brj-24033 14 1 in in ADP brj-24033 14 2 most most ADJ brj-24033 14 3 cases case NOUN brj-24033 14 4 , , PUNCT brj-24033 14 5 only only ADV brj-24033 14 6 those those DET brj-24033 14 7 tree tree NOUN brj-24033 14 8 species specie NOUN brj-24033 14 9 in in ADP brj-24033 14 10 a a DET brj-24033 14 11 specific specific ADJ brj-24033 14 12 region region NOUN brj-24033 14 13 ( ( PUNCT brj-24033 14 14 e.g. e.g. ADV brj-24033 14 15 , , PUNCT brj-24033 14 16 in in ADP brj-24033 14 17 heilongjiang heilongjiang PROPN brj-24033 14 18 province province NOUN brj-24033 14 19 of of ADP brj-24033 14 20 china china PROPN brj-24033 14 21 ) ) PUNCT brj-24033 14 22 or or CCONJ brj-24033 14 23 specific specific ADJ brj-24033 14 24 category category NOUN brj-24033 14 25 ( ( PUNCT brj-24033 14 26 e.g. e.g. ADV brj-24033 14 27 , , PUNCT brj-24033 14 28 mahogany mahogany NOUN brj-24033 14 29 ) ) PUNCT brj-24033 14 30 need need VERB brj-24033 14 31 classification classification NOUN brj-24033 14 32 . . PUNCT brj-24033 15 1 these these DET brj-24033 15 2 specific specific ADJ brj-24033 15 3 tree tree NOUN brj-24033 15 4 species specie NOUN brj-24033 15 5 are be AUX brj-24033 15 6 included include VERB brj-24033 15 7 in in ADP brj-24033 15 8 the the DET brj-24033 15 9 training training NOUN brj-24033 15 10 set set NOUN brj-24033 15 11 of of ADP brj-24033 15 12 the the DET brj-24033 15 13 above above ADV brj-24033 15 14 - - PUNCT brj-24033 15 15 mentioned mention VERB brj-24033 15 16 classification classification NOUN brj-24033 15 17 system system NOUN brj-24033 15 18 . . PUNCT brj-24033 16 1 however however ADV brj-24033 16 2 , , PUNCT brj-24033 16 3 those those DET brj-24033 16 4 tree tree NOUN brj-24033 16 5 species specie NOUN brj-24033 16 6 not not PART brj-24033 16 7 in in ADP brj-24033 16 8 the the DET brj-24033 16 9 training training NOUN brj-24033 16 10 set set NOUN brj-24033 16 11 should should AUX brj-24033 16 12 be be AUX brj-24033 16 13 rejected reject VERB brj-24033 16 14 correctly correctly ADV brj-24033 16 15 by by ADP brj-24033 16 16 this this DET brj-24033 16 17 classification classification NOUN brj-24033 16 18 system system NOUN brj-24033 16 19 . . PUNCT brj-24033 17 1 in in ADP brj-24033 17 2 summary summary NOUN brj-24033 17 3 , , PUNCT brj-24033 17 4 wood wood NOUN brj-24033 17 5 species specie NOUN brj-24033 17 6 classification classification NOUN brj-24033 17 7 should should AUX brj-24033 17 8 be be AUX brj-24033 17 9 studied study VERB brj-24033 17 10 in in ADP brj-24033 17 11 an an DET brj-24033 17 12 open open ADJ brj-24033 17 13 set set VERB brj-24033 17 14 scenario scenario NOUN brj-24033 17 15 in in ADP brj-24033 17 16 practice practice NOUN brj-24033 17 17 . . PUNCT brj-24033 18 1 however however ADV brj-24033 18 2 , , PUNCT brj-24033 18 3 most most ADJ brj-24033 18 4 wood wood NOUN brj-24033 18 5 species species NOUN brj-24033 18 6 classification classification NOUN brj-24033 18 7 investigations investigation NOUN brj-24033 18 8 are be AUX brj-24033 18 9 performed perform VERB brj-24033 18 10 in in ADP brj-24033 18 11 a a DET brj-24033 18 12 closed closed ADJ brj-24033 18 13 set set VERB brj-24033 18 14 scenario scenario NOUN brj-24033 18 15 with with ADP brj-24033 18 16 emphasis emphasis NOUN brj-24033 18 17 on on ADP brj-24033 18 18 classification classification NOUN brj-24033 18 19 methodologies methodology NOUN brj-24033 18 20 such such ADJ brj-24033 18 21 as as ADP brj-24033 18 22 spectral spectral ADJ brj-24033 18 23 analysis analysis NOUN brj-24033 18 24 and and CCONJ brj-24033 18 25 anatomical anatomical ADJ brj-24033 18 26 analysis analysis NOUN brj-24033 18 27 ( ( PUNCT brj-24033 18 28 zhan zhan PROPN brj-24033 18 29 et et PROPN brj-24033 18 30 al al PROPN brj-24033 18 31 . . PROPN brj-24033 18 32 2023 2023 NUM brj-24033 18 33 ; ; PUNCT brj-24033 18 34 ma ma PROPN brj-24033 18 35 et et PROPN brj-24033 18 36 al al PROPN brj-24033 18 37 . . PROPN brj-24033 18 38 2021 2021 NUM brj-24033 18 39 ; ; PUNCT brj-24033 18 40 park park NOUN brj-24033 18 41 et et PROPN brj-24033 18 42 al al PROPN brj-24033 18 43 . . PROPN brj-24033 18 44 2021 2021 NUM brj-24033 18 45 ; ; PUNCT brj-24033 18 46 tuncer tuncer NOUN brj-24033 18 47 et et PROPN brj-24033 18 48 al al PROPN brj-24033 18 49 . . PROPN brj-24033 18 50 2021 2021 NUM brj-24033 18 51 ) ) PUNCT brj-24033 18 52 . . PUNCT brj-24033 19 1 the the DET brj-24033 19 2 wood wood NOUN brj-24033 19 3 species specie NOUN brj-24033 19 4 in in ADP brj-24033 19 5 the the DET brj-24033 19 6 training training NOUN brj-24033 19 7 set set NOUN brj-24033 19 8 ( ( PUNCT brj-24033 19 9 i.e. i.e. X brj-24033 19 10 , , PUNCT brj-24033 19 11 the the DET brj-24033 19 12 known know VERB brj-24033 19 13 species specie NOUN brj-24033 19 14 ) ) PUNCT brj-24033 19 15 can can AUX brj-24033 19 16 be be AUX brj-24033 19 17 classified classify VERB brj-24033 19 18 correctly correctly ADV brj-24033 19 19 , , PUNCT brj-24033 19 20 whereas whereas SCONJ brj-24033 19 21 those those PRON brj-24033 19 22 not not PART brj-24033 19 23 in in ADP brj-24033 19 24 the the DET brj-24033 19 25 training training NOUN brj-24033 19 26 set set NOUN brj-24033 19 27 ( ( PUNCT brj-24033 19 28 i.e. i.e. X brj-24033 19 29 , , PUNCT brj-24033 19 30 the the DET brj-24033 19 31 unknown unknown ADJ brj-24033 19 32 species specie NOUN brj-24033 19 33 ) ) PUNCT brj-24033 19 34 will will AUX brj-24033 19 35 be be AUX brj-24033 19 36 misclassified misclassifie VERB brj-24033 19 37 as as ADP brj-24033 19 38 one one NUM brj-24033 19 39 known know VERB brj-24033 19 40 species specie NOUN brj-24033 19 41 in in ADP brj-24033 19 42 a a DET brj-24033 19 43 closed closed ADJ brj-24033 19 44 set set ADJ brj-24033 19 45 scenario scenario NOUN brj-24033 19 46 . . PUNCT brj-24033 20 1 in in ADP brj-24033 20 2 fact fact NOUN brj-24033 20 3 , , PUNCT brj-24033 20 4 to to ADP brj-24033 20 5 the the DET brj-24033 20 6 best good ADJ brj-24033 20 7 of of ADP brj-24033 20 8 our our PRON brj-24033 20 9 knowledge knowledge NOUN brj-24033 20 10 , , PUNCT brj-24033 20 11 the the DET brj-24033 20 12 wood wood NOUN brj-24033 20 13 species species NOUN brj-24033 20 14 number number NOUN brj-24033 20 15 in in ADP brj-24033 20 16 the the DET brj-24033 20 17 training training NOUN brj-24033 20 18 set set NOUN brj-24033 20 19 of of ADP brj-24033 20 20 almost almost ADV brj-24033 20 21 all all PRON brj-24033 20 22 wood wood NOUN brj-24033 20 23 species specie NOUN brj-24033 20 24 classification classification NOUN brj-24033 20 25 systems system NOUN brj-24033 20 26 is be AUX brj-24033 20 27 usually usually ADV brj-24033 20 28 under under ADP brj-24033 20 29 100 100 NUM brj-24033 20 30 . . PUNCT brj-24033 21 1 it it PRON brj-24033 21 2 is be AUX brj-24033 21 3 appropriate appropriate ADJ brj-24033 21 4 to to PART brj-24033 21 5 study study VERB brj-24033 21 6 wood wood NOUN brj-24033 21 7 species specie NOUN brj-24033 21 8 classification classification NOUN brj-24033 21 9 systems system NOUN brj-24033 21 10 in in ADP brj-24033 21 11 an an DET brj-24033 21 12 open open ADJ brj-24033 21 13 set set NOUN brj-24033 21 14 scenario scenario NOUN brj-24033 21 15 , , PUNCT brj-24033 21 16 since since SCONJ brj-24033 21 17 it it PRON brj-24033 21 18 is be AUX brj-24033 21 19 very very ADV brj-24033 21 20 likely likely ADJ brj-24033 21 21 that that SCONJ brj-24033 21 22 some some DET brj-24033 21 23 wood wood NOUN brj-24033 21 24 samples sample NOUN brj-24033 21 25 from from ADP brj-24033 21 26 unknown unknown ADJ brj-24033 21 27 wood wood NOUN brj-24033 21 28 species specie NOUN brj-24033 21 29 will will AUX brj-24033 21 30 be be AUX brj-24033 21 31 encountered encounter VERB brj-24033 21 32 by by ADP brj-24033 21 33 these these DET brj-24033 21 34 systems system NOUN brj-24033 21 35 . . PUNCT brj-24033 22 1 open open ADJ brj-24033 22 2 set set VERB brj-24033 22 3 recognition recognition NOUN brj-24033 22 4 ( ( PUNCT brj-24033 22 5 osr osr NOUN brj-24033 22 6 ) ) PUNCT brj-24033 22 7 has have AUX brj-24033 22 8 been be AUX brj-24033 22 9 investigated investigate VERB brj-24033 22 10 for for ADP brj-24033 22 11 more more ADJ brj-24033 22 12 than than ADP brj-24033 22 13 10 10 NUM brj-24033 22 14 years year NOUN brj-24033 22 15 . . PUNCT brj-24033 23 1 it it PRON brj-24033 23 2 can can AUX brj-24033 23 3 not not PART brj-24033 23 4 only only ADV brj-24033 23 5 classify classify VERB brj-24033 23 6 the the DET brj-24033 23 7 known know VERB brj-24033 23 8 classes class NOUN brj-24033 23 9 in in ADP brj-24033 23 10 the the DET brj-24033 23 11 training training NOUN brj-24033 23 12 set set VERB brj-24033 23 13 correctly correctly ADV brj-24033 23 14 , , PUNCT brj-24033 23 15 but but CCONJ brj-24033 23 16 also also ADV brj-24033 23 17 reject reject VERB brj-24033 23 18 those those DET brj-24033 23 19 peer peer NOUN brj-24033 23 20 - - PUNCT brj-24033 23 21 reviewed review VERB brj-24033 23 22 article article NOUN brj-24033 23 23 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 23 24 zhang zhang PROPN brj-24033 23 25 & & CCONJ brj-24033 23 26 zhao zhao PROPN brj-24033 23 27 ( ( PUNCT brj-24033 23 28 2025 2025 NUM brj-24033 23 29 ) ) PUNCT brj-24033 23 30 . . PUNCT brj-24033 24 1 β€œ " PUNCT brj-24033 24 2 wood wood NOUN brj-24033 24 3 image image NOUN brj-24033 24 4 classification classification NOUN brj-24033 24 5 , , PUNCT brj-24033 24 6 ” " PUNCT brj-24033 24 7 bioresources bioresource NOUN brj-24033 24 8 20(1 20(1 NUM brj-24033 24 9 ) ) PUNCT brj-24033 24 10 , , PUNCT brj-24033 24 11 944 944 NUM brj-24033 24 12 - - SYM brj-24033 24 13 955 955 NUM brj-24033 24 14 . . PUNCT brj-24033 25 1 945 945 NUM brj-24033 25 2 unknown unknown ADJ brj-24033 25 3 classes class NOUN brj-24033 25 4 not not PART brj-24033 25 5 included include VERB brj-24033 25 6 in in ADP brj-24033 25 7 the the DET brj-24033 25 8 training training NOUN brj-24033 25 9 set set VERB brj-24033 25 10 effectively effectively ADV brj-24033 25 11 ( ( PUNCT brj-24033 25 12 geng geng PROPN brj-24033 25 13 et et PROPN brj-24033 25 14 al al PROPN brj-24033 25 15 . . PROPN brj-24033 25 16 2021 2021 NUM brj-24033 25 17 ) ) PUNCT brj-24033 25 18 . . PUNCT brj-24033 26 1 scheirer scheirer AUX brj-24033 26 2 et et PROPN brj-24033 26 3 al al PROPN brj-24033 26 4 . . PROPN brj-24033 27 1 ( ( PUNCT brj-24033 27 2 2013 2013 NUM brj-24033 27 3 ) ) PUNCT brj-24033 27 4 proposed propose VERB brj-24033 27 5 a a DET brj-24033 27 6 1 1 NUM brj-24033 27 7 - - PUNCT brj-24033 27 8 vs vs ADP brj-24033 27 9 - - PUNCT brj-24033 27 10 set set ADJ brj-24033 27 11 machine machine NOUN brj-24033 27 12 based base VERB brj-24033 27 13 on on ADP brj-24033 27 14 the the DET brj-24033 27 15 one one NUM brj-24033 27 16 - - PUNCT brj-24033 27 17 class class NOUN brj-24033 27 18 support support NOUN brj-24033 27 19 vector vector NOUN brj-24033 27 20 machine machine NOUN brj-24033 27 21 ( ( PUNCT brj-24033 27 22 svm svm PROPN brj-24033 27 23 ) ) PUNCT brj-24033 27 24 for for ADP brj-24033 27 25 osr osr PROPN brj-24033 27 26 application application NOUN brj-24033 27 27 in in ADP brj-24033 27 28 image image NOUN brj-24033 27 29 processing processing NOUN brj-24033 27 30 . . PUNCT brj-24033 28 1 a a DET brj-24033 28 2 compact compact ADJ brj-24033 28 3 abating abate VERB brj-24033 28 4 probability probability NOUN brj-24033 28 5 ( ( PUNCT brj-24033 28 6 cap cap NOUN brj-24033 28 7 ) ) PUNCT brj-24033 28 8 model model NOUN brj-24033 28 9 was be AUX brj-24033 28 10 further far ADV brj-24033 28 11 proposed propose VERB brj-24033 28 12 later later ADV brj-24033 28 13 ( ( PUNCT brj-24033 28 14 scheirer scheirer X brj-24033 28 15 et et PROPN brj-24033 28 16 al al PROPN brj-24033 28 17 . . PROPN brj-24033 28 18 2014 2014 NUM brj-24033 28 19 ) ) PUNCT brj-24033 28 20 to to PART brj-24033 28 21 combine combine VERB brj-24033 28 22 with with ADP brj-24033 28 23 a a DET brj-24033 28 24 statistical statistical ADJ brj-24033 28 25 extreme extreme ADJ brj-24033 28 26 value value NOUN brj-24033 28 27 theory theory NOUN brj-24033 28 28 ( ( PUNCT brj-24033 28 29 evt evt PROPN brj-24033 28 30 ) ) PUNCT brj-24033 28 31 . . PUNCT brj-24033 29 1 a a DET brj-24033 29 2 weibull weibull PROPN brj-24033 29 3 distribution distribution NOUN brj-24033 29 4 calibrated calibrate VERB brj-24033 29 5 svm svm NOUN brj-24033 29 6 ( ( PUNCT brj-24033 29 7 w w NOUN brj-24033 29 8 - - PUNCT brj-24033 29 9 svm svm NOUN brj-24033 29 10 ) ) PUNCT brj-24033 29 11 was be AUX brj-24033 29 12 proposed propose VERB brj-24033 29 13 to to PART brj-24033 29 14 further far ADV brj-24033 29 15 improve improve VERB brj-24033 29 16 the the DET brj-24033 29 17 osr osr PROPN brj-24033 29 18 classification classification NOUN brj-24033 29 19 accuracy accuracy NOUN brj-24033 29 20 . . PUNCT brj-24033 30 1 jain jain PROPN brj-24033 30 2 et et PROPN brj-24033 30 3 al al PROPN brj-24033 30 4 . . PROPN brj-24033 31 1 ( ( PUNCT brj-24033 31 2 2014 2014 NUM brj-24033 31 3 ) ) PUNCT brj-24033 31 4 proposed propose VERB brj-24033 31 5 a a DET brj-24033 31 6 pi pi NOUN brj-24033 31 7 - - PUNCT brj-24033 31 8 svm svm NOUN brj-24033 31 9 , , PUNCT brj-24033 31 10 which which PRON brj-24033 31 11 also also ADV brj-24033 31 12 uses use VERB brj-24033 31 13 the the DET brj-24033 31 14 evt evt PROPN brj-24033 31 15 to to PART brj-24033 31 16 model model VERB brj-24033 31 17 the the DET brj-24033 31 18 positive positive ADJ brj-24033 31 19 training training NOUN brj-24033 31 20 samples sample NOUN brj-24033 31 21 in in ADP brj-24033 31 22 a a DET brj-24033 31 23 decision decision NOUN brj-24033 31 24 boundary boundary NOUN brj-24033 31 25 . . PUNCT brj-24033 32 1 except except SCONJ brj-24033 32 2 for for ADP brj-24033 32 3 the the DET brj-24033 32 4 osr osr PROPN brj-24033 32 5 schemes scheme NOUN brj-24033 32 6 based base VERB brj-24033 32 7 on on ADP brj-24033 32 8 svm svm PROPN brj-24033 32 9 , , PUNCT brj-24033 32 10 some some DET brj-24033 32 11 other other ADJ brj-24033 32 12 machine machine NOUN brj-24033 32 13 learning learning NOUN brj-24033 32 14 based base VERB brj-24033 32 15 osr osr PROPN brj-24033 32 16 schemes scheme NOUN brj-24033 32 17 have have AUX brj-24033 32 18 also also ADV brj-24033 32 19 been be AUX brj-24033 32 20 proposed propose VERB brj-24033 32 21 . . PUNCT brj-24033 33 1 for for ADP brj-24033 33 2 instance instance NOUN brj-24033 33 3 , , PUNCT brj-24033 33 4 zhang zhang PROPN brj-24033 33 5 and and CCONJ brj-24033 33 6 patel patel PROPN brj-24033 33 7 ( ( PUNCT brj-24033 33 8 2017 2017 NUM brj-24033 33 9 ) ) PUNCT brj-24033 33 10 proposed propose VERB brj-24033 33 11 a a DET brj-24033 33 12 sparse sparse ADJ brj-24033 33 13 representation representation NOUN brj-24033 33 14 based base VERB brj-24033 33 15 osr osr PROPN brj-24033 33 16 scheme scheme NOUN brj-24033 33 17 , , PUNCT brj-24033 33 18 which which PRON brj-24033 33 19 uses use VERB brj-24033 33 20 the the DET brj-24033 33 21 evt evt PROPN brj-24033 33 22 to to PART brj-24033 33 23 model model VERB brj-24033 33 24 the the DET brj-24033 33 25 tail tail NOUN brj-24033 33 26 distribution distribution NOUN brj-24033 33 27 of of ADP brj-24033 33 28 reconstruction reconstruction NOUN brj-24033 33 29 loss loss NOUN brj-24033 33 30 . . PUNCT brj-24033 34 1 junior junior ADV brj-24033 34 2 et et PROPN brj-24033 34 3 al al PROPN brj-24033 34 4 . . PROPN brj-24033 35 1 ( ( PUNCT brj-24033 35 2 2017 2017 NUM brj-24033 35 3 ) ) PUNCT brj-24033 35 4 proposed propose VERB brj-24033 35 5 an an DET brj-24033 35 6 open open ADJ brj-24033 35 7 set set VERB brj-24033 35 8 version version NOUN brj-24033 35 9 of of ADP brj-24033 35 10 nearest near ADJ brj-24033 35 11 neighbor neighbor NOUN brj-24033 35 12 classifier classifier NOUN brj-24033 35 13 ( ( PUNCT brj-24033 35 14 osnn osnn NOUN brj-24033 35 15 ) ) PUNCT brj-24033 35 16 . . PUNCT brj-24033 36 1 in in ADP brj-24033 36 2 this this DET brj-24033 36 3 scheme scheme NOUN brj-24033 36 4 , , PUNCT brj-24033 36 5 the the DET brj-24033 36 6 distances distance NOUN brj-24033 36 7 between between ADP brj-24033 36 8 a a DET brj-24033 36 9 detected detect VERB brj-24033 36 10 sample sample NOUN brj-24033 36 11 s s NOUN brj-24033 36 12 and and CCONJ brj-24033 36 13 the the DET brj-24033 36 14 two two NUM brj-24033 36 15 nearest near ADJ brj-24033 36 16 neighbor neighbor NOUN brj-24033 36 17 samples sample NOUN brj-24033 36 18 t t PROPN brj-24033 36 19 , , PUNCT brj-24033 36 20 u u NOUN brj-24033 36 21 from from ADP brj-24033 36 22 two two NUM brj-24033 36 23 different different ADJ brj-24033 36 24 classes class NOUN brj-24033 36 25 are be AUX brj-24033 36 26 calculated calculate VERB brj-24033 36 27 and and CCONJ brj-24033 36 28 the the DET brj-24033 36 29 ratio ratio NOUN brj-24033 36 30 is be AUX brj-24033 36 31 computed compute VERB brj-24033 36 32 as as ADP brj-24033 36 33 π‘…π‘Žπ‘‘π‘–π‘œ π‘…π‘Žπ‘‘π‘–π‘œ PROPN brj-24033 36 34 = = SYM brj-24033 36 35 𝑑(𝒔 𝑑(𝒔 PROPN brj-24033 36 36 , , PUNCT brj-24033 36 37 𝒕)/𝑑(𝒔 𝒕)/𝑑(𝒔 PROPN brj-24033 36 38 , , PUNCT brj-24033 36 39 𝒖 𝒖 NOUN brj-24033 36 40 ) ) PUNCT brj-24033 36 41 . . PUNCT brj-24033 37 1 if if SCONJ brj-24033 37 2 π‘…π‘Žπ‘‘π‘–π‘œ π‘…π‘Žπ‘‘π‘–π‘œ PROPN brj-24033 37 3 ≀ ≀ ADJ brj-24033 37 4 𝑇𝑅 𝑇𝑅 PROPN brj-24033 37 5 , , PUNCT brj-24033 37 6 then then ADV brj-24033 37 7 s s VERB brj-24033 37 8 is be AUX brj-24033 37 9 classified classify VERB brj-24033 37 10 into into ADP brj-24033 37 11 the the DET brj-24033 37 12 class class NOUN brj-24033 37 13 which which PRON brj-24033 37 14 includes include VERB brj-24033 37 15 the the DET brj-24033 37 16 sample sample NOUN brj-24033 37 17 t t PROPN brj-24033 37 18 ; ; PUNCT brj-24033 37 19 otherwise otherwise ADV brj-24033 37 20 , , PUNCT brj-24033 37 21 s s VERB brj-24033 37 22 is be AUX brj-24033 37 23 rejected reject VERB brj-24033 37 24 as as ADP brj-24033 37 25 an an DET brj-24033 37 26 unknown unknown ADJ brj-24033 37 27 class class NOUN brj-24033 37 28 . . PUNCT brj-24033 38 1 moreover moreover ADV brj-24033 38 2 , , PUNCT brj-24033 38 3 some some DET brj-24033 38 4 deep deep ADJ brj-24033 38 5 learning learning NOUN brj-24033 38 6 based base VERB brj-24033 38 7 osr osr PROPN brj-24033 38 8 schemes scheme NOUN brj-24033 38 9 have have AUX brj-24033 38 10 been be AUX brj-24033 38 11 proposed propose VERB brj-24033 38 12 such such ADJ brj-24033 38 13 as as ADP brj-24033 38 14 openmax openmax VERB brj-24033 38 15 neural neural ADJ brj-24033 38 16 network network NOUN brj-24033 38 17 ( ( PUNCT brj-24033 38 18 bendale bendale NOUN brj-24033 38 19 and and CCONJ brj-24033 38 20 boult boult NOUN brj-24033 38 21 , , PUNCT brj-24033 38 22 2016 2016 NUM brj-24033 38 23 ) ) PUNCT brj-24033 38 24 , , PUNCT brj-24033 38 25 where where SCONJ brj-24033 38 26 the the DET brj-24033 38 27 softmax softmax NOUN brj-24033 38 28 layer layer NOUN brj-24033 38 29 is be AUX brj-24033 38 30 replaced replace VERB brj-24033 38 31 by by ADP brj-24033 38 32 openmax openmax ADJ brj-24033 38 33 layer layer NOUN brj-24033 38 34 to to PART brj-24033 38 35 modify modify VERB brj-24033 38 36 the the DET brj-24033 38 37 membership membership NOUN brj-24033 38 38 probability probability NOUN brj-24033 38 39 of of ADP brj-24033 38 40 known known ADJ brj-24033 38 41 and and CCONJ brj-24033 38 42 unknown unknown ADJ brj-24033 38 43 classes class NOUN brj-24033 38 44 . . PUNCT brj-24033 39 1 bendale bendale NOUN brj-24033 39 2 and and CCONJ brj-24033 39 3 boult boult NOUN brj-24033 39 4 ( ( PUNCT brj-24033 39 5 2015 2015 NUM brj-24033 39 6 ) ) PUNCT brj-24033 39 7 extended extend VERB brj-24033 39 8 the the DET brj-24033 39 9 nearest near ADJ brj-24033 39 10 class class NOUN brj-24033 39 11 mean mean NOUN brj-24033 39 12 ( ( PUNCT brj-24033 39 13 ncm ncm NOUN brj-24033 39 14 ) ) PUNCT brj-24033 39 15 classifier classifier NOUN brj-24033 39 16 and and CCONJ brj-24033 39 17 propose propose VERB brj-24033 39 18 the the DET brj-24033 39 19 nearest near ADJ brj-24033 39 20 non non ADJ brj-24033 39 21 - - ADJ brj-24033 39 22 outlier outlier ADJ brj-24033 39 23 ( ( PUNCT brj-24033 39 24 nno nno NOUN brj-24033 39 25 ) ) PUNCT brj-24033 39 26 classifier classifier NOUN brj-24033 39 27 for for ADP brj-24033 39 28 osr osr NOUN brj-24033 39 29 use use NOUN brj-24033 39 30 . . PUNCT brj-24033 40 1 the the DET brj-24033 40 2 distances distance NOUN brj-24033 40 3 between between ADP brj-24033 40 4 a a DET brj-24033 40 5 detected detect VERB brj-24033 40 6 sample sample NOUN brj-24033 40 7 and and CCONJ brj-24033 40 8 each each DET brj-24033 40 9 known know VERB brj-24033 40 10 class class NOUN brj-24033 40 11 mean mean NOUN brj-24033 40 12 are be AUX brj-24033 40 13 computed compute VERB brj-24033 40 14 to to PART brj-24033 40 15 classify classify VERB brj-24033 40 16 the the DET brj-24033 40 17 known know VERB brj-24033 40 18 classes class NOUN brj-24033 40 19 and and CCONJ brj-24033 40 20 reject reject VERB brj-24033 40 21 the the DET brj-24033 40 22 unknown unknown ADJ brj-24033 40 23 classes class NOUN brj-24033 40 24 . . PUNCT brj-24033 41 1 moreover moreover ADV brj-24033 41 2 , , PUNCT brj-24033 41 3 a a DET brj-24033 41 4 metric metric ADJ brj-24033 41 5 learning learning NOUN brj-24033 41 6 ( ( PUNCT brj-24033 41 7 ml ml NOUN brj-24033 41 8 ) ) PUNCT brj-24033 41 9 algorithm algorithm NOUN brj-24033 41 10 ( ( PUNCT brj-24033 41 11 mensink mensink NOUN brj-24033 41 12 et et PROPN brj-24033 41 13 al al PROPN brj-24033 41 14 . . PROPN brj-24033 41 15 2013 2013 NUM brj-24033 41 16 ) ) PUNCT brj-24033 41 17 is be AUX brj-24033 41 18 used use VERB brj-24033 41 19 for for ADP brj-24033 41 20 feature feature NOUN brj-24033 41 21 dimension dimension NOUN brj-24033 41 22 reduction reduction NOUN brj-24033 41 23 . . PUNCT brj-24033 42 1 when when SCONJ brj-24033 42 2 a a DET brj-24033 42 3 new new ADJ brj-24033 42 4 class class NOUN brj-24033 42 5 is be AUX brj-24033 42 6 added add VERB brj-24033 42 7 into into ADP brj-24033 42 8 the the DET brj-24033 42 9 training training NOUN brj-24033 42 10 set set NOUN brj-24033 42 11 of of ADP brj-24033 42 12 the the DET brj-24033 42 13 classification classification NOUN brj-24033 42 14 system system NOUN brj-24033 42 15 , , PUNCT brj-24033 42 16 the the DET brj-24033 42 17 previous previous ADJ brj-24033 42 18 projection projection NOUN brj-24033 42 19 matrix matrix NOUN brj-24033 42 20 w w NOUN brj-24033 42 21 can can AUX brj-24033 42 22 still still ADV brj-24033 42 23 be be AUX brj-24033 42 24 used use VERB brj-24033 42 25 with with ADP brj-24033 42 26 near near ADV brj-24033 42 27 - - PUNCT brj-24033 42 28 zero zero NUM brj-24033 42 29 errors error NOUN brj-24033 42 30 , , PUNCT brj-24033 42 31 as as SCONJ brj-24033 42 32 pointed point VERB brj-24033 42 33 out out ADP brj-24033 42 34 by by ADP brj-24033 42 35 mensink mensink PROPN brj-24033 42 36 et et PROPN brj-24033 42 37 al al PROPN brj-24033 42 38 . . PROPN brj-24033 43 1 ( ( PUNCT brj-24033 43 2 2013 2013 NUM brj-24033 43 3 ) ) PUNCT brj-24033 43 4 . . PUNCT brj-24033 44 1 however however ADV brj-24033 44 2 , , PUNCT brj-24033 44 3 this this DET brj-24033 44 4 nno nno NOUN brj-24033 44 5 classifier classifier NOUN brj-24033 44 6 can can AUX brj-24033 44 7 achieve achieve VERB brj-24033 44 8 a a DET brj-24033 44 9 satisfied satisfied ADJ brj-24033 44 10 classification classification NOUN brj-24033 44 11 accuracy accuracy NOUN brj-24033 44 12 in in ADP brj-24033 44 13 open open ADJ brj-24033 44 14 set set VERB brj-24033 44 15 scenario scenario NOUN brj-24033 44 16 only only ADV brj-24033 44 17 when when SCONJ brj-24033 44 18 each each DET brj-24033 44 19 known know VERB brj-24033 44 20 class class NOUN brj-24033 44 21 is be AUX brj-24033 44 22 represented represent VERB brj-24033 44 23 by by ADP brj-24033 44 24 a a DET brj-24033 44 25 similar similar ADJ brj-24033 44 26 sphere sphere NOUN brj-24033 44 27 distribution distribution NOUN brj-24033 44 28 ( ( PUNCT brj-24033 44 29 i.e. i.e. X brj-24033 44 30 , , PUNCT brj-24033 44 31 with with ADP brj-24033 44 32 a a DET brj-24033 44 33 same same ADJ brj-24033 44 34 sphere sphere NOUN brj-24033 44 35 radius radius NOUN brj-24033 44 36 approximately approximately ADV brj-24033 44 37 ) ) PUNCT brj-24033 44 38 . . PUNCT brj-24033 45 1 in in ADP brj-24033 45 2 practice practice NOUN brj-24033 45 3 , , PUNCT brj-24033 45 4 this this DET brj-24033 45 5 strict strict ADJ brj-24033 45 6 constraint constraint NOUN brj-24033 45 7 is be AUX brj-24033 45 8 hard hard ADJ brj-24033 45 9 to to PART brj-24033 45 10 satisfy satisfy VERB brj-24033 45 11 . . PUNCT brj-24033 46 1 in in ADP brj-24033 46 2 this this DET brj-24033 46 3 article article NOUN brj-24033 46 4 , , PUNCT brj-24033 46 5 this this DET brj-24033 46 6 nno nno NOUN brj-24033 46 7 classifier classifier NOUN brj-24033 46 8 was be AUX brj-24033 46 9 improved improve VERB brj-24033 46 10 so so SCONJ brj-24033 46 11 that that SCONJ brj-24033 46 12 it it PRON brj-24033 46 13 can can AUX brj-24033 46 14 be be AUX brj-24033 46 15 used use VERB brj-24033 46 16 for for ADP brj-24033 46 17 known know VERB brj-24033 46 18 classes class NOUN brj-24033 46 19 with with ADP brj-24033 46 20 different different ADJ brj-24033 46 21 shape shape NOUN brj-24033 46 22 distributions distribution NOUN brj-24033 46 23 ( ( PUNCT brj-24033 46 24 e.g. e.g. ADV brj-24033 46 25 , , PUNCT brj-24033 46 26 sphere sphere NOUN brj-24033 46 27 structure structure NOUN brj-24033 46 28 or or CCONJ brj-24033 46 29 manifold manifold ADJ brj-24033 46 30 structure structure NOUN brj-24033 46 31 ) ) PUNCT brj-24033 46 32 in in ADP brj-24033 46 33 osr osr PROPN brj-24033 46 34 use use NOUN brj-24033 46 35 . . PUNCT brj-24033 47 1 specifically specifically ADV brj-24033 47 2 , , PUNCT brj-24033 47 3 every every DET brj-24033 47 4 known know VERB brj-24033 47 5 class class NOUN brj-24033 47 6 is be AUX brj-24033 47 7 further far ADV brj-24033 47 8 processed process VERB brj-24033 47 9 by by ADP brj-24033 47 10 an an DET brj-24033 47 11 automatic automatic ADJ brj-24033 47 12 cluster cluster NOUN brj-24033 47 13 analysis analysis NOUN brj-24033 47 14 to to PART brj-24033 47 15 get get VERB brj-24033 47 16 1 1 NUM brj-24033 47 17 to to PART brj-24033 47 18 3 3 NUM brj-24033 47 19 clusters cluster NOUN brj-24033 47 20 with with ADP brj-24033 47 21 a a DET brj-24033 47 22 sphere sphere NOUN brj-24033 47 23 structure structure NOUN brj-24033 47 24 approximately approximately ADV brj-24033 47 25 . . PUNCT brj-24033 48 1 in in ADP brj-24033 48 2 a a DET brj-24033 48 3 known know VERB brj-24033 48 4 class class NOUN brj-24033 48 5 , , PUNCT brj-24033 48 6 each each DET brj-24033 48 7 cluster cluster NOUN brj-24033 48 8 usually usually ADV brj-24033 48 9 has have VERB brj-24033 48 10 a a DET brj-24033 48 11 different different ADJ brj-24033 48 12 size size NOUN brj-24033 48 13 . . PUNCT brj-24033 49 1 then then ADV brj-24033 49 2 the the DET brj-24033 49 3 distances distance NOUN brj-24033 49 4 between between ADP brj-24033 49 5 a a DET brj-24033 49 6 detected detect VERB brj-24033 49 7 sample sample NOUN brj-24033 49 8 and and CCONJ brj-24033 49 9 each each DET brj-24033 49 10 cluster cluster NOUN brj-24033 49 11 of of ADP brj-24033 49 12 one one NUM brj-24033 49 13 known know VERB brj-24033 49 14 class class NOUN brj-24033 49 15 are be AUX brj-24033 49 16 computed compute VERB brj-24033 49 17 to to PART brj-24033 49 18 get get VERB brj-24033 49 19 membership membership NOUN brj-24033 49 20 probability probability NOUN brj-24033 49 21 of of ADP brj-24033 49 22 known know VERB brj-24033 49 23 classes class NOUN brj-24033 49 24 . . PUNCT brj-24033 50 1 moreover moreover ADV brj-24033 50 2 , , PUNCT brj-24033 50 3 each each DET brj-24033 50 4 cluster cluster NOUN brj-24033 50 5 ’s ’s PART brj-24033 50 6 size size NOUN brj-24033 50 7 threshold threshold NOUN brj-24033 50 8 is be AUX brj-24033 50 9 computed compute VERB brj-24033 50 10 by by ADP brj-24033 50 11 an an DET brj-24033 50 12 optimal optimal ADJ brj-24033 50 13 strategy strategy NOUN brj-24033 50 14 . . PUNCT brj-24033 51 1 in in ADP brj-24033 51 2 this this DET brj-24033 51 3 way way NOUN brj-24033 51 4 , , PUNCT brj-24033 51 5 the the DET brj-24033 51 6 proposed propose VERB brj-24033 51 7 improved improve VERB brj-24033 51 8 nno nno NOUN brj-24033 51 9 classifier classifier NOUN brj-24033 51 10 can can AUX brj-24033 51 11 achieve achieve VERB brj-24033 51 12 more more ADV brj-24033 51 13 accurate accurate ADJ brj-24033 51 14 classification classification NOUN brj-24033 51 15 results result NOUN brj-24033 51 16 in in ADP brj-24033 51 17 open open ADJ brj-24033 51 18 set set VERB brj-24033 51 19 in in ADP brj-24033 51 20 practice practice NOUN brj-24033 51 21 for for ADP brj-24033 51 22 known know VERB brj-24033 51 23 classes class NOUN brj-24033 51 24 with with ADP brj-24033 51 25 different different ADJ brj-24033 51 26 shape shape NOUN brj-24033 51 27 distributions distribution NOUN brj-24033 51 28 . . PUNCT brj-24033 52 1 as as ADP brj-24033 52 2 for for ADP brj-24033 52 3 the the DET brj-24033 52 4 classification classification NOUN brj-24033 52 5 feature feature NOUN brj-24033 52 6 , , PUNCT brj-24033 52 7 the the DET brj-24033 52 8 nir nir ADJ brj-24033 52 9 spectral spectral ADJ brj-24033 52 10 curve curve NOUN brj-24033 52 11 is be AUX brj-24033 52 12 used use VERB brj-24033 52 13 here here ADV brj-24033 52 14 , , PUNCT brj-24033 52 15 since since SCONJ brj-24033 52 16 it it PRON brj-24033 52 17 has have VERB brj-24033 52 18 the the DET brj-24033 52 19 advantages advantage NOUN brj-24033 52 20 of of ADP brj-24033 52 21 fast fast ADJ brj-24033 52 22 speed speed NOUN brj-24033 52 23 , , PUNCT brj-24033 52 24 high high ADJ brj-24033 52 25 accuracy accuracy NOUN brj-24033 52 26 , , PUNCT brj-24033 52 27 and and CCONJ brj-24033 52 28 non non ADJ brj-24033 52 29 - - ADJ brj-24033 52 30 destructive destructive ADJ brj-24033 52 31 testing testing NOUN brj-24033 52 32 ( ( PUNCT brj-24033 52 33 ma ma PROPN brj-24033 52 34 et et PROPN brj-24033 52 35 al al PROPN brj-24033 52 36 . . PROPN brj-24033 52 37 2021 2021 NUM brj-24033 52 38 ; ; PUNCT brj-24033 52 39 park park NOUN brj-24033 52 40 et et PROPN brj-24033 52 41 al al PROPN brj-24033 52 42 . . PROPN brj-24033 52 43 2021 2021 NUM brj-24033 52 44 ; ; PUNCT brj-24033 52 45 tuncer tuncer NOUN brj-24033 52 46 et et PROPN brj-24033 52 47 al al PROPN brj-24033 52 48 . . PROPN brj-24033 52 49 2021 2021 NUM brj-24033 52 50 ) ) PUNCT brj-24033 52 51 . . PUNCT brj-24033 53 1 experimental experimental ADJ brj-24033 53 2 there there PRON brj-24033 53 3 were be VERB brj-24033 53 4 in in ADP brj-24033 53 5 total total ADJ brj-24033 53 6 35 35 NUM brj-24033 53 7 wood wood NOUN brj-24033 53 8 species specie NOUN brj-24033 53 9 used use VERB brj-24033 53 10 in in ADP brj-24033 53 11 the the DET brj-24033 53 12 experimental experimental ADJ brj-24033 53 13 wood wood NOUN brj-24033 53 14 dataset dataset NOUN brj-24033 53 15 , , PUNCT brj-24033 53 16 which which PRON brj-24033 53 17 included include VERB brj-24033 53 18 both both CCONJ brj-24033 53 19 broadleaved broadleave VERB brj-24033 53 20 and and CCONJ brj-24033 53 21 coniferous coniferous ADJ brj-24033 53 22 tree tree NOUN brj-24033 53 23 species specie NOUN brj-24033 53 24 . . PUNCT brj-24033 54 1 the the DET brj-24033 54 2 wood wood NOUN brj-24033 54 3 dataset dataset NOUN brj-24033 54 4 contained contain VERB brj-24033 54 5 some some DET brj-24033 54 6 similar similar ADJ brj-24033 54 7 wood wood NOUN brj-24033 54 8 species specie NOUN brj-24033 54 9 with with ADP brj-24033 54 10 similar similar ADJ brj-24033 54 11 colors color NOUN brj-24033 54 12 and and CCONJ brj-24033 54 13 textures texture NOUN brj-24033 54 14 or or CCONJ brj-24033 54 15 within within ADP brj-24033 54 16 the the DET brj-24033 54 17 same same ADJ brj-24033 54 18 genus genus NOUN brj-24033 54 19 . . PUNCT brj-24033 55 1 the the DET brj-24033 55 2 specific specific ADJ brj-24033 55 3 tree tree NOUN brj-24033 55 4 species specie NOUN brj-24033 55 5 information information NOUN brj-24033 55 6 is be AUX brj-24033 55 7 illustrated illustrate VERB brj-24033 55 8 in in ADP brj-24033 55 9 table table NOUN brj-24033 55 10 1 1 NUM brj-24033 55 11 . . PUNCT brj-24033 56 1 the the DET brj-24033 56 2 cross cross NOUN brj-24033 56 3 sections section NOUN brj-24033 56 4 of of ADP brj-24033 56 5 these these DET brj-24033 56 6 wood wood NOUN brj-24033 56 7 peer peer NOUN brj-24033 56 8 - - PUNCT brj-24033 56 9 reviewed review VERB brj-24033 56 10 article article NOUN brj-24033 56 11 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 56 12 zhang zhang PROPN brj-24033 56 13 & & CCONJ brj-24033 56 14 zhao zhao PROPN brj-24033 56 15 ( ( PUNCT brj-24033 56 16 2025 2025 NUM brj-24033 56 17 ) ) PUNCT brj-24033 56 18 . . PUNCT brj-24033 57 1 β€œ " PUNCT brj-24033 57 2 wood wood NOUN brj-24033 57 3 image image NOUN brj-24033 57 4 classification classification NOUN brj-24033 57 5 , , PUNCT brj-24033 57 6 ” " PUNCT brj-24033 57 7 bioresources bioresource NOUN brj-24033 57 8 20(1 20(1 NUM brj-24033 57 9 ) ) PUNCT brj-24033 57 10 , , PUNCT brj-24033 57 11 944 944 NUM brj-24033 57 12 - - SYM brj-24033 57 13 955 955 NUM brj-24033 57 14 . . PUNCT brj-24033 58 1 946 946 NUM brj-24033 58 2 species specie NOUN brj-24033 58 3 were be AUX brj-24033 58 4 used use VERB brj-24033 58 5 for for ADP brj-24033 58 6 spectral spectral ADJ brj-24033 58 7 acquisition acquisition NOUN brj-24033 58 8 and and CCONJ brj-24033 58 9 are be AUX brj-24033 58 10 illustrated illustrate VERB brj-24033 58 11 in in ADP brj-24033 58 12 fig fig NOUN brj-24033 58 13 . . PUNCT brj-24033 59 1 1 1 X brj-24033 59 2 . . X brj-24033 60 1 the the DET brj-24033 60 2 ocean ocean NOUN brj-24033 60 3 optics optic VERB brj-24033 60 4 flame flame NOUN brj-24033 60 5 - - PUNCT brj-24033 60 6 nir nir NOUN brj-24033 60 7 micro micro ADJ brj-24033 60 8 spectrometer spectrometer NOUN brj-24033 61 1 was be AUX brj-24033 61 2 used use VERB brj-24033 61 3 to to PART brj-24033 61 4 pick pick VERB brj-24033 61 5 up up ADP brj-24033 61 6 the the DET brj-24033 61 7 nir nir ADJ brj-24033 61 8 spectral spectral ADJ brj-24033 61 9 curves curve NOUN brj-24033 61 10 . . PUNCT brj-24033 62 1 the the DET brj-24033 62 2 effective effective ADJ brj-24033 62 3 wavelength wavelength NOUN brj-24033 62 4 band band NOUN brj-24033 62 5 was be AUX brj-24033 62 6 950 950 NUM brj-24033 62 7 to to PART brj-24033 62 8 1650 1650 NUM brj-24033 62 9 nm nm NOUN brj-24033 62 10 , , PUNCT brj-24033 62 11 with with ADP brj-24033 62 12 a a DET brj-24033 62 13 wavelength wavelength NOUN brj-24033 62 14 resolution resolution NOUN brj-24033 62 15 of of ADP brj-24033 62 16 5.4 5.4 NUM brj-24033 62 17 nm nm NOUN brj-24033 62 18 , , PUNCT brj-24033 62 19 respectively respectively ADV brj-24033 62 20 . . PUNCT brj-24033 63 1 each each DET brj-24033 63 2 wood wood NOUN brj-24033 63 3 species specie NOUN brj-24033 63 4 consisted consist VERB brj-24033 63 5 of of ADP brj-24033 63 6 50 50 NUM brj-24033 63 7 samples sample NOUN brj-24033 63 8 ( ( PUNCT brj-24033 63 9 i.e. i.e. X brj-24033 63 10 , , PUNCT brj-24033 63 11 nir nir ADJ brj-24033 63 12 spectral spectral ADJ brj-24033 63 13 curves curve NOUN brj-24033 63 14 ) ) PUNCT brj-24033 63 15 so so SCONJ brj-24033 63 16 that that SCONJ brj-24033 63 17 there there PRON brj-24033 63 18 were be VERB brj-24033 63 19 a a DET brj-24033 63 20 total total NOUN brj-24033 63 21 of of ADP brj-24033 63 22 1750 1750 NUM brj-24033 63 23 samples sample NOUN brj-24033 63 24 . . PUNCT brj-24033 64 1 each each DET brj-24033 64 2 spectral spectral ADJ brj-24033 64 3 curve curve NOUN brj-24033 64 4 was be AUX brj-24033 64 5 represented represent VERB brj-24033 64 6 by by ADP brj-24033 64 7 a a DET brj-24033 64 8 128dimensional 128dimensional PROPN brj-24033 64 9 ( ( PUNCT brj-24033 64 10 128d 128d NOUN brj-24033 64 11 ) ) PUNCT brj-24033 64 12 vector vector NOUN brj-24033 64 13 . . PUNCT brj-24033 64 14 table table NOUN brj-24033 64 15 1 1 NUM brj-24033 64 16 . . PUNCT brj-24033 65 1 information information NOUN brj-24033 65 2 on on ADP brj-24033 65 3 experimental experimental ADJ brj-24033 65 4 samples sample NOUN brj-24033 65 5 no no INTJ brj-24033 65 6 . . PUNCT brj-24033 66 1 genus genus ADJ brj-24033 66 2 species species PROPN brj-24033 66 3 1 1 NUM brj-24033 66 4 acer acer NOUN brj-24033 66 5 davidii davidii NOUN brj-24033 66 6 2 2 NUM brj-24033 66 7 amygdalus amygdalus ADJ brj-24033 66 8 davidiana davidiana PROPN brj-24033 66 9 3 3 NUM brj-24033 66 10 aucoumea aucoumea PROPN brj-24033 66 11 klaineana klaineana VERB brj-24033 66 12 4 4 NUM brj-24033 66 13 betula betula ADJ brj-24033 66 14 alnoides alnoide NOUN brj-24033 66 15 5 5 NUM brj-24033 66 16 betula betula ADJ brj-24033 66 17 platyphylla platyphylla NOUN brj-24033 66 18 6 6 NUM brj-24033 66 19 calophyllum calophyllum PROPN brj-24033 66 20 inophyllum inophyllum VERB brj-24033 66 21 7 7 NUM brj-24033 66 22 chamaecyparis chamaecyparis NOUN brj-24033 66 23 nootkatensis nootkatensis NOUN brj-24033 66 24 8 8 NUM brj-24033 66 25 cinnamomum cinnamomum ADJ brj-24033 66 26 camphora camphora NOUN brj-24033 66 27 9 9 NUM brj-24033 66 28 cyclobalanopsis cyclobalanopsis NOUN brj-24033 66 29 glauca glauca NOUN brj-24033 66 30 10 10 NUM brj-24033 66 31 dipterocarpus dipterocarpu NOUN brj-24033 66 32 alatus alatus NOUN brj-24033 66 33 11 11 NUM brj-24033 66 34 entandrophragma entandrophragma NOUN brj-24033 66 35 candollei candollei VERB brj-24033 66 36 12 12 NUM brj-24033 66 37 fraxinus fraxinus NOUN brj-24033 66 38 mandshurica mandshurica PROPN brj-24033 66 39 13 13 NUM brj-24033 66 40 fraxinus fraxinus PROPN brj-24033 66 41 chinensis chinensis NOUN brj-24033 66 42 14 14 NUM brj-24033 66 43 guibourtia guibourtia NOUN brj-24033 66 44 demeusei demeusei VERB brj-24033 66 45 15 15 NUM brj-24033 66 46 guibourtia guibourtia NOUN brj-24033 66 47 ehie ehie VERB brj-24033 66 48 16 16 NUM brj-24033 66 49 intsia intsia NOUN brj-24033 66 50 bijuga bijuga ADP brj-24033 66 51 17 17 NUM brj-24033 66 52 juglans juglan NOUN brj-24033 66 53 mandshurica mandshurica VERB brj-24033 66 54 18 18 NUM brj-24033 66 55 juglans juglan NOUN brj-24033 66 56 nigra nigra PROPN brj-24033 66 57 19 19 NUM brj-24033 66 58 larix larix NOUN brj-24033 66 59 gmelinii gmelinii ADJ brj-24033 66 60 20 20 NUM brj-24033 66 61 magnolia magnolia NOUN brj-24033 66 62 fordiana fordiana PROPN brj-24033 66 63 21 21 NUM brj-24033 66 64 millettia millettia NOUN brj-24033 66 65 laurentii laurentii VERB brj-24033 66 66 22 22 NUM brj-24033 66 67 picea picea NOUN brj-24033 66 68 asperata asperata NOUN brj-24033 66 69 23 23 NUM brj-24033 66 70 pinus pinus NOUN brj-24033 66 71 radiata radiata NOUN brj-24033 66 72 24 24 NUM brj-24033 66 73 pinus pinus NOUN brj-24033 66 74 koraiensis koraiensis NOUN brj-24033 66 75 25 25 NUM brj-24033 66 76 pinus pinus NOUN brj-24033 66 77 massoniana massoniana NOUN brj-24033 66 78 26 26 NUM brj-24033 66 79 shorea shorea PROPN brj-24033 66 80 contorta contorta PROPN brj-24033 66 81 27 27 NUM brj-24033 66 82 shorea shorea NOUN brj-24033 66 83 laevis laevis ADJ brj-24033 66 84 28 28 NUM brj-24033 66 85 sophora sophora PROPN brj-24033 66 86 japonica japonica NOUN brj-24033 66 87 29 29 NUM brj-24033 66 88 swietenia swietenia NOUN brj-24033 66 89 mahagoni mahagoni VERB brj-24033 66 90 30 30 NUM brj-24033 66 91 tectona tectona NOUN brj-24033 66 92 grandis grandis NOUN brj-24033 66 93 31 31 NUM brj-24033 66 94 terminalia terminalia PROPN brj-24033 66 95 catappa catappa NOUN brj-24033 66 96 32 32 NUM brj-24033 66 97 tilia tilia NOUN brj-24033 66 98 mandshurica mandshurica PROPN brj-24033 66 99 33 33 NUM brj-24033 66 100 toona toona PROPN brj-24033 66 101 ciliata ciliata NOUN brj-24033 66 102 34 34 NUM brj-24033 66 103 ulmus ulmus PROPN brj-24033 66 104 glabra glabra PROPN brj-24033 66 105 35 35 NUM brj-24033 66 106 vernicia vernicia NOUN brj-24033 66 107 fordii fordii NOUN brj-24033 66 108 before before ADP brj-24033 66 109 the the DET brj-24033 66 110 wood wood NOUN brj-24033 66 111 spectral spectral ADJ brj-24033 66 112 collection collection NOUN brj-24033 66 113 , , PUNCT brj-24033 66 114 the the DET brj-24033 66 115 wood wood NOUN brj-24033 66 116 sample sample NOUN brj-24033 66 117 pre pre ADJ brj-24033 66 118 - - ADJ brj-24033 66 119 processing processing NOUN brj-24033 66 120 was be AUX brj-24033 66 121 performed perform VERB brj-24033 66 122 . . PUNCT brj-24033 67 1 first first ADV brj-24033 67 2 , , PUNCT brj-24033 67 3 25 25 NUM brj-24033 67 4 wood wood NOUN brj-24033 67 5 blocks block NOUN brj-24033 67 6 from from ADP brj-24033 67 7 different different ADJ brj-24033 67 8 trees tree NOUN brj-24033 67 9 were be AUX brj-24033 67 10 selected select VERB brj-24033 67 11 for for ADP brj-24033 67 12 every every DET brj-24033 67 13 tree tree NOUN brj-24033 67 14 species specie NOUN brj-24033 67 15 . . PUNCT brj-24033 68 1 these these DET brj-24033 68 2 25 25 NUM brj-24033 68 3 wood wood NOUN brj-24033 68 4 blocks block NOUN brj-24033 68 5 are be AUX brj-24033 68 6 then then ADV brj-24033 68 7 cut cut VERB brj-24033 68 8 into into ADP brj-24033 68 9 small small ADJ brj-24033 68 10 wood wood NOUN brj-24033 68 11 samples sample NOUN brj-24033 68 12 with with ADP brj-24033 68 13 size size NOUN brj-24033 68 14 of of ADP brj-24033 68 15 2 2 NUM brj-24033 68 16 Γ— Γ— NOUN brj-24033 68 17 2 2 NUM brj-24033 68 18 Γ— Γ— NOUN brj-24033 68 19 3 3 NUM brj-24033 68 20 cm cm NOUN brj-24033 68 21 . . PUNCT brj-24033 69 1 the the DET brj-24033 69 2 2 2 NUM brj-24033 69 3 Γ— Γ— NOUN brj-24033 69 4 2 2 NUM brj-24033 69 5 cm cm NOUN brj-24033 69 6 surface surface NOUN brj-24033 69 7 was be AUX brj-24033 69 8 the the DET brj-24033 69 9 cross cross NOUN brj-24033 69 10 section section NOUN brj-24033 69 11 , , PUNCT brj-24033 69 12 while while SCONJ brj-24033 69 13 the the DET brj-24033 69 14 2 2 NUM brj-24033 69 15 Γ— Γ— NOUN brj-24033 69 16 3 3 NUM brj-24033 69 17 cm cm NOUN brj-24033 69 18 surface surface NOUN brj-24033 69 19 was be AUX brj-24033 69 20 the the DET brj-24033 69 21 radial radial ADJ brj-24033 69 22 or or CCONJ brj-24033 69 23 tangential tangential ADJ brj-24033 69 24 section section NOUN brj-24033 69 25 . . PUNCT brj-24033 70 1 second second ADJ brj-24033 70 2 , , PUNCT brj-24033 70 3 2 2 NUM brj-24033 70 4 wood wood NOUN brj-24033 70 5 samples sample NOUN brj-24033 70 6 from from ADP brj-24033 70 7 every every DET brj-24033 70 8 wood wood NOUN brj-24033 70 9 block block NOUN brj-24033 70 10 were be AUX brj-24033 70 11 randomly randomly ADV brj-24033 70 12 selected select VERB brj-24033 70 13 so so ADV brj-24033 70 14 as as SCONJ brj-24033 70 15 to to PART brj-24033 70 16 obtain obtain VERB brj-24033 70 17 50 50 NUM brj-24033 70 18 wood wood NOUN brj-24033 70 19 samples sample NOUN brj-24033 70 20 with with ADP brj-24033 70 21 size size NOUN brj-24033 70 22 of of ADP brj-24033 70 23 2 2 NUM brj-24033 70 24 Γ— Γ— NOUN brj-24033 70 25 2 2 NUM brj-24033 70 26 Γ— Γ— NOUN brj-24033 70 27 3 3 NUM brj-24033 70 28 cm cm NOUN brj-24033 70 29 in in ADP brj-24033 70 30 total total NOUN brj-24033 70 31 for for ADP brj-24033 70 32 each each DET brj-24033 70 33 wood wood NOUN brj-24033 70 34 species specie NOUN brj-24033 70 35 . . PUNCT brj-24033 71 1 to to PART brj-24033 71 2 delete delete VERB brj-24033 71 3 the the DET brj-24033 71 4 uneven uneven ADJ brj-24033 71 5 burrs burrs NOUN brj-24033 71 6 from from ADP brj-24033 71 7 the the DET brj-24033 71 8 wood wood NOUN brj-24033 71 9 cutting cutting NOUN brj-24033 71 10 procedure procedure NOUN brj-24033 71 11 , , PUNCT brj-24033 71 12 sandpaper sandpaper NOUN brj-24033 71 13 of of ADP brj-24033 71 14 800 800 NUM brj-24033 71 15 to to ADP brj-24033 71 16 peer peer NOUN brj-24033 71 17 - - PUNCT brj-24033 71 18 reviewed review VERB brj-24033 71 19 article article NOUN brj-24033 71 20 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 71 21 zhang zhang PROPN brj-24033 71 22 & & CCONJ brj-24033 71 23 zhao zhao PROPN brj-24033 71 24 ( ( PUNCT brj-24033 71 25 2025 2025 NUM brj-24033 71 26 ) ) PUNCT brj-24033 71 27 . . PUNCT brj-24033 72 1 β€œ " PUNCT brj-24033 72 2 wood wood NOUN brj-24033 72 3 image image NOUN brj-24033 72 4 classification classification NOUN brj-24033 72 5 , , PUNCT brj-24033 72 6 ” " PUNCT brj-24033 72 7 bioresources bioresource NOUN brj-24033 72 8 20(1 20(1 NUM brj-24033 72 9 ) ) PUNCT brj-24033 72 10 , , PUNCT brj-24033 72 11 944 944 NUM brj-24033 72 12 - - SYM brj-24033 72 13 955 955 NUM brj-24033 72 14 . . PUNCT brj-24033 73 1 947 947 NUM brj-24033 73 2 1200 1200 NUM brj-24033 73 3 mesh mesh NOUN brj-24033 73 4 was be AUX brj-24033 73 5 used use VERB brj-24033 73 6 to to PART brj-24033 73 7 polish polish VERB brj-24033 73 8 the the DET brj-24033 73 9 cross cross NOUN brj-24033 73 10 sections section NOUN brj-24033 73 11 of of ADP brj-24033 73 12 wood wood NOUN brj-24033 73 13 samples sample NOUN brj-24033 73 14 . . PUNCT brj-24033 74 1 finally finally ADV brj-24033 74 2 , , PUNCT brj-24033 74 3 the the DET brj-24033 74 4 wood wood NOUN brj-24033 74 5 nir nir ADJ brj-24033 74 6 spectral spectral ADJ brj-24033 74 7 curves curve NOUN brj-24033 74 8 may may AUX brj-24033 74 9 be be AUX brj-24033 74 10 sensitive sensitive ADJ brj-24033 74 11 to to ADP brj-24033 74 12 some some DET brj-24033 74 13 external external ADJ brj-24033 74 14 environmental environmental ADJ brj-24033 74 15 factors factor NOUN brj-24033 74 16 such such ADJ brj-24033 74 17 as as ADP brj-24033 74 18 temperature temperature NOUN brj-24033 74 19 and and CCONJ brj-24033 74 20 humidity humidity NOUN brj-24033 74 21 so so SCONJ brj-24033 74 22 that that SCONJ brj-24033 74 23 the the DET brj-24033 74 24 spectral spectral ADJ brj-24033 74 25 acquisition acquisition NOUN brj-24033 74 26 was be AUX brj-24033 74 27 performed perform VERB brj-24033 74 28 in in ADP brj-24033 74 29 a a DET brj-24033 74 30 room room NOUN brj-24033 74 31 with with ADP brj-24033 74 32 temperature temperature NOUN brj-24033 74 33 at at ADP brj-24033 74 34 24 24 NUM brj-24033 74 35 Β° Β° NOUN brj-24033 74 36 c c NOUN brj-24033 74 37 and and CCONJ brj-24033 74 38 humidity humidity NOUN brj-24033 74 39 at at ADP brj-24033 74 40 35 35 NUM brj-24033 74 41 % % NOUN brj-24033 74 42 . . PUNCT brj-24033 75 1 it it PRON brj-24033 75 2 should should AUX brj-24033 75 3 be be AUX brj-24033 75 4 noted note VERB brj-24033 75 5 that that SCONJ brj-24033 75 6 the the DET brj-24033 75 7 physical physical ADJ brj-24033 75 8 property property NOUN brj-24033 75 9 of of ADP brj-24033 75 10 wood wood NOUN brj-24033 75 11 samples sample NOUN brj-24033 75 12 is be AUX brj-24033 75 13 influenced influence VERB brj-24033 75 14 by by ADP brj-24033 75 15 some some DET brj-24033 75 16 variables variable NOUN brj-24033 75 17 such such ADJ brj-24033 75 18 as as ADP brj-24033 75 19 the the DET brj-24033 75 20 age age NOUN brj-24033 75 21 of of ADP brj-24033 75 22 trees tree NOUN brj-24033 75 23 , , PUNCT brj-24033 75 24 geographic geographic ADJ brj-24033 75 25 origin origin NOUN brj-24033 75 26 , , PUNCT brj-24033 75 27 growth growth NOUN brj-24033 75 28 ring ring NOUN brj-24033 75 29 position position NOUN brj-24033 75 30 , , PUNCT brj-24033 75 31 and and CCONJ brj-24033 75 32 proportion proportion NOUN brj-24033 75 33 of of ADP brj-24033 75 34 latewood latewood NOUN brj-24033 75 35 versus versus ADP brj-24033 75 36 earlywood earlywood NOUN brj-24033 75 37 . . PUNCT brj-24033 76 1 these these DET brj-24033 76 2 variables variable NOUN brj-24033 76 3 are be AUX brj-24033 76 4 controlled control VERB brj-24033 76 5 effectively effectively ADV brj-24033 76 6 in in ADP brj-24033 76 7 wood wood NOUN brj-24033 76 8 spectral spectral ADJ brj-24033 76 9 acquisition acquisition NOUN brj-24033 76 10 so so SCONJ brj-24033 76 11 that that SCONJ brj-24033 76 12 the the DET brj-24033 76 13 within within ADP brj-24033 76 14 - - PUNCT brj-24033 76 15 class class NOUN brj-24033 76 16 difference difference NOUN brj-24033 76 17 of of ADP brj-24033 76 18 spectral spectral ADJ brj-24033 76 19 curves curve NOUN brj-24033 76 20 for for ADP brj-24033 76 21 each each DET brj-24033 76 22 wood wood NOUN brj-24033 76 23 species specie NOUN brj-24033 76 24 is be AUX brj-24033 76 25 adequately adequately ADV brj-24033 76 26 small small ADJ brj-24033 76 27 . . PUNCT brj-24033 77 1 this this DET brj-24033 77 2 control control NOUN brj-24033 77 3 is be AUX brj-24033 77 4 implemented implement VERB brj-24033 77 5 in in ADP brj-24033 77 6 practice practice NOUN brj-24033 77 7 by by ADP brj-24033 77 8 ensuring ensure VERB brj-24033 77 9 that that SCONJ brj-24033 77 10 π‘‘π‘Ÿπ‘Žπ‘π‘’(π‘Ίπ’˜ π‘‘π‘Ÿπ‘Žπ‘π‘’(π‘Ίπ’˜ ADV brj-24033 77 11 ) ) PUNCT brj-24033 77 12 is be AUX brj-24033 77 13 small small ADJ brj-24033 77 14 or or CCONJ brj-24033 77 15 less less ADJ brj-24033 77 16 than than ADP brj-24033 77 17 a a DET brj-24033 77 18 threshold threshold NOUN brj-24033 77 19 for for ADP brj-24033 77 20 every every DET brj-24033 77 21 species specie NOUN brj-24033 77 22 ( ( PUNCT brj-24033 77 23 i.e. i.e. X brj-24033 77 24 , , PUNCT brj-24033 77 25 𝑺𝑀 𝑺𝑀 PROPN brj-24033 77 26 denotes denote VERB brj-24033 77 27 the the DET brj-24033 77 28 within within ADP brj-24033 77 29 - - PUNCT brj-24033 77 30 class class NOUN brj-24033 77 31 scatter scatter NOUN brj-24033 77 32 matrix matrix NOUN brj-24033 77 33 ) ) PUNCT brj-24033 77 34 . . PUNCT brj-24033 78 1 fig fig NOUN brj-24033 78 2 . . PUNCT brj-24033 79 1 1 1 X brj-24033 79 2 . . X brj-24033 79 3 cross cross NOUN brj-24033 79 4 sections section NOUN brj-24033 79 5 of of ADP brj-24033 79 6 the the DET brj-24033 79 7 35 35 NUM brj-24033 79 8 wood wood NOUN brj-24033 79 9 species specie NOUN brj-24033 79 10 ( ( PUNCT brj-24033 79 11 the the DET brj-24033 79 12 serial serial ADJ brj-24033 79 13 number number NOUN brj-24033 79 14 in in ADP brj-24033 79 15 figure figure NOUN brj-24033 79 16 1 1 NUM brj-24033 79 17 is be AUX brj-24033 79 18 same same ADJ brj-24033 79 19 as as ADP brj-24033 79 20 that that PRON brj-24033 79 21 in in ADP brj-24033 79 22 table table NOUN brj-24033 79 23 1 1 NUM brj-24033 79 24 ) ) PUNCT brj-24033 79 25 . . PUNCT brj-24033 80 1 figure figure NOUN brj-24033 80 2 2 2 NUM brj-24033 80 3 shows show VERB brj-24033 80 4 the the DET brj-24033 80 5 experimental experimental ADJ brj-24033 80 6 spectral spectral ADJ brj-24033 80 7 collection collection NOUN brj-24033 80 8 setup setup NOUN brj-24033 80 9 . . PUNCT brj-24033 81 1 this this DET brj-24033 81 2 spectral spectral ADJ brj-24033 81 3 collection collection NOUN brj-24033 81 4 setup setup NOUN brj-24033 81 5 mainly mainly ADV brj-24033 81 6 consists consist VERB brj-24033 81 7 of of ADP brj-24033 81 8 a a DET brj-24033 81 9 computer computer NOUN brj-24033 81 10 , , PUNCT brj-24033 81 11 spectrometer spectrometer NOUN brj-24033 81 12 , , PUNCT brj-24033 81 13 optical optical ADJ brj-24033 81 14 fiber fiber NOUN brj-24033 81 15 , , PUNCT brj-24033 81 16 and and CCONJ brj-24033 81 17 radian radian ADJ brj-24033 81 18 ( ( PUNCT brj-24033 81 19 i.e. i.e. X brj-24033 81 20 , , PUNCT brj-24033 81 21 halogen halogen PROPN brj-24033 81 22 lamp lamp PROPN brj-24033 81 23 ) ) PUNCT brj-24033 81 24 . . PUNCT brj-24033 82 1 the the DET brj-24033 82 2 spectral spectral ADJ brj-24033 82 3 acquisition acquisition NOUN brj-24033 82 4 is be AUX brj-24033 82 5 performed perform VERB brj-24033 82 6 as as ADP brj-24033 82 7 following follow VERB brj-24033 82 8 steps step NOUN brj-24033 82 9 . . PUNCT brj-24033 83 1 a a DET brj-24033 83 2 spectral spectral ADJ brj-24033 83 3 calibration calibration NOUN brj-24033 83 4 is be AUX brj-24033 83 5 performed perform VERB brj-24033 83 6 by by ADP brj-24033 83 7 using use VERB brj-24033 83 8 a a DET brj-24033 83 9 standard standard ADJ brj-24033 83 10 whiteboard whiteboard NOUN brj-24033 83 11 . . PUNCT brj-24033 84 1 then then ADV brj-24033 84 2 one one NUM brj-24033 84 3 wood wood NOUN brj-24033 84 4 sample sample NOUN brj-24033 84 5 is be AUX brj-24033 84 6 placed place VERB brj-24033 84 7 on on ADP brj-24033 84 8 the the DET brj-24033 84 9 holder holder NOUN brj-24033 84 10 , , PUNCT brj-24033 84 11 and and CCONJ brj-24033 84 12 the the DET brj-24033 84 13 distance distance NOUN brj-24033 84 14 between between ADP brj-24033 84 15 this this DET brj-24033 84 16 wood wood NOUN brj-24033 84 17 sample sample NOUN brj-24033 84 18 and and CCONJ brj-24033 84 19 the the DET brj-24033 84 20 fiber fiber NOUN brj-24033 84 21 probe probe NOUN brj-24033 84 22 is be AUX brj-24033 84 23 adjusted adjust VERB brj-24033 84 24 . . PUNCT brj-24033 85 1 finally finally ADV brj-24033 85 2 , , PUNCT brj-24033 85 3 the the DET brj-24033 85 4 spectral spectral ADJ brj-24033 85 5 reflectance reflectance NOUN brj-24033 85 6 curves curve NOUN brj-24033 85 7 are be AUX brj-24033 85 8 picked pick VERB brj-24033 85 9 up up ADP brj-24033 85 10 and and CCONJ brj-24033 85 11 are be AUX brj-24033 85 12 saved save VERB brj-24033 85 13 in in ADP brj-24033 85 14 the the DET brj-24033 85 15 computer computer NOUN brj-24033 85 16 . . PUNCT brj-24033 86 1 fig fig NOUN brj-24033 86 2 . . PUNCT brj-24033 87 1 2 2 X brj-24033 87 2 . . X brj-24033 87 3 experimental experimental ADJ brj-24033 87 4 spectral spectral ADJ brj-24033 87 5 acquisition acquisition NOUN brj-24033 87 6 setup setup NOUN brj-24033 87 7 peer peer NOUN brj-24033 87 8 - - PUNCT brj-24033 87 9 reviewed review VERB brj-24033 87 10 article article NOUN brj-24033 87 11 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 87 12 zhang zhang PROPN brj-24033 87 13 & & CCONJ brj-24033 87 14 zhao zhao PROPN brj-24033 87 15 ( ( PUNCT brj-24033 87 16 2025 2025 NUM brj-24033 87 17 ) ) PUNCT brj-24033 87 18 . . PUNCT brj-24033 88 1 β€œ " PUNCT brj-24033 88 2 wood wood NOUN brj-24033 88 3 image image NOUN brj-24033 88 4 classification classification NOUN brj-24033 88 5 , , PUNCT brj-24033 88 6 ” " PUNCT brj-24033 88 7 bioresources bioresource NOUN brj-24033 88 8 20(1 20(1 NUM brj-24033 88 9 ) ) PUNCT brj-24033 88 10 , , PUNCT brj-24033 88 11 944 944 NUM brj-24033 88 12 - - SYM brj-24033 88 13 955 955 NUM brj-24033 88 14 . . PUNCT brj-24033 89 1 948 948 NUM brj-24033 89 2 spectral spectral ADJ brj-24033 89 3 dimension dimension NOUN brj-24033 89 4 reduction reduction NOUN brj-24033 89 5 before before ADP brj-24033 89 6 the the DET brj-24033 89 7 wood wood NOUN brj-24033 89 8 spectral spectral ADJ brj-24033 89 9 dimension dimension NOUN brj-24033 89 10 reduction reduction NOUN brj-24033 89 11 , , PUNCT brj-24033 89 12 a a DET brj-24033 89 13 wood wood NOUN brj-24033 89 14 spectral spectral ADJ brj-24033 89 15 pre pre ADJ brj-24033 89 16 - - ADJ brj-24033 89 17 processing processing ADJ brj-24033 89 18 procedure procedure NOUN brj-24033 89 19 is be AUX brj-24033 89 20 required require VERB brj-24033 89 21 . . PUNCT brj-24033 90 1 figure figure NOUN brj-24033 90 2 3 3 NUM brj-24033 90 3 illustrates illustrate VERB brj-24033 90 4 the the DET brj-24033 90 5 spectral spectral ADJ brj-24033 90 6 reflectance reflectance NOUN brj-24033 90 7 curves curve NOUN brj-24033 90 8 of of ADP brj-24033 90 9 the the DET brj-24033 90 10 35 35 NUM brj-24033 90 11 wood wood NOUN brj-24033 90 12 species specie NOUN brj-24033 90 13 . . PUNCT brj-24033 91 1 a a DET brj-24033 91 2 standard standard ADJ brj-24033 91 3 normal normal ADJ brj-24033 91 4 variation variation NOUN brj-24033 91 5 ( ( PUNCT brj-24033 91 6 snv snv PROPN brj-24033 91 7 ) ) PUNCT brj-24033 91 8 correction correction NOUN brj-24033 91 9 and and CCONJ brj-24033 91 10 a a DET brj-24033 91 11 smoothing smooth VERB brj-24033 91 12 correction correction NOUN brj-24033 91 13 with with ADP brj-24033 91 14 a a DET brj-24033 91 15 moving move VERB brj-24033 91 16 window window NOUN brj-24033 91 17 of of ADP brj-24033 91 18 5 5 NUM brj-24033 91 19 Γ— Γ— NOUN brj-24033 91 20 5 5 NUM brj-24033 91 21 size size NOUN brj-24033 91 22 are be AUX brj-24033 91 23 often often ADV brj-24033 91 24 used use VERB brj-24033 91 25 for for ADP brj-24033 91 26 the the DET brj-24033 91 27 nir nir ADJ brj-24033 91 28 spectral spectral ADJ brj-24033 91 29 curves curve NOUN brj-24033 91 30 to to PART brj-24033 91 31 ensure ensure VERB brj-24033 91 32 a a DET brj-24033 91 33 good good ADJ brj-24033 91 34 classification classification NOUN brj-24033 91 35 accuracy accuracy NOUN brj-24033 91 36 . . PUNCT brj-24033 92 1 fig fig NOUN brj-24033 92 2 . . PUNCT brj-24033 93 1 3 3 X brj-24033 93 2 . . X brj-24033 93 3 spectral spectral ADJ brj-24033 93 4 reflectance reflectance NOUN brj-24033 93 5 curves curve NOUN brj-24033 93 6 of of ADP brj-24033 93 7 cross cross NOUN brj-24033 93 8 sections section NOUN brj-24033 93 9 of of ADP brj-24033 93 10 35 35 NUM brj-24033 93 11 wood wood NOUN brj-24033 93 12 species specie NOUN brj-24033 93 13 the the DET brj-24033 93 14 nir nir ADJ brj-24033 93 15 spectral spectral ADJ brj-24033 93 16 curve curve NOUN brj-24033 93 17 may may AUX brj-24033 93 18 be be AUX brj-24033 93 19 a a DET brj-24033 93 20 128d 128d NUM brj-24033 93 21 vector vector NOUN brj-24033 93 22 , , PUNCT brj-24033 93 23 so so SCONJ brj-24033 93 24 that that SCONJ brj-24033 93 25 a a DET brj-24033 93 26 spectral spectral ADJ brj-24033 93 27 dimension dimension NOUN brj-24033 93 28 reduction reduction NOUN brj-24033 93 29 is be AUX brj-24033 93 30 usually usually ADV brj-24033 93 31 performed perform VERB brj-24033 93 32 to to PART brj-24033 93 33 decrease decrease VERB brj-24033 93 34 the the DET brj-24033 93 35 redundant redundant ADJ brj-24033 93 36 information information NOUN brj-24033 93 37 and and CCONJ brj-24033 93 38 increase increase VERB brj-24033 93 39 the the DET brj-24033 93 40 computational computational ADJ brj-24033 93 41 efficiency efficiency NOUN brj-24033 93 42 . . PUNCT brj-24033 94 1 some some DET brj-24033 94 2 feature feature NOUN brj-24033 94 3 dimension dimension NOUN brj-24033 94 4 reduction reduction NOUN brj-24033 94 5 algorithms algorithm NOUN brj-24033 94 6 can can AUX brj-24033 94 7 be be AUX brj-24033 94 8 used use VERB brj-24033 94 9 such such ADJ brj-24033 94 10 as as ADP brj-24033 94 11 principal principal ADJ brj-24033 94 12 component component NOUN brj-24033 94 13 analysis analysis NOUN brj-24033 94 14 ( ( PUNCT brj-24033 94 15 pca pca NOUN brj-24033 94 16 ) ) PUNCT brj-24033 94 17 ( ( PUNCT brj-24033 94 18 reddy reddy PROPN brj-24033 94 19 et et PROPN brj-24033 94 20 al al PROPN brj-24033 94 21 . . PROPN brj-24033 94 22 2020 2020 NUM brj-24033 94 23 ) ) PUNCT brj-24033 94 24 , , PUNCT brj-24033 94 25 multidimensional multidimensional ADJ brj-24033 94 26 scaling scaling NOUN brj-24033 94 27 ( ( PUNCT brj-24033 94 28 mds mds PROPN brj-24033 94 29 ) ) PUNCT brj-24033 94 30 ( ( PUNCT brj-24033 94 31 mignotte mignotte NOUN brj-24033 94 32 2011 2011 NUM brj-24033 94 33 ) ) PUNCT brj-24033 94 34 , , PUNCT brj-24033 94 35 locally locally ADV brj-24033 94 36 linear linear VERB brj-24033 94 37 embedding embed VERB brj-24033 94 38 ( ( PUNCT brj-24033 94 39 lle lle PROPN brj-24033 94 40 ) ) PUNCT brj-24033 94 41 ( ( PUNCT brj-24033 94 42 yu yu PROPN brj-24033 94 43 et et PROPN brj-24033 94 44 al al PROPN brj-24033 94 45 . . PROPN brj-24033 94 46 2020 2020 NUM brj-24033 94 47 ) ) PUNCT brj-24033 94 48 , , PUNCT brj-24033 94 49 laplacian laplacian X brj-24033 94 50 ( ( PUNCT brj-24033 94 51 belkin belkin NOUN brj-24033 94 52 and and CCONJ brj-24033 94 53 niyogi niyogi ADJ brj-24033 94 54 2003 2003 NUM brj-24033 94 55 ) ) PUNCT brj-24033 94 56 , , PUNCT brj-24033 94 57 and and CCONJ brj-24033 94 58 kernel kernel PROPN brj-24033 94 59 pca pca PROPN brj-24033 94 60 ( ( PUNCT brj-24033 94 61 alhayani alhayani NOUN brj-24033 94 62 and and CCONJ brj-24033 94 63 ilhan ilhan PROPN brj-24033 94 64 2017 2017 NUM brj-24033 94 65 ) ) PUNCT brj-24033 94 66 . . PUNCT brj-24033 95 1 in in ADP brj-24033 95 2 this this DET brj-24033 95 3 work work NOUN brj-24033 95 4 , , PUNCT brj-24033 95 5 a a DET brj-24033 95 6 metric metric ADJ brj-24033 95 7 learning learning NOUN brj-24033 95 8 ( ( PUNCT brj-24033 95 9 ml ml NOUN brj-24033 95 10 ) ) PUNCT brj-24033 95 11 algorithm algorithm NOUN brj-24033 95 12 proposed propose VERB brj-24033 95 13 by by ADP brj-24033 95 14 mensink mensink PROPN brj-24033 95 15 et et PROPN brj-24033 95 16 al al PROPN brj-24033 95 17 . . PROPN brj-24033 95 18 ( ( PUNCT brj-24033 95 19 2013 2013 NUM brj-24033 95 20 ) ) PUNCT brj-24033 95 21 was be AUX brj-24033 95 22 used use VERB brj-24033 95 23 for for ADP brj-24033 95 24 spectral spectral ADJ brj-24033 95 25 dimension dimension NOUN brj-24033 95 26 reduction reduction NOUN brj-24033 95 27 . . PUNCT brj-24033 96 1 when when SCONJ brj-24033 96 2 a a DET brj-24033 96 3 new new ADJ brj-24033 96 4 class class NOUN brj-24033 96 5 is be AUX brj-24033 96 6 added add VERB brj-24033 96 7 into into ADP brj-24033 96 8 the the DET brj-24033 96 9 training training NOUN brj-24033 96 10 set set NOUN brj-24033 96 11 of of ADP brj-24033 96 12 the the DET brj-24033 96 13 open open ADJ brj-24033 96 14 set set NOUN brj-24033 96 15 classifier classifier NOUN brj-24033 96 16 , , PUNCT brj-24033 96 17 the the DET brj-24033 96 18 previous previous ADJ brj-24033 96 19 projection projection NOUN brj-24033 96 20 matrix matrix NOUN brj-24033 96 21 w w NOUN brj-24033 96 22 can can AUX brj-24033 96 23 still still ADV brj-24033 96 24 be be AUX brj-24033 96 25 used use VERB brj-24033 96 26 with with ADP brj-24033 96 27 near near ADV brj-24033 96 28 - - PUNCT brj-24033 96 29 zero zero NUM brj-24033 96 30 errors error NOUN brj-24033 96 31 , , PUNCT brj-24033 96 32 as as SCONJ brj-24033 96 33 pointed point VERB brj-24033 96 34 out out ADP brj-24033 96 35 by by ADP brj-24033 96 36 mensink mensink PROPN brj-24033 96 37 et et PROPN brj-24033 96 38 al al PROPN brj-24033 96 39 . . PROPN brj-24033 97 1 ( ( PUNCT brj-24033 97 2 2013 2013 NUM brj-24033 97 3 ) ) PUNCT brj-24033 97 4 . . PUNCT brj-24033 98 1 therefore therefore ADV brj-24033 98 2 , , PUNCT brj-24033 98 3 this this DET brj-24033 98 4 projection projection NOUN brj-24033 98 5 matrix matrix NOUN brj-24033 98 6 w w NOUN brj-24033 98 7 can can AUX brj-24033 98 8 be be AUX brj-24033 98 9 used use VERB brj-24033 98 10 in in ADP brj-24033 98 11 an an DET brj-24033 98 12 incremental incremental ADJ brj-24033 98 13 learning learning NOUN brj-24033 98 14 classifier classifier NOUN brj-24033 98 15 , , PUNCT brj-24033 98 16 in in ADP brj-24033 98 17 which which PRON brj-24033 98 18 the the DET brj-24033 98 19 number number NOUN brj-24033 98 20 of of ADP brj-24033 98 21 known know VERB brj-24033 98 22 classes class NOUN brj-24033 98 23 is be AUX brj-24033 98 24 increased increase VERB brj-24033 98 25 gradually gradually ADV brj-24033 98 26 . . PUNCT brj-24033 99 1 this this DET brj-24033 99 2 incremental incremental ADJ brj-24033 99 3 learning learning NOUN brj-24033 99 4 classifier classifier NOUN brj-24033 99 5 is be AUX brj-24033 99 6 usually usually ADV brj-24033 99 7 used use VERB brj-24033 99 8 in in ADP brj-24033 99 9 an an DET brj-24033 99 10 open open ADJ brj-24033 99 11 set set NOUN brj-24033 99 12 scenario scenario NOUN brj-24033 99 13 , , PUNCT brj-24033 99 14 since since SCONJ brj-24033 99 15 one one PRON brj-24033 99 16 often often ADV brj-24033 99 17 wishes wish VERB brj-24033 99 18 to to PART brj-24033 99 19 increase increase VERB brj-24033 99 20 the the DET brj-24033 99 21 number number NOUN brj-24033 99 22 of of ADP brj-24033 99 23 known know VERB brj-24033 99 24 classes class NOUN brj-24033 99 25 in in ADP brj-24033 99 26 a a DET brj-24033 99 27 training training NOUN brj-24033 99 28 set set VERB brj-24033 99 29 so so SCONJ brj-24033 99 30 as as SCONJ brj-24033 99 31 to to PART brj-24033 99 32 classify classify VERB brj-24033 99 33 more more ADV brj-24033 99 34 known know VERB brj-24033 99 35 classes class NOUN brj-24033 99 36 and and CCONJ brj-24033 99 37 reject reject VERB brj-24033 99 38 fewer few ADJ brj-24033 99 39 unknown unknown ADJ brj-24033 99 40 classes class NOUN brj-24033 99 41 correctly correctly ADV brj-24033 99 42 . . PUNCT brj-24033 100 1 due due ADP brj-24033 100 2 to to ADP brj-24033 100 3 the the DET brj-24033 100 4 above above ADV brj-24033 100 5 - - PUNCT brj-24033 100 6 mentioned mention VERB brj-24033 100 7 advantages advantage NOUN brj-24033 100 8 in in ADP brj-24033 100 9 the the DET brj-24033 100 10 incremental incremental ADJ brj-24033 100 11 learning learning NOUN brj-24033 100 12 classifier classifier NOUN brj-24033 100 13 , , PUNCT brj-24033 100 14 this this DET brj-24033 100 15 ml ml NOUN brj-24033 100 16 algorithm algorithm NOUN brj-24033 100 17 is be AUX brj-24033 100 18 also also ADV brj-24033 100 19 used use VERB brj-24033 100 20 in in ADP brj-24033 100 21 the the DET brj-24033 100 22 nno nno NOUN brj-24033 100 23 classifier classifier NOUN brj-24033 100 24 ( ( PUNCT brj-24033 100 25 bendale bendale NOUN brj-24033 100 26 and and CCONJ brj-24033 100 27 boult boult NOUN brj-24033 100 28 2015 2015 NUM brj-24033 100 29 ) ) PUNCT brj-24033 100 30 . . PUNCT brj-24033 101 1 if if SCONJ brj-24033 101 2 one one NUM brj-24033 101 3 defines define VERB brj-24033 101 4 an an DET brj-24033 101 5 original original ADJ brj-24033 101 6 128d 128d NOUN brj-24033 101 7 spectral spectral ADJ brj-24033 101 8 vector vector NOUN brj-24033 101 9 as as ADP brj-24033 101 10 𝒗 𝒗 PROPN brj-24033 101 11 , , PUNCT brj-24033 101 12 then then ADV brj-24033 101 13 a a DET brj-24033 101 14 new new ADJ brj-24033 101 15 spectral spectral ADJ brj-24033 101 16 vector vector NOUN brj-24033 101 17 after after SCONJ brj-24033 101 18 spectral spectral ADJ brj-24033 101 19 dimension dimension NOUN brj-24033 101 20 reduction reduction NOUN brj-24033 101 21 is be AUX brj-24033 101 22 denoted denote VERB brj-24033 101 23 as as SCONJ brj-24033 101 24 𝑾𝒗. 𝑾𝒗. PROPN brj-24033 101 25 the the DET brj-24033 101 26 detailed detailed ADJ brj-24033 101 27 computation computation NOUN brj-24033 101 28 procedure procedure NOUN brj-24033 101 29 is be AUX brj-24033 101 30 illustrated illustrate VERB brj-24033 101 31 by by ADP brj-24033 101 32 mensink mensink PROPN brj-24033 101 33 et et PROPN brj-24033 101 34 al al PROPN brj-24033 101 35 . . PROPN brj-24033 102 1 ( ( PUNCT brj-24033 102 2 2013 2013 NUM brj-24033 102 3 ) ) PUNCT brj-24033 102 4 , , PUNCT brj-24033 102 5 which which PRON brj-24033 102 6 is be AUX brj-24033 102 7 omitted omit VERB brj-24033 102 8 here here ADV brj-24033 102 9 . . PUNCT brj-24033 103 1 original original ADJ brj-24033 103 2 nno nno PROPN brj-24033 103 3 classifier classifier PROPN brj-24033 103 4 the the DET brj-24033 103 5 original original ADJ brj-24033 103 6 nno nno PROPN brj-24033 103 7 classifier classifier NOUN brj-24033 103 8 was be AUX brj-24033 103 9 proposed propose VERB brj-24033 103 10 by by ADP brj-24033 103 11 bendale bendale NOUN brj-24033 103 12 and and CCONJ brj-24033 103 13 boult boult NOUN brj-24033 103 14 ( ( PUNCT brj-24033 103 15 2015 2015 NUM brj-24033 103 16 ) ) PUNCT brj-24033 103 17 . . PUNCT brj-24033 104 1 in in ADP brj-24033 104 2 summary summary NOUN brj-24033 104 3 , , PUNCT brj-24033 104 4 an an DET brj-24033 104 5 original original ADJ brj-24033 104 6 nno nno NOUN brj-24033 104 7 classifier classifier NOUN brj-24033 104 8 is be AUX brj-24033 104 9 an an DET brj-24033 104 10 extension extension NOUN brj-24033 104 11 version version NOUN brj-24033 104 12 of of ADP brj-24033 104 13 ncm ncm PROPN brj-24033 104 14 classifier classifier PROPN brj-24033 104 15 for for ADP brj-24033 104 16 osr osr PROPN brj-24033 104 17 wavelength wavelength NOUN brj-24033 104 18 ( ( PUNCT brj-24033 104 19 nm nm NOUN brj-24033 104 20 ) ) PUNCT brj-24033 105 1 r r NOUN brj-24033 105 2 e e NOUN brj-24033 105 3 fl fl NOUN brj-24033 105 4 e e NOUN brj-24033 105 5 c c NOUN brj-24033 105 6 ti ti X brj-24033 105 7 v v ADP brj-24033 105 8 it it PRON brj-24033 105 9 y y PROPN brj-24033 105 10 peer peer NOUN brj-24033 105 11 - - PUNCT brj-24033 105 12 reviewed review VERB brj-24033 105 13 article article NOUN brj-24033 105 14 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 105 15 zhang zhang PROPN brj-24033 105 16 & & CCONJ brj-24033 105 17 zhao zhao PROPN brj-24033 105 18 ( ( PUNCT brj-24033 105 19 2025 2025 NUM brj-24033 105 20 ) ) PUNCT brj-24033 105 21 . . PUNCT brj-24033 106 1 β€œ " PUNCT brj-24033 106 2 wood wood NOUN brj-24033 106 3 image image NOUN brj-24033 106 4 classification classification NOUN brj-24033 106 5 , , PUNCT brj-24033 106 6 ” " PUNCT brj-24033 106 7 bioresources bioresource NOUN brj-24033 106 8 20(1 20(1 NUM brj-24033 106 9 ) ) PUNCT brj-24033 106 10 , , PUNCT brj-24033 106 11 944 944 NUM brj-24033 106 12 - - SYM brj-24033 106 13 955 955 NUM brj-24033 106 14 . . PUNCT brj-24033 107 1 949 949 NUM brj-24033 107 2 use use NOUN brj-24033 107 3 . . PUNCT brj-24033 108 1 in in ADP brj-24033 108 2 an an DET brj-24033 108 3 nno nno NOUN brj-24033 108 4 classifier classifier NOUN brj-24033 108 5 , , PUNCT brj-24033 108 6 a a DET brj-24033 108 7 confidence confidence NOUN brj-24033 108 8 score score NOUN brj-24033 108 9 for for ADP brj-24033 108 10 one one NUM brj-24033 108 11 known know VERB brj-24033 108 12 class class NOUN brj-24033 108 13 y y PROPN brj-24033 108 14 is be AUX brj-24033 108 15 defined define VERB brj-24033 108 16 as as ADP brj-24033 108 17 follows follow VERB brj-24033 108 18 . . PUNCT brj-24033 109 1 𝑠𝑦(𝒙 𝑠𝑦(𝒙 ADJ brj-24033 109 2 , , PUNCT brj-24033 109 3 𝜏 𝜏 NOUN brj-24033 109 4 ) ) PUNCT brj-24033 109 5 = = SYM brj-24033 110 1 π‘πœ π‘πœ NOUN brj-24033 110 2 ( ( PUNCT brj-24033 110 3 1 1 NUM brj-24033 110 4 βˆ’ βˆ’ PROPN brj-24033 110 5 1 1 NUM brj-24033 110 6 𝜏 𝜏 PRON brj-24033 110 7 𝑑𝑾(𝒙 𝑑𝑾(𝒙 NUM brj-24033 110 8 , , PUNCT brj-24033 110 9 𝝁𝑦 𝝁𝑦 NOUN brj-24033 110 10 ) ) PUNCT brj-24033 110 11 ) ) PUNCT brj-24033 110 12 ( ( PUNCT brj-24033 110 13 1 1 X brj-24033 110 14 ) ) PUNCT brj-24033 110 15 where where SCONJ brj-24033 110 16 𝒙 𝒙 PROPN brj-24033 110 17 is be AUX brj-24033 110 18 a a DET brj-24033 110 19 detected detect VERB brj-24033 110 20 sample sample NOUN brj-24033 110 21 vector vector NOUN brj-24033 110 22 , , PUNCT brj-24033 110 23 and and CCONJ brj-24033 110 24 the the DET brj-24033 110 25 parameter parameter NOUN brj-24033 110 26 𝜏 𝜏 NOUN brj-24033 110 27 is be AUX brj-24033 110 28 a a DET brj-24033 110 29 fixed fix VERB brj-24033 110 30 threshold threshold NOUN brj-24033 110 31 value value NOUN brj-24033 110 32 to to PART brj-24033 110 33 define define VERB brj-24033 110 34 a a DET brj-24033 110 35 sphere sphere NOUN brj-24033 110 36 radius radius NOUN brj-24033 110 37 around around ADP brj-24033 110 38 each each DET brj-24033 110 39 known know VERB brj-24033 110 40 class class NOUN brj-24033 110 41 mean mean NOUN brj-24033 110 42 𝝁𝑦. 𝝁𝑦. NUM brj-24033 111 1 this this DET brj-24033 111 2 𝜏 𝜏 NOUN brj-24033 111 3 is be AUX brj-24033 111 4 determined determine VERB brj-24033 111 5 in in ADP brj-24033 111 6 advance advance NOUN brj-24033 111 7 by by ADP brj-24033 111 8 an an DET brj-24033 111 9 experienced experienced ADJ brj-24033 111 10 expert expert NOUN brj-24033 111 11 and and CCONJ brj-24033 111 12 it it PRON brj-24033 111 13 is be AUX brj-24033 111 14 a a DET brj-24033 111 15 constant constant ADJ brj-24033 111 16 value value NOUN brj-24033 111 17 for for ADP brj-24033 111 18 all all DET brj-24033 111 19 known know VERB brj-24033 111 20 classes class NOUN brj-24033 111 21 in in ADP brj-24033 111 22 the the DET brj-24033 111 23 original original ADJ brj-24033 111 24 nno nno NOUN brj-24033 111 25 classifier classifier NOUN brj-24033 111 26 . . PUNCT brj-24033 112 1 the the DET brj-24033 112 2 distance distance NOUN brj-24033 112 3 𝑑𝑾(𝒙 𝑑𝑾(𝒙 NUM brj-24033 112 4 , , PUNCT brj-24033 112 5 𝝁𝑦 𝝁𝑦 NOUN brj-24033 112 6 ) ) PUNCT brj-24033 112 7 = = PUNCT brj-24033 113 1 ‖𝑾𝒙 ‖𝑾𝒙 ADJ brj-24033 113 2 βˆ’ βˆ’ NUM brj-24033 114 1 𝑾𝝁𝑦‖ 𝑾𝝁𝑦‖ PROPN brj-24033 114 2 2 2 NUM brj-24033 114 3 . . PUNCT brj-24033 115 1 the the DET brj-24033 115 2 normalization normalization NOUN brj-24033 115 3 factor factor NOUN brj-24033 115 4 π‘πœ π‘πœ PROPN brj-24033 115 5 is be AUX brj-24033 115 6 defined define VERB brj-24033 115 7 as as ADP brj-24033 115 8 π‘πœ π‘πœ PROPN brj-24033 115 9 = = PUNCT brj-24033 115 10 ( ( PUNCT brj-24033 115 11 Π³ Π³ PROPN brj-24033 115 12 ( ( PUNCT brj-24033 115 13 π‘š π‘š PROPN brj-24033 115 14 2 2 NUM brj-24033 115 15 + + NUM brj-24033 115 16 1 1 NUM brj-24033 115 17 ) ) PUNCT brj-24033 115 18 ) ) PUNCT brj-24033 115 19 /(πœ‹π‘š/2πœπ‘š /(πœ‹π‘š/2πœπ‘š PUNCT brj-24033 115 20 ) ) PUNCT brj-24033 116 1 so so SCONJ brj-24033 116 2 that that SCONJ brj-24033 116 3 𝑠𝑦 𝑠𝑦 ADP brj-24033 116 4 integrates integrate NOUN brj-24033 116 5 to to ADP brj-24033 116 6 1 1 NUM brj-24033 116 7 in in ADP brj-24033 116 8 the the DET brj-24033 116 9 domain domain NOUN brj-24033 116 10 𝑠𝑦(βˆ™ 𝑠𝑦(βˆ™ PROPN brj-24033 116 11 ) ) PUNCT brj-24033 116 12 > > X brj-24033 116 13 0 0 PUNCT brj-24033 117 1 ( ( PUNCT brj-24033 117 2 i.e. i.e. X brj-24033 117 3 , , PUNCT brj-24033 117 4 Π³ Π³ PROPN brj-24033 117 5 represents represent VERB brj-24033 117 6 a a DET brj-24033 117 7 standard standard ADJ brj-24033 117 8 gamma gamma NOUN brj-24033 117 9 function function NOUN brj-24033 117 10 ) ) PUNCT brj-24033 117 11 . . PUNCT brj-24033 118 1 in in ADP brj-24033 118 2 fact fact NOUN brj-24033 118 3 , , PUNCT brj-24033 118 4 π‘πœ π‘πœ PROPN brj-24033 118 5 is be AUX brj-24033 118 6 the the DET brj-24033 118 7 reverse reverse NOUN brj-24033 118 8 of of ADP brj-24033 118 9 sphere sphere NOUN brj-24033 118 10 volume volume NOUN brj-24033 118 11 with with ADP brj-24033 118 12 radius radius NOUN brj-24033 118 13 𝜏 𝜏 NOUN brj-24033 118 14 and and CCONJ brj-24033 118 15 dimension dimension NOUN brj-24033 118 16 m. m. NOUN brj-24033 118 17 as as ADP brj-24033 118 18 for for ADP brj-24033 118 19 open open ADJ brj-24033 118 20 set set ADJ brj-24033 118 21 classification classification NOUN brj-24033 118 22 , , PUNCT brj-24033 118 23 a a DET brj-24033 118 24 detected detect VERB brj-24033 118 25 sample sample NOUN brj-24033 118 26 𝒙 𝒙 PRON brj-24033 118 27 is be AUX brj-24033 118 28 rejected reject VERB brj-24033 118 29 by by ADP brj-24033 118 30 one one NUM brj-24033 118 31 known know VERB brj-24033 118 32 class class NOUN brj-24033 118 33 y y PROPN brj-24033 118 34 when when SCONJ brj-24033 118 35 𝑠𝑦(𝒙 𝑠𝑦(𝒙 PROPN brj-24033 118 36 , , PUNCT brj-24033 118 37 𝜏 𝜏 NOUN brj-24033 118 38 ) ) PUNCT brj-24033 118 39 ≀ ≀ NOUN brj-24033 118 40 0 0 NUM brj-24033 118 41 , , PUNCT brj-24033 118 42 and and CCONJ brj-24033 118 43 this this DET brj-24033 118 44 𝒙 𝒙 NOUN brj-24033 118 45 is be AUX brj-24033 118 46 rejected reject VERB brj-24033 118 47 as as ADP brj-24033 118 48 an an DET brj-24033 118 49 unknown unknown ADJ brj-24033 118 50 class class NOUN brj-24033 118 51 only only ADV brj-24033 118 52 when when SCONJ brj-24033 118 53 it it PRON brj-24033 118 54 is be AUX brj-24033 118 55 rejected reject VERB brj-24033 118 56 by by ADP brj-24033 118 57 all all DET brj-24033 118 58 known know VERB brj-24033 118 59 classes class NOUN brj-24033 118 60 . . PUNCT brj-24033 119 1 otherwise otherwise ADV brj-24033 119 2 , , PUNCT brj-24033 119 3 this this DET brj-24033 119 4 𝒙 𝒙 NOUN brj-24033 119 5 is be AUX brj-24033 119 6 classified classify VERB brj-24033 119 7 as as ADP brj-24033 119 8 a a DET brj-24033 119 9 known know VERB brj-24033 119 10 class class NOUN brj-24033 119 11 with with ADP brj-24033 119 12 the the DET brj-24033 119 13 largest large ADJ brj-24033 119 14 positive positive ADJ brj-24033 119 15 𝑠𝑦(𝒙 𝑠𝑦(𝒙 NOUN brj-24033 119 16 , , PUNCT brj-24033 119 17 𝜏 𝜏 NOUN brj-24033 119 18 ) ) PUNCT brj-24033 119 19 . . PUNCT brj-24033 120 1 the the DET brj-24033 120 2 projection projection NOUN brj-24033 120 3 matrix matrix NOUN brj-24033 120 4 w w NOUN brj-24033 120 5 is be AUX brj-24033 120 6 learned learn VERB brj-24033 120 7 offline offline ADJ brj-24033 120 8 in in ADP brj-24033 120 9 an an DET brj-24033 120 10 initial initial ADJ brj-24033 120 11 training training NOUN brj-24033 120 12 set set NOUN brj-24033 120 13 of of ADP brj-24033 120 14 known know VERB brj-24033 120 15 classes class NOUN brj-24033 120 16 . . PUNCT brj-24033 121 1 this this DET brj-24033 121 2 matrix matrix NOUN brj-24033 121 3 can can AUX brj-24033 121 4 still still ADV brj-24033 121 5 be be AUX brj-24033 121 6 used use VERB brj-24033 121 7 with with ADP brj-24033 121 8 near near ADV brj-24033 121 9 - - PUNCT brj-24033 121 10 zero zero NUM brj-24033 121 11 errors error NOUN brj-24033 121 12 in in ADP brj-24033 121 13 an an DET brj-24033 121 14 incremental incremental ADJ brj-24033 121 15 learning learning NOUN brj-24033 121 16 classifier classifier NOUN brj-24033 121 17 where where SCONJ brj-24033 121 18 some some DET brj-24033 121 19 new new ADJ brj-24033 121 20 classes class NOUN brj-24033 121 21 are be AUX brj-24033 121 22 added add VERB brj-24033 121 23 into into ADP brj-24033 121 24 the the DET brj-24033 121 25 training training NOUN brj-24033 121 26 set set VERB brj-24033 121 27 gradually gradually ADV brj-24033 121 28 , , PUNCT brj-24033 121 29 as as SCONJ brj-24033 121 30 pointed point VERB brj-24033 121 31 out out ADP brj-24033 121 32 by by ADP brj-24033 121 33 mensink mensink PROPN brj-24033 121 34 et et PROPN brj-24033 121 35 al al PROPN brj-24033 121 36 . . PROPN brj-24033 122 1 ( ( PUNCT brj-24033 122 2 2013 2013 NUM brj-24033 122 3 ) ) PUNCT brj-24033 122 4 . . PUNCT brj-24033 123 1 proposed propose VERB brj-24033 123 2 improved improve VERB brj-24033 123 3 nno nno PROPN brj-24033 123 4 classifier classifier NOUN brj-24033 123 5 the the DET brj-24033 123 6 original original ADJ brj-24033 123 7 nno nno NOUN brj-24033 123 8 classifier classifier NOUN brj-24033 123 9 ( ( PUNCT brj-24033 123 10 bendale bendale NOUN brj-24033 123 11 and and CCONJ brj-24033 123 12 boult boult NOUN brj-24033 123 13 2015 2015 NUM brj-24033 123 14 ) ) PUNCT brj-24033 123 15 has have VERB brj-24033 123 16 some some DET brj-24033 123 17 disadvantages disadvantage NOUN brj-24033 123 18 . . PUNCT brj-24033 124 1 first first ADV brj-24033 124 2 , , PUNCT brj-24033 124 3 a a DET brj-24033 124 4 constant constant ADJ brj-24033 124 5 threshold threshold NOUN brj-24033 124 6 𝜏 𝜏 NOUN brj-24033 124 7 is be AUX brj-24033 124 8 used use VERB brj-24033 124 9 for for ADP brj-24033 124 10 all all DET brj-24033 124 11 known know VERB brj-24033 124 12 classes class NOUN brj-24033 124 13 . . PUNCT brj-24033 125 1 assuming assume VERB brj-24033 125 2 that that SCONJ brj-24033 125 3 all all DET brj-24033 125 4 known know VERB brj-24033 125 5 classes class NOUN brj-24033 125 6 have have VERB brj-24033 125 7 sphere sphere NOUN brj-24033 125 8 distribution distribution NOUN brj-24033 125 9 structures structure NOUN brj-24033 125 10 , , PUNCT brj-24033 125 11 these these DET brj-24033 125 12 sphere sphere NOUN brj-24033 125 13 radii radius NOUN brj-24033 125 14 are be AUX brj-24033 125 15 usually usually ADV brj-24033 125 16 different different ADJ brj-24033 125 17 . . PUNCT brj-24033 126 1 therefore therefore ADV brj-24033 126 2 , , PUNCT brj-24033 126 3 different different ADJ brj-24033 126 4 threshold threshold NOUN brj-24033 126 5 𝜏 𝜏 DET brj-24033 126 6 values value NOUN brj-24033 126 7 should should AUX brj-24033 126 8 be be AUX brj-24033 126 9 used use VERB brj-24033 126 10 . . PUNCT brj-24033 127 1 second second ADJ brj-24033 127 2 , , PUNCT brj-24033 127 3 in in ADP brj-24033 127 4 practice practice NOUN brj-24033 127 5 , , PUNCT brj-24033 127 6 all all DET brj-24033 127 7 known know VERB brj-24033 127 8 classes class NOUN brj-24033 127 9 may may AUX brj-24033 127 10 have have VERB brj-24033 127 11 different different ADJ brj-24033 127 12 topological topological ADJ brj-24033 127 13 distribution distribution NOUN brj-24033 127 14 structures structure NOUN brj-24033 127 15 such such ADJ brj-24033 127 16 as as ADP brj-24033 127 17 sphere sphere NOUN brj-24033 127 18 structure structure NOUN brj-24033 127 19 and and CCONJ brj-24033 127 20 manifold manifold ADJ brj-24033 127 21 structure structure NOUN brj-24033 127 22 . . PUNCT brj-24033 128 1 the the DET brj-24033 128 2 manifold manifold ADJ brj-24033 128 3 structure structure NOUN brj-24033 128 4 can can AUX brj-24033 128 5 be be AUX brj-24033 128 6 usually usually ADV brj-24033 128 7 divided divide VERB brj-24033 128 8 into into ADP brj-24033 128 9 different different ADJ brj-24033 128 10 clusters cluster NOUN brj-24033 128 11 and and CCONJ brj-24033 128 12 this this DET brj-24033 128 13 division division NOUN brj-24033 128 14 can can AUX brj-24033 128 15 be be AUX brj-24033 128 16 fulfilled fulfil VERB brj-24033 128 17 by by ADP brj-24033 128 18 a a DET brj-24033 128 19 cluster cluster NOUN brj-24033 128 20 analysis analysis NOUN brj-24033 128 21 . . PUNCT brj-24033 129 1 to to PART brj-24033 129 2 overcome overcome VERB brj-24033 129 3 the the DET brj-24033 129 4 above above ADV brj-24033 129 5 - - PUNCT brj-24033 129 6 mentioned mention VERB brj-24033 129 7 two two NUM brj-24033 129 8 disadvantages disadvantage NOUN brj-24033 129 9 , , PUNCT brj-24033 129 10 an an DET brj-24033 129 11 improved improved ADJ brj-24033 129 12 nno nno NOUN brj-24033 129 13 classifier classifier PROPN brj-24033 129 14 version version PROPN brj-24033 129 15 was be AUX brj-24033 129 16 proposed propose VERB brj-24033 129 17 here here ADV brj-24033 129 18 . . PUNCT brj-24033 130 1 for for ADP brj-24033 130 2 every every DET brj-24033 130 3 known know VERB brj-24033 130 4 class class NOUN brj-24033 130 5 , , PUNCT brj-24033 130 6 a a DET brj-24033 130 7 density density NOUN brj-24033 130 8 peak peak NOUN brj-24033 130 9 clustering clustering NOUN brj-24033 130 10 ( ( PUNCT brj-24033 130 11 dpc dpc PROPN brj-24033 130 12 ) ) PUNCT brj-24033 130 13 algorithm algorithm NOUN brj-24033 130 14 is be AUX brj-24033 130 15 used use VERB brj-24033 130 16 to to PART brj-24033 130 17 perform perform VERB brj-24033 130 18 a a DET brj-24033 130 19 cluster cluster NOUN brj-24033 130 20 analysis analysis NOUN brj-24033 130 21 ( ( PUNCT brj-24033 130 22 rodriguez rodriguez NOUN brj-24033 130 23 and and CCONJ brj-24033 130 24 laio laio NOUN brj-24033 130 25 2014 2014 NUM brj-24033 130 26 ) ) PUNCT brj-24033 130 27 . . PUNCT brj-24033 131 1 in in ADP brj-24033 131 2 this this DET brj-24033 131 3 clustering clustering ADJ brj-24033 131 4 algorithm algorithm NOUN brj-24033 131 5 , , PUNCT brj-24033 131 6 the the DET brj-24033 131 7 number number NOUN brj-24033 131 8 of of ADP brj-24033 131 9 clusters cluster NOUN brj-24033 131 10 is be AUX brj-24033 131 11 not not PART brj-24033 131 12 required require VERB brj-24033 131 13 to to PART brj-24033 131 14 be be AUX brj-24033 131 15 determined determine VERB brj-24033 131 16 in in ADP brj-24033 131 17 advance advance NOUN brj-24033 131 18 , , PUNCT brj-24033 131 19 and and CCONJ brj-24033 131 20 this this DET brj-24033 131 21 number number NOUN brj-24033 131 22 can can AUX brj-24033 131 23 be be AUX brj-24033 131 24 determined determine VERB brj-24033 131 25 automatically automatically ADV brj-24033 131 26 in in ADP brj-24033 131 27 the the DET brj-24033 131 28 clustering clustering ADJ brj-24033 131 29 process process NOUN brj-24033 131 30 . . PUNCT brj-24033 132 1 this this DET brj-24033 132 2 algorithm algorithm NOUN brj-24033 132 3 is be AUX brj-24033 132 4 hardly hardly ADV brj-24033 132 5 influenced influence VERB brj-24033 132 6 by by ADP brj-24033 132 7 outliers outlier NOUN brj-24033 132 8 . . PUNCT brj-24033 133 1 the the DET brj-24033 133 2 detailed detailed ADJ brj-24033 133 3 clustering clustering ADJ brj-24033 133 4 procedure procedure NOUN brj-24033 133 5 is be AUX brj-24033 133 6 illustrated illustrate VERB brj-24033 133 7 as as SCONJ brj-24033 133 8 follows follow VERB brj-24033 133 9 . . PUNCT brj-24033 134 1 the the DET brj-24033 134 2 clustering cluster VERB brj-24033 134 3 centers center NOUN brj-24033 134 4 have have VERB brj-24033 134 5 relatively relatively ADV brj-24033 134 6 high high ADJ brj-24033 134 7 local local ADJ brj-24033 134 8 density density NOUN brj-24033 134 9 , , PUNCT brj-24033 134 10 and and CCONJ brj-24033 134 11 they they PRON brj-24033 134 12 are be AUX brj-24033 134 13 relatively relatively ADV brj-24033 134 14 far far ADV brj-24033 134 15 from from ADP brj-24033 134 16 those those DET brj-24033 134 17 points point NOUN brj-24033 134 18 with with ADP brj-24033 134 19 higher high ADJ brj-24033 134 20 local local ADJ brj-24033 134 21 density density NOUN brj-24033 134 22 . . PUNCT brj-24033 135 1 the the DET brj-24033 135 2 local local ADJ brj-24033 135 3 density density NOUN brj-24033 135 4 πœŒπ‘– πœŒπ‘– ADP brj-24033 135 5 and and CCONJ brj-24033 135 6 distance distance NOUN brj-24033 135 7 𝛿𝑖 𝛿𝑖 PROPN brj-24033 135 8 were be AUX brj-24033 135 9 calculated calculate VERB brj-24033 135 10 for for ADP brj-24033 135 11 each each DET brj-24033 135 12 sample sample NOUN brj-24033 135 13 𝝑𝑖. 𝝑𝑖. AUX brj-24033 135 14 a a DET brj-24033 135 15 local local ADJ brj-24033 135 16 density density NOUN brj-24033 135 17 of of ADP brj-24033 135 18 each each DET brj-24033 135 19 sample sample NOUN brj-24033 135 20 is be AUX brj-24033 135 21 calculated calculate VERB brj-24033 135 22 by by ADP brj-24033 135 23 either either CCONJ brj-24033 135 24 a a DET brj-24033 135 25 cutoff cutoff NOUN brj-24033 135 26 kernel kernel NOUN brj-24033 135 27 eq eq ADJ brj-24033 135 28 . . PROPN brj-24033 135 29 2 2 NUM brj-24033 135 30 or or CCONJ brj-24033 135 31 a a DET brj-24033 135 32 gaussian gaussian ADJ brj-24033 135 33 kernel kernel NOUN brj-24033 135 34 eq eq ADP brj-24033 135 35 . . PROPN brj-24033 135 36 3 3 NUM brj-24033 135 37 , , PUNCT brj-24033 135 38 πœŒπ‘– πœŒπ‘– X brj-24033 135 39 = = SYM brj-24033 135 40 βˆ‘ βˆ‘ PROPN brj-24033 135 41 πœ’(𝑑𝑖𝑗 πœ’(𝑑𝑖𝑗 PROPN brj-24033 135 42 βˆ’ βˆ’ PROPN brj-24033 135 43 𝑑𝑐)βˆ€π‘—β‰ π‘– 𝑑𝑐)βˆ€π‘—β‰ π‘– NOUN brj-24033 135 44 ( ( PUNCT brj-24033 135 45 2 2 NUM brj-24033 135 46 ) ) PUNCT brj-24033 135 47 πœŒπ‘– πœŒπ‘– X brj-24033 135 48 = = SYM brj-24033 135 49 βˆ‘ βˆ‘ NOUN brj-24033 135 50 𝑒π‘₯𝑝[βˆ’(𝑑𝑖𝑗/𝑑𝑐 𝑒π‘₯𝑝[βˆ’(𝑑𝑖𝑗/𝑑𝑐 ADJ brj-24033 135 51 ) ) PUNCT brj-24033 135 52 2 2 NUM brj-24033 135 53 βˆ€π‘—β‰ π‘– βˆ€π‘—β‰ π‘– NOUN brj-24033 135 54 ] ] X brj-24033 135 55 ( ( PUNCT brj-24033 135 56 3 3 X brj-24033 135 57 ) ) PUNCT brj-24033 135 58 πœ’(𝑑 πœ’(𝑑 NOUN brj-24033 135 59 ) ) PUNCT brj-24033 136 1 = = PRON brj-24033 136 2 { { PUNCT brj-24033 136 3 1 1 NUM brj-24033 136 4 , , PUNCT brj-24033 136 5 𝑑 𝑑 PROPN brj-24033 136 6 < < X brj-24033 136 7 0 0 NUM brj-24033 136 8 0 0 NUM brj-24033 136 9 , , PUNCT brj-24033 136 10 𝑑 𝑑 PROPN brj-24033 136 11 β‰₯ β‰₯ NOUN brj-24033 136 12 0 0 NUM brj-24033 136 13 ( ( PUNCT brj-24033 136 14 4 4 NUM brj-24033 136 15 ) ) PUNCT brj-24033 136 16 where where SCONJ brj-24033 136 17 𝑑𝑖𝑗 𝑑𝑖𝑗 PROPN brj-24033 136 18 is be AUX brj-24033 136 19 the the DET brj-24033 136 20 distance distance NOUN brj-24033 136 21 between between ADP brj-24033 136 22 sample sample NOUN brj-24033 136 23 𝝑𝑖 𝝑𝑖 ADV brj-24033 136 24 and and CCONJ brj-24033 136 25 sample sample NOUN brj-24033 136 26 𝝑𝑗 𝝑𝑗 NOUN brj-24033 136 27 ; ; PUNCT brj-24033 136 28 𝑑𝑐 𝑑𝑐 X brj-24033 136 29 is be AUX brj-24033 136 30 the the DET brj-24033 136 31 cut cut NOUN brj-24033 136 32 - - PUNCT brj-24033 136 33 off off ADP brj-24033 136 34 distance distance NOUN brj-24033 136 35 that that PRON brj-24033 136 36 is be AUX brj-24033 136 37 determined determine VERB brj-24033 136 38 in in ADP brj-24033 136 39 advance advance NOUN brj-24033 136 40 by by ADP brj-24033 136 41 an an DET brj-24033 136 42 expert expert NOUN brj-24033 136 43 ; ; PUNCT brj-24033 136 44 πœ’(𝑑 πœ’(𝑑 X brj-24033 136 45 ) ) PUNCT brj-24033 136 46 is be AUX brj-24033 136 47 a a DET brj-24033 136 48 0 0 NUM brj-24033 136 49 - - SYM brj-24033 136 50 1 1 NUM brj-24033 136 51 function function NOUN brj-24033 136 52 defined define VERB brj-24033 136 53 as as ADP brj-24033 136 54 eq eq NOUN brj-24033 136 55 . . PROPN brj-24033 136 56 4 4 X brj-24033 136 57 . . X brj-24033 137 1 therefore therefore ADV brj-24033 137 2 , , PUNCT brj-24033 137 3 the the DET brj-24033 137 4 πœŒπ‘– πœŒπ‘– NOUN brj-24033 137 5 computed compute VERB brj-24033 137 6 by by ADP brj-24033 137 7 eq eq PROPN brj-24033 137 8 . . PROPN brj-24033 137 9 2 2 NUM brj-24033 137 10 is be AUX brj-24033 137 11 a a DET brj-24033 137 12 discrete discrete ADJ brj-24033 137 13 value value NOUN brj-24033 137 14 , , PUNCT brj-24033 137 15 whereas whereas SCONJ brj-24033 137 16 that that PRON brj-24033 137 17 by by ADP brj-24033 137 18 eq eq NOUN brj-24033 137 19 . . PROPN brj-24033 137 20 3 3 NUM brj-24033 137 21 is be AUX brj-24033 137 22 a a DET brj-24033 137 23 continuous continuous ADJ brj-24033 137 24 value value NOUN brj-24033 137 25 . . PUNCT brj-24033 138 1 peer peer NOUN brj-24033 138 2 - - PUNCT brj-24033 138 3 reviewed review VERB brj-24033 138 4 article article NOUN brj-24033 138 5 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 138 6 zhang zhang PROPN brj-24033 138 7 & & CCONJ brj-24033 138 8 zhao zhao PROPN brj-24033 138 9 ( ( PUNCT brj-24033 138 10 2025 2025 NUM brj-24033 138 11 ) ) PUNCT brj-24033 138 12 . . PUNCT brj-24033 139 1 β€œ " PUNCT brj-24033 139 2 wood wood NOUN brj-24033 139 3 image image NOUN brj-24033 139 4 classification classification NOUN brj-24033 139 5 , , PUNCT brj-24033 139 6 ” " PUNCT brj-24033 139 7 bioresources bioresource NOUN brj-24033 139 8 20(1 20(1 NUM brj-24033 139 9 ) ) PUNCT brj-24033 139 10 , , PUNCT brj-24033 139 11 944 944 NUM brj-24033 139 12 - - SYM brj-24033 139 13 955 955 NUM brj-24033 139 14 . . PUNCT brj-24033 140 1 950 950 NUM brj-24033 140 2 the the DET brj-24033 140 3 𝛿𝑖 𝛿𝑖 PROPN brj-24033 140 4 is be AUX brj-24033 140 5 the the DET brj-24033 140 6 distance distance NOUN brj-24033 140 7 between between ADP brj-24033 140 8 a a DET brj-24033 140 9 sample sample NOUN brj-24033 140 10 𝝑𝑖 𝝑𝑖 ADV brj-24033 140 11 and and CCONJ brj-24033 140 12 its its PRON brj-24033 140 13 nearest near ADJ brj-24033 140 14 sample sample NOUN brj-24033 140 15 𝝑𝑗 𝝑𝑗 ADP brj-24033 140 16 among among ADP brj-24033 140 17 those those DET brj-24033 140 18 samples sample NOUN brj-24033 140 19 with with ADP brj-24033 140 20 higher high ADJ brj-24033 140 21 local local ADJ brj-24033 140 22 density density NOUN brj-24033 140 23 , , PUNCT brj-24033 140 24 as as SCONJ brj-24033 140 25 illustrated illustrate VERB brj-24033 140 26 by by ADP brj-24033 140 27 eq eq PROPN brj-24033 140 28 . . PROPN brj-24033 140 29 5 5 X brj-24033 140 30 . . X brj-24033 141 1 therefore therefore ADV brj-24033 141 2 , , PUNCT brj-24033 141 3 a a DET brj-24033 141 4 sample sample NOUN brj-24033 141 5 is be AUX brj-24033 141 6 possibly possibly ADV brj-24033 141 7 a a DET brj-24033 141 8 clustering clustering ADJ brj-24033 141 9 center center NOUN brj-24033 141 10 when when SCONJ brj-24033 141 11 it it PRON brj-24033 141 12 has have VERB brj-24033 141 13 a a DET brj-24033 141 14 relatively relatively ADV brj-24033 141 15 large large ADJ brj-24033 141 16 πœŒπ‘– πœŒπ‘– NOUN brj-24033 141 17 and and CCONJ brj-24033 141 18 𝛿𝑖 𝛿𝑖 NOUN brj-24033 141 19 so so SCONJ brj-24033 141 20 that that SCONJ brj-24033 141 21 the the DET brj-24033 141 22 probability probability NOUN brj-24033 141 23 of of ADP brj-24033 141 24 one one NUM brj-24033 141 25 sample sample NOUN brj-24033 141 26 being be AUX brj-24033 141 27 a a DET brj-24033 141 28 clustering clustering ADJ brj-24033 141 29 center center NOUN brj-24033 141 30 can can AUX brj-24033 141 31 be be AUX brj-24033 141 32 computed compute VERB brj-24033 141 33 by by ADP brj-24033 141 34 eq eq PROPN brj-24033 141 35 . . PROPN brj-24033 142 1 6 6 NUM brj-24033 142 2 . . PUNCT brj-24033 143 1 once once ADV brj-24033 143 2 the the DET brj-24033 143 3 clustering clustering ADJ brj-24033 143 4 centers center NOUN brj-24033 143 5 are be AUX brj-24033 143 6 determined determine VERB brj-24033 143 7 , , PUNCT brj-24033 143 8 a a DET brj-24033 143 9 detected detect VERB brj-24033 143 10 sample sample NOUN brj-24033 143 11 is be AUX brj-24033 143 12 classified classify VERB brj-24033 143 13 into into ADP brj-24033 143 14 the the DET brj-24033 143 15 cluster cluster NOUN brj-24033 143 16 which which PRON brj-24033 143 17 consists consist VERB brj-24033 143 18 of of ADP brj-24033 143 19 the the DET brj-24033 143 20 nearest near ADJ brj-24033 143 21 neighbor neighbor NOUN brj-24033 143 22 of of ADP brj-24033 143 23 this this DET brj-24033 143 24 sample sample NOUN brj-24033 143 25 with with ADP brj-24033 143 26 a a DET brj-24033 143 27 higher high ADJ brj-24033 143 28 local local ADJ brj-24033 143 29 density density NOUN brj-24033 143 30 . . PUNCT brj-24033 144 1 𝛿𝑖 𝛿𝑖 PUNCT brj-24033 145 1 = = NOUN brj-24033 145 2 π‘šπ‘–π‘›βˆ€π‘—,πœŒπ‘—>πœŒπ‘– π‘šπ‘–π‘›βˆ€π‘—,πœŒπ‘—>πœŒπ‘– X brj-24033 145 3 ( ( PUNCT brj-24033 145 4 𝑑𝑖𝑗 𝑑𝑖𝑗 PROPN brj-24033 145 5 ) ) PUNCT brj-24033 145 6 ( ( PUNCT brj-24033 145 7 5 5 X brj-24033 145 8 ) ) PUNCT brj-24033 145 9 𝛾𝑖 𝛾𝑖 NOUN brj-24033 145 10 = = NOUN brj-24033 145 11 πœŒπ‘–π›Ώπ‘– πœŒπ‘–π›Ώπ‘– NOUN brj-24033 145 12 ( ( PUNCT brj-24033 145 13 6 6 NUM brj-24033 145 14 ) ) PUNCT brj-24033 145 15 the the DET brj-24033 145 16 dpc dpc PROPN brj-24033 145 17 algorithm algorithm NOUN brj-24033 145 18 is be AUX brj-24033 145 19 applied apply VERB brj-24033 145 20 to to ADP brj-24033 145 21 the the DET brj-24033 145 22 spectral spectral ADJ brj-24033 145 23 vectors vector NOUN brj-24033 145 24 after after ADP brj-24033 145 25 spectral spectral ADJ brj-24033 145 26 dimension dimension NOUN brj-24033 145 27 reduction reduction NOUN brj-24033 145 28 by by ADP brj-24033 145 29 using use VERB brj-24033 145 30 the the DET brj-24033 145 31 above above ADJ brj-24033 145 32 ml ml ADP brj-24033 145 33 algorithm algorithm NOUN brj-24033 145 34 ( ( PUNCT brj-24033 145 35 mensink mensink NOUN brj-24033 145 36 et et PROPN brj-24033 145 37 al al PROPN brj-24033 145 38 . . PROPN brj-24033 145 39 2013 2013 NUM brj-24033 145 40 ) ) PUNCT brj-24033 145 41 . . PUNCT brj-24033 146 1 this this DET brj-24033 146 2 spectral spectral ADJ brj-24033 146 3 vector vector NOUN brj-24033 146 4 is be AUX brj-24033 146 5 denoted denote VERB brj-24033 146 6 as as ADP brj-24033 146 7 οΏ½ οΏ½ PROPN brj-24033 146 8 Μƒ Μƒ PROPN brj-24033 146 9 οΏ½ οΏ½ NOUN brj-24033 146 10 = = SYM brj-24033 146 11 𝑾𝒙. 𝑾𝒙. PROPN brj-24033 146 12 in in ADP brj-24033 146 13 the the DET brj-24033 146 14 following follow VERB brj-24033 146 15 wood wood NOUN brj-24033 146 16 spectral spectral ADJ brj-24033 146 17 classification classification NOUN brj-24033 146 18 experiments experiment NOUN brj-24033 146 19 , , PUNCT brj-24033 146 20 1 1 NUM brj-24033 146 21 to to PART brj-24033 146 22 3 3 NUM brj-24033 146 23 clusters cluster NOUN brj-24033 146 24 were be AUX brj-24033 146 25 obtained obtain VERB brj-24033 146 26 for for ADP brj-24033 146 27 each each DET brj-24033 146 28 known know VERB brj-24033 146 29 wood wood NOUN brj-24033 146 30 species specie NOUN brj-24033 146 31 . . PUNCT brj-24033 147 1 all all DET brj-24033 147 2 these these DET brj-24033 147 3 clusters cluster NOUN brj-24033 147 4 can can AUX brj-24033 147 5 be be AUX brj-24033 147 6 approximated approximate VERB brj-24033 147 7 by by ADP brj-24033 147 8 spheres sphere NOUN brj-24033 147 9 with with ADP brj-24033 147 10 different different ADJ brj-24033 147 11 radii radius NOUN brj-24033 147 12 𝜏 𝜏 NOUN brj-24033 147 13 ( ( PUNCT brj-24033 147 14 i.e. i.e. X brj-24033 147 15 , , PUNCT brj-24033 147 16 this this DET brj-24033 147 17 𝜏 𝜏 NOUN brj-24033 147 18 is be AUX brj-24033 147 19 a a DET brj-24033 147 20 threshold threshold NOUN brj-24033 147 21 for for ADP brj-24033 147 22 a a DET brj-24033 147 23 cluster cluster NOUN brj-24033 147 24 sphere sphere ADV brj-24033 147 25 ) ) PUNCT brj-24033 147 26 . . PUNCT brj-24033 148 1 then then ADV brj-24033 148 2 the the DET brj-24033 148 3 original original ADJ brj-24033 148 4 nno nno NOUN brj-24033 148 5 classification classification NOUN brj-24033 148 6 is be AUX brj-24033 148 7 performed perform VERB brj-24033 148 8 by by ADP brj-24033 148 9 using use VERB brj-24033 148 10 these these DET brj-24033 148 11 clusters cluster NOUN brj-24033 148 12 with with ADP brj-24033 148 13 different different ADJ brj-24033 148 14 sizes size NOUN brj-24033 148 15 . . PUNCT brj-24033 149 1 please please INTJ brj-24033 149 2 note note VERB brj-24033 149 3 that that SCONJ brj-24033 149 4 the the DET brj-24033 149 5 𝑑𝑾(𝒙 𝑑𝑾(𝒙 NUM brj-24033 149 6 , , PUNCT brj-24033 149 7 𝝁𝑦 𝝁𝑦 NOUN brj-24033 149 8 ) ) PUNCT brj-24033 149 9 in in ADP brj-24033 149 10 eq eq NOUN brj-24033 149 11 . . PROPN brj-24033 149 12 1 1 NUM brj-24033 149 13 is be AUX brj-24033 149 14 computed compute VERB brj-24033 149 15 by by ADP brj-24033 149 16 a a DET brj-24033 149 17 mahalanobis mahalanobis ADJ brj-24033 149 18 distance distance NOUN brj-24033 149 19 instead instead ADV brj-24033 149 20 of of ADP brj-24033 149 21 euclidean euclidean ADJ brj-24033 149 22 distance distance NOUN brj-24033 149 23 , , PUNCT brj-24033 149 24 as as SCONJ brj-24033 149 25 illustrated illustrate VERB brj-24033 149 26 by by ADP brj-24033 149 27 eq eq PROPN brj-24033 149 28 . . PROPN brj-24033 149 29 7 7 NUM brj-24033 149 30 , , PUNCT brj-24033 149 31 𝑑𝑾(𝒙 𝑑𝑾(𝒙 NUM brj-24033 149 32 , , PUNCT brj-24033 149 33 𝝁𝑦 𝝁𝑦 NOUN brj-24033 149 34 ) ) PUNCT brj-24033 149 35 = = PUNCT brj-24033 150 1 ‖𝑾𝒙 ‖𝑾𝒙 ADJ brj-24033 150 2 βˆ’ βˆ’ NUM brj-24033 150 3 𝑾𝝁𝑦‖ 𝑾𝝁𝑦‖ PROPN brj-24033 150 4 2 2 NUM brj-24033 150 5 = = SYM brj-24033 150 6 β€– β€– PROPN brj-24033 150 7 οΏ½ οΏ½ PROPN brj-24033 150 8 Μƒ Μƒ PROPN brj-24033 150 9 οΏ½ οΏ½ PROPN brj-24033 150 10 βˆ’ βˆ’ PROPN brj-24033 150 11 οΏ½ οΏ½ PROPN brj-24033 150 12 Μƒ Μƒ PROPN brj-24033 150 13 οΏ½ οΏ½ NOUN brj-24033 150 14 𝑦‖ 𝑦‖ ADJ brj-24033 150 15 2 2 NUM brj-24033 150 16 = = SYM brj-24033 150 17 √ √ PROPN brj-24033 150 18 ( ( PUNCT brj-24033 150 19 οΏ½ οΏ½ PROPN brj-24033 150 20 Μƒ Μƒ PROPN brj-24033 150 21 οΏ½ οΏ½ PROPN brj-24033 150 22 βˆ’ βˆ’ PROPN brj-24033 150 23 οΏ½ οΏ½ PROPN brj-24033 150 24 Μƒ Μƒ PROPN brj-24033 150 25 οΏ½ οΏ½ NOUN brj-24033 150 26 𝑦 𝑦 NOUN brj-24033 150 27 ) ) PUNCT brj-24033 150 28 𝑇 𝑇 PROPN brj-24033 150 29 π‘ͺβˆ’πŸ π‘ͺβˆ’πŸ NOUN brj-24033 150 30 ( ( PUNCT brj-24033 150 31 οΏ½ οΏ½ PROPN brj-24033 150 32 Μƒ Μƒ PROPN brj-24033 150 33 οΏ½ οΏ½ PROPN brj-24033 150 34 βˆ’ βˆ’ PROPN brj-24033 150 35 οΏ½ οΏ½ PROPN brj-24033 150 36 Μƒ Μƒ PROPN brj-24033 150 37 οΏ½ οΏ½ NOUN brj-24033 150 38 𝑦 𝑦 NOUN brj-24033 150 39 ) ) PUNCT brj-24033 150 40 ( ( PUNCT brj-24033 150 41 7 7 X brj-24033 150 42 ) ) PUNCT brj-24033 150 43 where where SCONJ brj-24033 150 44 οΏ½ οΏ½ PROPN brj-24033 150 45 Μƒ Μƒ PROPN brj-24033 150 46 οΏ½ οΏ½ NOUN brj-24033 150 47 𝑦 𝑦 NOUN brj-24033 150 48 = = PUNCT brj-24033 150 49 𝑾𝝁𝑦 𝑾𝝁𝑦 PROPN brj-24033 150 50 is be AUX brj-24033 150 51 a a DET brj-24033 150 52 cluster cluster NOUN brj-24033 150 53 center center NOUN brj-24033 150 54 after after ADP brj-24033 150 55 spectral spectral ADJ brj-24033 150 56 dimension dimension NOUN brj-24033 150 57 reduction reduction NOUN brj-24033 150 58 ; ; PUNCT brj-24033 150 59 π‘ͺ π‘ͺ PROPN brj-24033 150 60 is be AUX brj-24033 150 61 the the DET brj-24033 150 62 covariance covariance NOUN brj-24033 150 63 matrix matrix NOUN brj-24033 150 64 for for ADP brj-24033 150 65 the the DET brj-24033 150 66 cluster cluster NOUN brj-24033 150 67 whose whose DET brj-24033 150 68 cluster cluster NOUN brj-24033 150 69 center center NOUN brj-24033 150 70 is be AUX brj-24033 150 71 οΏ½ οΏ½ PROPN brj-24033 150 72 Μƒ Μƒ PROPN brj-24033 150 73 οΏ½ οΏ½ PROPN brj-24033 150 74 𝑦. 𝑦. NOUN brj-24033 150 75 the the DET brj-24033 150 76 different different ADJ brj-24033 150 77 thresholds threshold NOUN brj-24033 150 78 𝜏 𝜏 X brj-24033 150 79 of of ADP brj-24033 150 80 different different ADJ brj-24033 150 81 clusters cluster NOUN brj-24033 150 82 can can AUX brj-24033 150 83 be be AUX brj-24033 150 84 obtained obtain VERB brj-24033 150 85 by by ADP brj-24033 150 86 an an DET brj-24033 150 87 optimal optimal ADJ brj-24033 150 88 grid grid NOUN brj-24033 150 89 search search NOUN brj-24033 150 90 . . PUNCT brj-24033 151 1 more more ADV brj-24033 151 2 accurate accurate ADJ brj-24033 151 3 classification classification NOUN brj-24033 151 4 is be AUX brj-24033 151 5 achieved achieve VERB brj-24033 151 6 because because SCONJ brj-24033 151 7 the the DET brj-24033 151 8 nno nno PROPN brj-24033 151 9 classifier classifier NOUN brj-24033 151 10 is be AUX brj-24033 151 11 applied apply VERB brj-24033 151 12 with with ADP brj-24033 151 13 those those DET brj-24033 151 14 extracted extract VERB brj-24033 151 15 clusters cluster NOUN brj-24033 151 16 with with ADP brj-24033 151 17 different different ADJ brj-24033 151 18 radius radius NOUN brj-24033 151 19 thresholds threshold NOUN brj-24033 151 20 . . PUNCT brj-24033 152 1 however however ADV brj-24033 152 2 , , PUNCT brj-24033 152 3 the the DET brj-24033 152 4 original original ADJ brj-24033 152 5 nno nno NOUN brj-24033 152 6 classifier classifier NOUN brj-24033 152 7 is be AUX brj-24033 152 8 applied apply VERB brj-24033 152 9 with with ADP brj-24033 152 10 those those DET brj-24033 152 11 original original ADJ brj-24033 152 12 known know VERB brj-24033 152 13 classes class NOUN brj-24033 152 14 with with ADP brj-24033 152 15 a a DET brj-24033 152 16 same same ADJ brj-24033 152 17 radius radius NOUN brj-24033 152 18 𝜏 𝜏 NOUN brj-24033 152 19 , , PUNCT brj-24033 152 20 even even ADV brj-24033 152 21 some some DET brj-24033 152 22 classes class NOUN brj-24033 152 23 may may AUX brj-24033 152 24 have have VERB brj-24033 152 25 manifold manifold ADJ brj-24033 152 26 distribution distribution NOUN brj-24033 152 27 structures structure NOUN brj-24033 152 28 . . PUNCT brj-24033 153 1 this this DET brj-24033 153 2 situation situation NOUN brj-24033 153 3 may may AUX brj-24033 153 4 produce produce VERB brj-24033 153 5 large large ADJ brj-24033 153 6 classification classification NOUN brj-24033 153 7 errors error NOUN brj-24033 153 8 . . PUNCT brj-24033 154 1 results result NOUN brj-24033 154 2 and and CCONJ brj-24033 154 3 discussion discussion NOUN brj-24033 154 4 classification classification NOUN brj-24033 154 5 performance performance NOUN brj-24033 154 6 evaluations evaluation NOUN brj-24033 154 7 performance performance NOUN brj-24033 154 8 evaluation evaluation NOUN brj-24033 154 9 measures measure NOUN brj-24033 154 10 play play VERB brj-24033 154 11 an an DET brj-24033 154 12 important important ADJ brj-24033 154 13 role role NOUN brj-24033 154 14 in in ADP brj-24033 154 15 judging judge VERB brj-24033 154 16 the the DET brj-24033 154 17 classification classification NOUN brj-24033 154 18 performance performance NOUN brj-24033 154 19 of of ADP brj-24033 154 20 the the DET brj-24033 154 21 osr osr PROPN brj-24033 154 22 classifier classifier NOUN brj-24033 154 23 . . PUNCT brj-24033 155 1 to to PART brj-24033 155 2 comprehensively comprehensively ADV brj-24033 155 3 assess assess VERB brj-24033 155 4 the the DET brj-24033 155 5 classification classification NOUN brj-24033 155 6 performance performance NOUN brj-24033 155 7 in in ADP brj-24033 155 8 osr osr PROPN brj-24033 155 9 , , PUNCT brj-24033 155 10 three three NUM brj-24033 155 11 measures measure NOUN brj-24033 155 12 such such ADJ brj-24033 155 13 as as ADP brj-24033 155 14 f f NOUN brj-24033 155 15 - - PUNCT brj-24033 155 16 score score NOUN brj-24033 155 17 , , PUNCT brj-24033 155 18 kappa kappa ADJ brj-24033 155 19 coefficient coefficient NOUN brj-24033 155 20 , , PUNCT brj-24033 155 21 and and CCONJ brj-24033 155 22 overall overall ADJ brj-24033 155 23 recognition recognition NOUN brj-24033 155 24 accuracy accuracy NOUN brj-24033 155 25 ( ( PUNCT brj-24033 155 26 ora ora INTJ brj-24033 155 27 ) ) PUNCT brj-24033 155 28 are be AUX brj-24033 155 29 used use VERB brj-24033 155 30 . . PUNCT brj-24033 156 1 the the DET brj-24033 156 2 f f NOUN brj-24033 156 3 - - PUNCT brj-24033 156 4 score score NOUN brj-24033 156 5 computation computation NOUN brj-24033 156 6 is be AUX brj-24033 156 7 based base VERB brj-24033 156 8 on on ADP brj-24033 156 9 the the DET brj-24033 156 10 precision precision NOUN brj-24033 156 11 and and CCONJ brj-24033 156 12 recall recall NOUN brj-24033 156 13 , , PUNCT brj-24033 156 14 as as SCONJ brj-24033 156 15 illustrated illustrate VERB brj-24033 156 16 in in ADP brj-24033 156 17 eqs eqs PROPN brj-24033 156 18 . . PROPN brj-24033 156 19 8 8 NUM brj-24033 156 20 to to PART brj-24033 156 21 10 10 NUM brj-24033 156 22 . . PUNCT brj-24033 157 1 here here ADV brj-24033 157 2 π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ NOUN brj-24033 157 3 represents represent VERB brj-24033 157 4 the the DET brj-24033 157 5 number number NOUN brj-24033 157 6 of of ADP brj-24033 157 7 correctly correctly ADV brj-24033 157 8 classified classify VERB brj-24033 157 9 samples sample NOUN brj-24033 157 10 in in ADP brj-24033 157 11 known known ADJ brj-24033 157 12 classes class NOUN brj-24033 157 13 , , PUNCT brj-24033 157 14 whereas whereas SCONJ brj-24033 157 15 π‘π‘œπ‘’π‘›π‘‘π‘˜_𝑒 π‘π‘œπ‘’π‘›π‘‘π‘˜_𝑒 ADV brj-24033 157 16 and and CCONJ brj-24033 157 17 π‘π‘œπ‘’π‘›π‘‘π‘’π‘˜_𝑒 π‘π‘œπ‘’π‘›π‘‘π‘’π‘˜_𝑒 X brj-24033 157 18 represent represent VERB brj-24033 157 19 the the DET brj-24033 157 20 number number NOUN brj-24033 157 21 of of ADP brj-24033 157 22 misclassified misclassifie VERB brj-24033 157 23 samples sample NOUN brj-24033 157 24 in in ADP brj-24033 157 25 known know VERB brj-24033 157 26 and and CCONJ brj-24033 157 27 unknown unknown ADJ brj-24033 157 28 classes class NOUN brj-24033 157 29 , , PUNCT brj-24033 157 30 respectively respectively ADV brj-24033 157 31 . . PUNCT brj-24033 157 32 precision precision NOUN brj-24033 157 33 = = SYM brj-24033 157 34 π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ/(π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ/(π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ X brj-24033 157 35 + + X brj-24033 157 36 π‘π‘œπ‘’π‘›π‘‘π‘’π‘˜_𝑒 π‘π‘œπ‘’π‘›π‘‘π‘’π‘˜_𝑒 X brj-24033 157 37 ) ) PUNCT brj-24033 157 38 ( ( PUNCT brj-24033 157 39 8) 8) NUM brj-24033 157 40 recall recall NOUN brj-24033 157 41 = = SYM brj-24033 157 42 π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ/(π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ/(π‘π‘œπ‘’π‘›π‘‘π‘˜_π‘Ÿ X brj-24033 157 43 + + PUNCT brj-24033 157 44 π‘π‘œπ‘’π‘›π‘‘π‘˜_𝑒 π‘π‘œπ‘’π‘›π‘‘π‘˜_𝑒 NUM brj-24033 157 45 ) ) PUNCT brj-24033 157 46 ( ( PUNCT brj-24033 157 47 9 9 X brj-24033 157 48 ) ) PUNCT brj-24033 157 49 f f NOUN brj-24033 157 50 βˆ’ βˆ’ NOUN brj-24033 157 51 score score NOUN brj-24033 157 52 = = NOUN brj-24033 157 53 2precisionβˆ™recall 2precisionβˆ™recall NUM brj-24033 157 54 precision+recall precision+recall PROPN brj-24033 157 55 ( ( PUNCT brj-24033 157 56 10 10 NUM brj-24033 157 57 ) ) PUNCT brj-24033 157 58 peer peer NOUN brj-24033 157 59 - - PUNCT brj-24033 157 60 reviewed review VERB brj-24033 157 61 article article NOUN brj-24033 157 62 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 158 1 zhang zhang PROPN brj-24033 158 2 & & CCONJ brj-24033 158 3 zhao zhao PROPN brj-24033 158 4 ( ( PUNCT brj-24033 158 5 2025 2025 NUM brj-24033 158 6 ) ) PUNCT brj-24033 158 7 . . PUNCT brj-24033 159 1 β€œ " PUNCT brj-24033 159 2 wood wood NOUN brj-24033 159 3 image image NOUN brj-24033 159 4 classification classification NOUN brj-24033 159 5 , , PUNCT brj-24033 159 6 ” " PUNCT brj-24033 159 7 bioresources bioresource NOUN brj-24033 159 8 20(1 20(1 NUM brj-24033 159 9 ) ) PUNCT brj-24033 159 10 , , PUNCT brj-24033 159 11 944 944 NUM brj-24033 159 12 - - SYM brj-24033 159 13 955 955 NUM brj-24033 159 14 . . PUNCT brj-24033 160 1 951 951 NUM brj-24033 160 2 dataset dataset ADJ brj-24033 160 3 partition partition NOUN brj-24033 160 4 there there PRON brj-24033 160 5 are be VERB brj-24033 160 6 35 35 NUM brj-24033 160 7 wood wood NOUN brj-24033 160 8 species specie NOUN brj-24033 160 9 in in ADP brj-24033 160 10 total total NOUN brj-24033 160 11 in in ADP brj-24033 160 12 the the DET brj-24033 160 13 wood wood NOUN brj-24033 160 14 dataset dataset NOUN brj-24033 160 15 as as SCONJ brj-24033 160 16 illustrated illustrate VERB brj-24033 160 17 in in ADP brj-24033 160 18 table table NOUN brj-24033 160 19 1 1 NUM brj-24033 160 20 , , PUNCT brj-24033 160 21 and and CCONJ brj-24033 160 22 each each DET brj-24033 160 23 wood wood NOUN brj-24033 160 24 species specie NOUN brj-24033 160 25 consists consist VERB brj-24033 160 26 of of ADP brj-24033 160 27 50 50 NUM brj-24033 160 28 spectral spectral ADJ brj-24033 160 29 samples sample NOUN brj-24033 160 30 . . PUNCT brj-24033 161 1 the the DET brj-24033 161 2 wood wood NOUN brj-24033 161 3 dataset dataset NOUN brj-24033 161 4 is be AUX brj-24033 161 5 divided divide VERB brj-24033 161 6 into into ADP brj-24033 161 7 3 3 NUM brj-24033 161 8 groups group NOUN brj-24033 161 9 to to PART brj-24033 161 10 testify testify VERB brj-24033 161 11 the the DET brj-24033 161 12 proposed propose VERB brj-24033 161 13 improved improve VERB brj-24033 161 14 nno nno NOUN brj-24033 161 15 classifier classifier NOUN brj-24033 161 16 in in ADP brj-24033 161 17 open open ADJ brj-24033 161 18 set set VERB brj-24033 161 19 scenario scenario NOUN brj-24033 161 20 . . PUNCT brj-24033 162 1 these these DET brj-24033 162 2 3 3 NUM brj-24033 162 3 groups group NOUN brj-24033 162 4 are be AUX brj-24033 162 5 explained explain VERB brj-24033 162 6 as as SCONJ brj-24033 162 7 follows follow VERB brj-24033 162 8 . . PUNCT brj-24033 163 1 group group NOUN brj-24033 163 2 1 1 NUM brj-24033 163 3 : : PUNCT brj-24033 163 4 the the DET brj-24033 163 5 initial initial ADJ brj-24033 163 6 5 5 NUM brj-24033 163 7 wood wood NOUN brj-24033 163 8 species specie NOUN brj-24033 163 9 are be AUX brj-24033 163 10 selected select VERB brj-24033 163 11 randomly randomly ADV brj-24033 163 12 as as ADP brj-24033 163 13 the the DET brj-24033 163 14 known know VERB brj-24033 163 15 species specie NOUN brj-24033 163 16 to to PART brj-24033 163 17 form form VERB brj-24033 163 18 the the DET brj-24033 163 19 training training NOUN brj-24033 163 20 set set NOUN brj-24033 163 21 of of ADP brj-24033 163 22 the the DET brj-24033 163 23 nno nno PROPN brj-24033 163 24 classifier classifier NOUN brj-24033 163 25 , , PUNCT brj-24033 163 26 and and CCONJ brj-24033 163 27 these these DET brj-24033 163 28 5 5 NUM brj-24033 163 29 wood wood NOUN brj-24033 163 30 species specie NOUN brj-24033 163 31 are be AUX brj-24033 163 32 used use VERB brj-24033 163 33 in in ADP brj-24033 163 34 the the DET brj-24033 163 35 ml ml PROPN brj-24033 163 36 algorithm algorithm NOUN brj-24033 163 37 ( ( PUNCT brj-24033 163 38 mensink mensink NOUN brj-24033 163 39 et et PROPN brj-24033 163 40 al al PROPN brj-24033 163 41 . . PROPN brj-24033 163 42 2013 2013 NUM brj-24033 163 43 ) ) PUNCT brj-24033 163 44 to to PART brj-24033 163 45 obtain obtain VERB brj-24033 163 46 the the DET brj-24033 163 47 projection projection NOUN brj-24033 163 48 matrix matrix NOUN brj-24033 163 49 w. w. NOUN brj-24033 163 50 then then ADV brj-24033 163 51 in in ADP brj-24033 163 52 the the DET brj-24033 163 53 incremental incremental ADJ brj-24033 163 54 learning learning NOUN brj-24033 163 55 process process NOUN brj-24033 163 56 , , PUNCT brj-24033 163 57 another another DET brj-24033 163 58 5 5 NUM brj-24033 163 59 wood wood NOUN brj-24033 163 60 species specie NOUN brj-24033 163 61 are be AUX brj-24033 163 62 selected select VERB brj-24033 163 63 randomly randomly ADV brj-24033 163 64 and and CCONJ brj-24033 163 65 added add VERB brj-24033 163 66 into into ADP brj-24033 163 67 the the DET brj-24033 163 68 training training NOUN brj-24033 163 69 set set NOUN brj-24033 163 70 of of ADP brj-24033 163 71 the the DET brj-24033 163 72 improved improved ADJ brj-24033 163 73 nno nno NOUN brj-24033 163 74 classifier classifier NOUN brj-24033 163 75 . . PUNCT brj-24033 164 1 this this DET brj-24033 164 2 incremental incremental ADJ brj-24033 164 3 learning learning NOUN brj-24033 164 4 process process NOUN brj-24033 164 5 is be AUX brj-24033 164 6 repeated repeat VERB brj-24033 164 7 for for ADP brj-24033 164 8 4 4 NUM brj-24033 164 9 times time NOUN brj-24033 164 10 . . PUNCT brj-24033 165 1 finally finally ADV brj-24033 165 2 , , PUNCT brj-24033 165 3 the the DET brj-24033 165 4 training training NOUN brj-24033 165 5 dataset dataset NOUN brj-24033 165 6 consists consist VERB brj-24033 165 7 of of ADP brj-24033 165 8 25 25 NUM brj-24033 165 9 known know VERB brj-24033 165 10 wood wood NOUN brj-24033 165 11 species specie NOUN brj-24033 165 12 . . PUNCT brj-24033 166 1 the the DET brj-24033 166 2 unknown unknown ADJ brj-24033 166 3 wood wood NOUN brj-24033 166 4 dataset dataset NOUN brj-24033 166 5 consists consist VERB brj-24033 166 6 of of ADP brj-24033 166 7 5 5 NUM brj-24033 166 8 wood wood NOUN brj-24033 166 9 species specie NOUN brj-24033 166 10 . . PUNCT brj-24033 167 1 this this DET brj-24033 167 2 wood wood NOUN brj-24033 167 3 dataset dataset NOUN brj-24033 167 4 partition partition NOUN brj-24033 167 5 is be AUX brj-24033 167 6 illustrated illustrate VERB brj-24033 167 7 in in ADP brj-24033 167 8 table table NOUN brj-24033 167 9 2 2 NUM brj-24033 167 10 . . PUNCT brj-24033 167 11 group group NOUN brj-24033 167 12 2 2 NUM brj-24033 167 13 : : PUNCT brj-24033 167 14 the the DET brj-24033 167 15 initial initial ADJ brj-24033 167 16 5 5 NUM brj-24033 167 17 wood wood NOUN brj-24033 167 18 species specie NOUN brj-24033 167 19 are be AUX brj-24033 167 20 selected select VERB brj-24033 167 21 randomly randomly ADV brj-24033 167 22 as as ADP brj-24033 167 23 the the DET brj-24033 167 24 known know VERB brj-24033 167 25 species specie NOUN brj-24033 167 26 to to PART brj-24033 167 27 form form VERB brj-24033 167 28 the the DET brj-24033 167 29 training training NOUN brj-24033 167 30 set set NOUN brj-24033 167 31 , , PUNCT brj-24033 167 32 and and CCONJ brj-24033 167 33 these these DET brj-24033 167 34 5 5 NUM brj-24033 167 35 wood wood NOUN brj-24033 167 36 species specie NOUN brj-24033 167 37 are be AUX brj-24033 167 38 used use VERB brj-24033 167 39 in in ADP brj-24033 167 40 the the DET brj-24033 167 41 ml ml PROPN brj-24033 167 42 algorithm algorithm NOUN brj-24033 167 43 ( ( PUNCT brj-24033 167 44 mensink mensink NOUN brj-24033 167 45 et et PROPN brj-24033 167 46 al al PROPN brj-24033 167 47 . . PROPN brj-24033 167 48 2013 2013 NUM brj-24033 167 49 ) ) PUNCT brj-24033 167 50 to to PART brj-24033 167 51 obtain obtain VERB brj-24033 167 52 the the DET brj-24033 167 53 projection projection NOUN brj-24033 167 54 matrix matrix NOUN brj-24033 167 55 w. w. NOUN brj-24033 167 56 then then ADV brj-24033 167 57 in in ADP brj-24033 167 58 the the DET brj-24033 167 59 incremental incremental ADJ brj-24033 167 60 learning learning NOUN brj-24033 167 61 process process NOUN brj-24033 167 62 , , PUNCT brj-24033 167 63 another another DET brj-24033 167 64 5 5 NUM brj-24033 167 65 wood wood NOUN brj-24033 167 66 species specie NOUN brj-24033 167 67 are be AUX brj-24033 167 68 added add VERB brj-24033 167 69 into into ADP brj-24033 167 70 the the DET brj-24033 167 71 training training NOUN brj-24033 167 72 set set NOUN brj-24033 167 73 . . PUNCT brj-24033 168 1 this this DET brj-24033 168 2 incremental incremental ADJ brj-24033 168 3 learning learning NOUN brj-24033 168 4 process process NOUN brj-24033 168 5 is be AUX brj-24033 168 6 repeated repeat VERB brj-24033 168 7 for for ADP brj-24033 168 8 3 3 NUM brj-24033 168 9 times time NOUN brj-24033 168 10 . . PUNCT brj-24033 169 1 finally finally ADV brj-24033 169 2 , , PUNCT brj-24033 169 3 the the DET brj-24033 169 4 training training NOUN brj-24033 169 5 dataset dataset NOUN brj-24033 169 6 consists consist VERB brj-24033 169 7 of of ADP brj-24033 169 8 20 20 NUM brj-24033 169 9 known know VERB brj-24033 169 10 wood wood NOUN brj-24033 169 11 species specie NOUN brj-24033 169 12 . . PUNCT brj-24033 170 1 the the DET brj-24033 170 2 unknown unknown ADJ brj-24033 170 3 wood wood NOUN brj-24033 170 4 dataset dataset NOUN brj-24033 170 5 consists consist VERB brj-24033 170 6 of of ADP brj-24033 170 7 10 10 NUM brj-24033 170 8 wood wood NOUN brj-24033 170 9 species specie NOUN brj-24033 170 10 . . PUNCT brj-24033 171 1 this this DET brj-24033 171 2 wood wood NOUN brj-24033 171 3 dataset dataset NOUN brj-24033 171 4 partition partition NOUN brj-24033 171 5 is be AUX brj-24033 171 6 illustrated illustrate VERB brj-24033 171 7 in in ADP brj-24033 171 8 table table NOUN brj-24033 171 9 3 3 NUM brj-24033 171 10 . . PUNCT brj-24033 171 11 group group NOUN brj-24033 171 12 3 3 NUM brj-24033 171 13 : : PUNCT brj-24033 171 14 the the DET brj-24033 171 15 initial initial ADJ brj-24033 171 16 10 10 NUM brj-24033 171 17 wood wood NOUN brj-24033 171 18 species specie NOUN brj-24033 171 19 are be AUX brj-24033 171 20 selected select VERB brj-24033 171 21 randomly randomly ADV brj-24033 171 22 as as ADP brj-24033 171 23 the the DET brj-24033 171 24 known know VERB brj-24033 171 25 species specie NOUN brj-24033 171 26 to to PART brj-24033 171 27 form form VERB brj-24033 171 28 the the DET brj-24033 171 29 training training NOUN brj-24033 171 30 set set NOUN brj-24033 171 31 , , PUNCT brj-24033 171 32 and and CCONJ brj-24033 171 33 these these DET brj-24033 171 34 10 10 NUM brj-24033 171 35 wood wood NOUN brj-24033 171 36 species specie NOUN brj-24033 171 37 are be AUX brj-24033 171 38 used use VERB brj-24033 171 39 in in ADP brj-24033 171 40 the the DET brj-24033 171 41 ml ml PROPN brj-24033 171 42 algorithm algorithm NOUN brj-24033 171 43 ( ( PUNCT brj-24033 171 44 mensink mensink NOUN brj-24033 171 45 et et PROPN brj-24033 171 46 al al PROPN brj-24033 171 47 . . PROPN brj-24033 171 48 2013 2013 NUM brj-24033 171 49 ) ) PUNCT brj-24033 171 50 to to PART brj-24033 171 51 obtain obtain VERB brj-24033 171 52 the the DET brj-24033 171 53 projection projection NOUN brj-24033 171 54 matrix matrix NOUN brj-24033 171 55 w. w. NOUN brj-24033 171 56 then then ADV brj-24033 171 57 in in ADP brj-24033 171 58 the the DET brj-24033 171 59 incremental incremental ADJ brj-24033 171 60 learning learning NOUN brj-24033 171 61 process process NOUN brj-24033 171 62 , , PUNCT brj-24033 171 63 another another DET brj-24033 171 64 5 5 NUM brj-24033 171 65 wood wood NOUN brj-24033 171 66 species specie NOUN brj-24033 171 67 are be AUX brj-24033 171 68 added add VERB brj-24033 171 69 into into ADP brj-24033 171 70 the the DET brj-24033 171 71 training training NOUN brj-24033 171 72 set set NOUN brj-24033 171 73 . . PUNCT brj-24033 172 1 this this DET brj-24033 172 2 incremental incremental ADJ brj-24033 172 3 learning learning NOUN brj-24033 172 4 process process NOUN brj-24033 172 5 is be AUX brj-24033 172 6 repeated repeat VERB brj-24033 172 7 for for ADP brj-24033 172 8 3 3 NUM brj-24033 172 9 times time NOUN brj-24033 172 10 . . PUNCT brj-24033 173 1 finally finally ADV brj-24033 173 2 , , PUNCT brj-24033 173 3 the the DET brj-24033 173 4 training training NOUN brj-24033 173 5 dataset dataset NOUN brj-24033 173 6 consists consist VERB brj-24033 173 7 of of ADP brj-24033 173 8 25 25 NUM brj-24033 173 9 known know VERB brj-24033 173 10 wood wood NOUN brj-24033 173 11 species specie NOUN brj-24033 173 12 . . PUNCT brj-24033 174 1 the the DET brj-24033 174 2 unknown unknown ADJ brj-24033 174 3 wood wood NOUN brj-24033 174 4 dataset dataset NOUN brj-24033 174 5 consists consist VERB brj-24033 174 6 of of ADP brj-24033 174 7 5 5 NUM brj-24033 174 8 wood wood NOUN brj-24033 174 9 species specie NOUN brj-24033 174 10 . . PUNCT brj-24033 175 1 this this DET brj-24033 175 2 wood wood NOUN brj-24033 175 3 dataset dataset NOUN brj-24033 175 4 partition partition NOUN brj-24033 175 5 is be AUX brj-24033 175 6 illustrated illustrate VERB brj-24033 175 7 in in ADP brj-24033 175 8 table table NOUN brj-24033 175 9 4 4 NUM brj-24033 175 10 . . PUNCT brj-24033 175 11 table table NOUN brj-24033 175 12 2 2 NUM brj-24033 175 13 . . PUNCT brj-24033 176 1 the the DET brj-24033 176 2 known know VERB brj-24033 176 3 and and CCONJ brj-24033 176 4 unknown unknown ADJ brj-24033 176 5 wood wood NOUN brj-24033 176 6 species species NOUN brj-24033 176 7 number number NOUN brj-24033 176 8 partition partition NOUN brj-24033 176 9 in in ADP brj-24033 176 10 group group NOUN brj-24033 176 11 1 1 NUM brj-24033 176 12 initial initial ADJ brj-24033 176 13 training training NOUN brj-24033 176 14 incremental incremental ADJ brj-24033 176 15 learning learning NOUN brj-24033 176 16 training training NOUN brj-24033 176 17 set set NOUN brj-24033 176 18 ( ( PUNCT brj-24033 176 19 known know VERB brj-24033 176 20 species specie NOUN brj-24033 176 21 ) ) PUNCT brj-24033 176 22 5 5 NUM brj-24033 176 23 10 10 NUM brj-24033 176 24 15 15 NUM brj-24033 176 25 20 20 NUM brj-24033 176 26 open open ADJ brj-24033 176 27 set set NOUN brj-24033 176 28 ( ( PUNCT brj-24033 176 29 unknown unknown ADJ brj-24033 176 30 species specie NOUN brj-24033 176 31 ) ) PUNCT brj-24033 176 32 5 5 NUM brj-24033 176 33 5 5 NUM brj-24033 176 34 5 5 NUM brj-24033 176 35 5 5 NUM brj-24033 176 36 table table NOUN brj-24033 176 37 3 3 NUM brj-24033 176 38 . . PUNCT brj-24033 177 1 the the DET brj-24033 177 2 known know VERB brj-24033 177 3 and and CCONJ brj-24033 177 4 unknown unknown ADJ brj-24033 177 5 wood wood NOUN brj-24033 177 6 species species NOUN brj-24033 177 7 number number NOUN brj-24033 177 8 partition partition NOUN brj-24033 177 9 in in ADP brj-24033 177 10 group group NOUN brj-24033 177 11 2 2 NUM brj-24033 177 12 initial initial ADJ brj-24033 177 13 training training NOUN brj-24033 177 14 incremental incremental ADJ brj-24033 177 15 learning learning NOUN brj-24033 177 16 training training NOUN brj-24033 177 17 set set NOUN brj-24033 177 18 ( ( PUNCT brj-24033 177 19 known know VERB brj-24033 177 20 species specie NOUN brj-24033 177 21 ) ) PUNCT brj-24033 177 22 5 5 NUM brj-24033 177 23 10 10 NUM brj-24033 177 24 15 15 NUM brj-24033 177 25 20 20 NUM brj-24033 177 26 open open ADJ brj-24033 177 27 set set NOUN brj-24033 177 28 ( ( PUNCT brj-24033 177 29 unknown unknown ADJ brj-24033 177 30 species specie NOUN brj-24033 177 31 ) ) PUNCT brj-24033 177 32 10 10 NUM brj-24033 177 33 10 10 NUM brj-24033 177 34 10 10 NUM brj-24033 177 35 10 10 NUM brj-24033 177 36 table table NOUN brj-24033 177 37 4 4 NUM brj-24033 177 38 . . PUNCT brj-24033 178 1 the the DET brj-24033 178 2 known know VERB brj-24033 178 3 and and CCONJ brj-24033 178 4 unknown unknown ADJ brj-24033 178 5 wood wood NOUN brj-24033 178 6 species species NOUN brj-24033 178 7 number number NOUN brj-24033 178 8 partition partition NOUN brj-24033 178 9 in in ADP brj-24033 178 10 group group NOUN brj-24033 178 11 3 3 NUM brj-24033 178 12 initial initial ADJ brj-24033 178 13 training training NOUN brj-24033 178 14 incremental incremental ADJ brj-24033 178 15 learning learning NOUN brj-24033 178 16 training training NOUN brj-24033 178 17 set set NOUN brj-24033 178 18 ( ( PUNCT brj-24033 178 19 known know VERB brj-24033 178 20 species specie NOUN brj-24033 178 21 ) ) PUNCT brj-24033 178 22 10 10 NUM brj-24033 178 23 15 15 NUM brj-24033 178 24 20 20 NUM brj-24033 178 25 25 25 NUM brj-24033 178 26 open open ADJ brj-24033 178 27 set set NOUN brj-24033 178 28 ( ( PUNCT brj-24033 178 29 unknown unknown ADJ brj-24033 178 30 species specie NOUN brj-24033 178 31 ) ) PUNCT brj-24033 178 32 5 5 NUM brj-24033 178 33 5 5 NUM brj-24033 178 34 5 5 NUM brj-24033 178 35 5 5 NUM brj-24033 178 36 wood wood NOUN brj-24033 178 37 species species NOUN brj-24033 178 38 classification classification NOUN brj-24033 178 39 comparisons comparison NOUN brj-24033 178 40 the the DET brj-24033 178 41 proposed propose VERB brj-24033 178 42 improved improve VERB brj-24033 178 43 nno nno PROPN brj-24033 178 44 classifier classifier NOUN brj-24033 178 45 was be AUX brj-24033 178 46 compared compare VERB brj-24033 178 47 in in ADP brj-24033 178 48 open open ADJ brj-24033 178 49 set set VERB brj-24033 178 50 with with ADP brj-24033 178 51 other other ADJ brj-24033 178 52 5 5 NUM brj-24033 178 53 representative representative ADJ brj-24033 178 54 osr osr PROPN brj-24033 178 55 classifiers classifier NOUN brj-24033 178 56 . . PUNCT brj-24033 179 1 the the DET brj-24033 179 2 original original ADJ brj-24033 179 3 nno nno NOUN brj-24033 179 4 classifier classifier NOUN brj-24033 179 5 ( ( PUNCT brj-24033 179 6 bendale bendale NOUN brj-24033 179 7 and and CCONJ brj-24033 179 8 boult boult NOUN brj-24033 179 9 2015 2015 NUM brj-24033 179 10 ) ) PUNCT brj-24033 179 11 was be AUX brj-24033 179 12 used use VERB brj-24033 179 13 for for ADP brj-24033 179 14 a a DET brj-24033 179 15 baseline baseline ADJ brj-24033 179 16 comparison comparison NOUN brj-24033 179 17 . . PUNCT brj-24033 180 1 another another DET brj-24033 180 2 one one NUM brj-24033 180 3 conventional conventional ADJ brj-24033 180 4 osr osr NOUN brj-24033 180 5 classifier classifier NOUN brj-24033 180 6 was be AUX brj-24033 180 7 used use VERB brj-24033 180 8 , , PUNCT brj-24033 180 9 which which PRON brj-24033 180 10 consisted consist VERB brj-24033 180 11 of of ADP brj-24033 180 12 two two NUM brj-24033 180 13 parts part NOUN brj-24033 180 14 . . PUNCT brj-24033 181 1 peer peer NOUN brj-24033 181 2 - - PUNCT brj-24033 181 3 reviewed review VERB brj-24033 181 4 article article NOUN brj-24033 181 5 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 181 6 zhang zhang PROPN brj-24033 181 7 & & CCONJ brj-24033 181 8 zhao zhao PROPN brj-24033 181 9 ( ( PUNCT brj-24033 181 10 2025 2025 NUM brj-24033 181 11 ) ) PUNCT brj-24033 181 12 . . PUNCT brj-24033 182 1 β€œ " PUNCT brj-24033 182 2 wood wood NOUN brj-24033 182 3 image image NOUN brj-24033 182 4 classification classification NOUN brj-24033 182 5 , , PUNCT brj-24033 182 6 ” " PUNCT brj-24033 182 7 bioresources bioresource NOUN brj-24033 182 8 20(1 20(1 NUM brj-24033 182 9 ) ) PUNCT brj-24033 182 10 , , PUNCT brj-24033 182 11 944 944 NUM brj-24033 182 12 - - SYM brj-24033 182 13 955 955 NUM brj-24033 182 14 . . PUNCT brj-24033 183 1 952 952 NUM brj-24033 183 2 table table NOUN brj-24033 183 3 5 5 NUM brj-24033 183 4 . . PUNCT brj-24033 184 1 the the DET brj-24033 184 2 osr osr PROPN brj-24033 184 3 classification classification NOUN brj-24033 184 4 performance performance NOUN brj-24033 184 5 comparisons comparison NOUN brj-24033 184 6 in in ADP brj-24033 184 7 group group NOUN brj-24033 184 8 1 1 NUM brj-24033 184 9 ora ora PROPN brj-24033 184 10 kappa kappa PROPN brj-24033 184 11 coefficient coefficient NOUN brj-24033 184 12 f f X brj-24033 184 13 - - PUNCT brj-24033 184 14 score score NOUN brj-24033 184 15 known know VERB brj-24033 184 16 class class NOUN brj-24033 184 17 number number NOUN brj-24033 184 18 5 5 NUM brj-24033 184 19 10 10 NUM brj-24033 184 20 15 15 NUM brj-24033 184 21 20 20 NUM brj-24033 184 22 5 5 NUM brj-24033 184 23 10 10 NUM brj-24033 184 24 15 15 NUM brj-24033 184 25 20 20 NUM brj-24033 184 26 5 5 NUM brj-24033 184 27 10 10 NUM brj-24033 184 28 15 15 NUM brj-24033 184 29 20 20 NUM brj-24033 184 30 original original ADJ brj-24033 184 31 nno nno NOUN brj-24033 184 32 67.00 67.00 NUM brj-24033 184 33 % % NOUN brj-24033 184 34 62.11 62.11 NUM brj-24033 184 35 % % NOUN brj-24033 184 36 58.53 58.53 NUM brj-24033 184 37 % % NOUN brj-24033 184 38 42.62 42.62 NUM brj-24033 184 39 % % NOUN brj-24033 184 40 66.64 66.64 NUM brj-24033 184 41 % % NOUN brj-24033 184 42 61.57 61.57 NUM brj-24033 184 43 % % NOUN brj-24033 184 44 57.98 57.98 NUM brj-24033 184 45 % % NOUN brj-24033 184 46 41.43 41.43 NUM brj-24033 184 47 % % NOUN brj-24033 184 48 5.77 5.77 NUM brj-24033 184 49 % % NOUN brj-24033 184 50 6.15 6.15 NUM brj-24033 184 51 % % NOUN brj-24033 184 52 4.08 4.08 NUM brj-24033 184 53 % % NOUN brj-24033 184 54 6.25 6.25 NUM brj-24033 184 55 % % NOUN brj-24033 184 56 nno nno NOUN brj-24033 185 1 + + CCONJ brj-24033 185 2 k k X brj-24033 185 3 - - PUNCT brj-24033 185 4 means mean VERB brj-24033 185 5 97.43 97.43 NUM brj-24033 185 6 % % NOUN brj-24033 185 7 91.25 91.25 NUM brj-24033 185 8 % % NOUN brj-24033 185 9 38.73 38.73 NUM brj-24033 185 10 % % NOUN brj-24033 185 11 29.03 29.03 NUM brj-24033 185 12 % % NOUN brj-24033 185 13 97.36 97.36 NUM brj-24033 185 14 % % NOUN brj-24033 185 15 91.09 91.09 NUM brj-24033 185 16 % % NOUN brj-24033 185 17 37.92 37.92 NUM brj-24033 185 18 % % NOUN brj-24033 185 19 27.96 27.96 NUM brj-24033 185 20 % % NOUN brj-24033 185 21 81.08 81.08 NUM brj-24033 185 22 % % NOUN brj-24033 185 23 61.76 61.76 NUM brj-24033 185 24 % % NOUN brj-24033 185 25 19.25 19.25 NUM brj-24033 185 26 % % NOUN brj-24033 185 27 21.43 21.43 NUM brj-24033 185 28 % % NOUN brj-24033 185 29 openmax openmax ADJ brj-24033 185 30 network network NOUN brj-24033 185 31 29.00 29.00 NUM brj-24033 185 32 % % NOUN brj-24033 185 33 39.33 39.33 NUM brj-24033 185 34 % % NOUN brj-24033 185 35 49.50 49.50 NUM brj-24033 185 36 % % NOUN brj-24033 185 37 50.00 50.00 NUM brj-24033 185 38 % % NOUN brj-24033 185 39 24.23 24.23 NUM brj-24033 185 40 % % NOUN brj-24033 185 41 37.04 37.04 NUM brj-24033 185 42 % % NOUN brj-24033 185 43 47.83 47.83 NUM brj-24033 185 44 % % NOUN brj-24033 185 45 48.63 48.63 NUM brj-24033 185 46 % % NOUN brj-24033 185 47 61.05 61.05 NUM brj-24033 185 48 % % NOUN brj-24033 185 49 63.44 63.44 NUM brj-24033 185 50 % % NOUN brj-24033 185 51 73.06 73.06 NUM brj-24033 185 52 % % NOUN brj-24033 185 53 74.63 74.63 NUM brj-24033 185 54 % % NOUN brj-24033 185 55 osnn osnn NOUN brj-24033 185 56 35.28 35.28 NUM brj-24033 185 57 % % NOUN brj-24033 185 58 73.60 73.60 NUM brj-24033 185 59 % % NOUN brj-24033 185 60 77.80 77.80 NUM brj-24033 185 61 % % NOUN brj-24033 185 62 65.42 65.42 NUM brj-24033 185 63 % % NOUN brj-24033 185 64 35.05 35.05 NUM brj-24033 185 65 % % NOUN brj-24033 185 66 73.27 73.27 NUM brj-24033 185 67 % % NOUN brj-24033 185 68 77.63 77.63 NUM brj-24033 185 69 % % NOUN brj-24033 185 70 65.02 65.02 NUM brj-24033 185 71 % % NOUN brj-24033 185 72 1.12 1.12 NUM brj-24033 185 73 % % NOUN brj-24033 185 74 1.31 1.31 NUM brj-24033 185 75 % % NOUN brj-24033 185 76 7.09 7.09 NUM brj-24033 185 77 % % NOUN brj-24033 185 78 0.93 0.93 NUM brj-24033 185 79 % % NOUN brj-24033 185 80 weight weight NOUN brj-24033 185 81 - - PUNCT brj-24033 185 82 svdd+libsvm svdd+libsvm PROPN brj-24033 185 83 94.12 94.12 NUM brj-24033 185 84 % % NOUN brj-24033 185 85 90.24 90.24 NUM brj-24033 185 86 % % NOUN brj-24033 185 87 84.76 84.76 NUM brj-24033 185 88 % % NOUN brj-24033 185 89 69.50 69.50 NUM brj-24033 185 90 % % NOUN brj-24033 185 91 92.93 92.93 NUM brj-24033 185 92 % % NOUN brj-24033 185 93 88.62 88.62 NUM brj-24033 185 94 % % NOUN brj-24033 185 95 82.57 82.57 NUM brj-24033 185 96 % % NOUN brj-24033 185 97 66.46 66.46 NUM brj-24033 185 98 % % NOUN brj-24033 185 99 50.00 50.00 NUM brj-24033 185 100 % % NOUN brj-24033 185 101 57.97 57.97 NUM brj-24033 185 102 % % NOUN brj-24033 185 103 44.19 44.19 NUM brj-24033 185 104 % % NOUN brj-24033 185 105 35.80 35.80 NUM brj-24033 185 106 % % NOUN brj-24033 185 107 ours ours PROPN brj-24033 185 108 ( ( PUNCT brj-24033 185 109 nno+dpc nno+dpc NOUN brj-24033 185 110 ) ) PUNCT brj-24033 185 111 98.53 98.53 NUM brj-24033 185 112 % % NOUN brj-24033 185 113 97.98 97.98 NUM brj-24033 185 114 % % NOUN brj-24033 185 115 96.83 96.83 NUM brj-24033 185 116 % % NOUN brj-24033 185 117 93.84 93.84 NUM brj-24033 185 118 % % NOUN brj-24033 185 119 98.49 98.49 NUM brj-24033 185 120 % % NOUN brj-24033 185 121 97.94 97.94 NUM brj-24033 185 122 % % NOUN brj-24033 185 123 96.77 96.77 NUM brj-24033 185 124 % % NOUN brj-24033 185 125 93.73 93.73 NUM brj-24033 185 126 % % NOUN brj-24033 185 127 90.91 90.91 NUM brj-24033 185 128 % % NOUN brj-24033 185 129 93.18 93.18 NUM brj-24033 185 130 % % NOUN brj-24033 185 131 91.67 91.67 NUM brj-24033 185 132 % % NOUN brj-24033 185 133 86.96 86.96 NUM brj-24033 185 134 % % NOUN brj-24033 185 135 table table NOUN brj-24033 185 136 6 6 NUM brj-24033 185 137 . . PUNCT brj-24033 186 1 the the DET brj-24033 186 2 osr osr PROPN brj-24033 186 3 classification classification NOUN brj-24033 186 4 performance performance NOUN brj-24033 186 5 comparisons comparison NOUN brj-24033 186 6 in in ADP brj-24033 186 7 group group NOUN brj-24033 186 8 2 2 NUM brj-24033 186 9 ora ora PROPN brj-24033 186 10 kappa kappa NOUN brj-24033 186 11 coefficient coefficient NOUN brj-24033 186 12 f f X brj-24033 186 13 - - PUNCT brj-24033 186 14 score score NOUN brj-24033 186 15 known know VERB brj-24033 186 16 class class NOUN brj-24033 186 17 number number NOUN brj-24033 186 18 5 5 NUM brj-24033 186 19 10 10 NUM brj-24033 186 20 15 15 NUM brj-24033 186 21 20 20 NUM brj-24033 186 22 5 5 NUM brj-24033 186 23 10 10 NUM brj-24033 186 24 15 15 NUM brj-24033 186 25 20 20 NUM brj-24033 186 26 5 5 NUM brj-24033 186 27 10 10 NUM brj-24033 186 28 15 15 NUM brj-24033 186 29 20 20 NUM brj-24033 186 30 original original ADJ brj-24033 186 31 nno nno NOUN brj-24033 186 32 53.20 53.20 NUM brj-24033 186 33 % % NOUN brj-24033 186 34 40.21 40.21 NUM brj-24033 186 35 % % NOUN brj-24033 186 36 38.81 38.81 NUM brj-24033 186 37 % % NOUN brj-24033 186 38 26.62 26.62 NUM brj-24033 186 39 % % NOUN brj-24033 186 40 52.92 52.92 NUM brj-24033 186 41 % % NOUN brj-24033 186 42 39.72 39.72 NUM brj-24033 186 43 % % NOUN brj-24033 186 44 38.34 38.34 NUM brj-24033 186 45 % % NOUN brj-24033 186 46 25.77 25.77 NUM brj-24033 186 47 % % NOUN brj-24033 186 48 2.29 2.29 NUM brj-24033 186 49 % % NOUN brj-24033 186 50 2.29 2.29 NUM brj-24033 186 51 % % NOUN brj-24033 186 52 1.63 1.63 NUM brj-24033 186 53 % % NOUN brj-24033 186 54 3.00 3.00 NUM brj-24033 186 55 % % NOUN brj-24033 186 56 nno nno NOUN brj-24033 186 57 + + CCONJ brj-24033 186 58 kmeans kmean NOUN brj-24033 186 59 97.13 97.13 NUM brj-24033 186 60 % % NOUN brj-24033 186 61 94.52 94.52 NUM brj-24033 186 62 % % NOUN brj-24033 186 63 61.42 61.42 NUM brj-24033 186 64 % % NOUN brj-24033 186 65 48.90 48.90 NUM brj-24033 186 66 % % NOUN brj-24033 186 67 97.09 97.09 NUM brj-24033 186 68 % % NOUN brj-24033 186 69 94.47 94.47 NUM brj-24033 186 70 % % NOUN brj-24033 186 71 61.18 61.18 NUM brj-24033 186 72 % % NOUN brj-24033 186 73 48.48 48.48 NUM brj-24033 186 74 % % NOUN brj-24033 186 75 66.67 66.67 NUM brj-24033 186 76 % % NOUN brj-24033 186 77 57.14 57.14 NUM brj-24033 186 78 % % NOUN brj-24033 186 79 16.79 16.79 NUM brj-24033 186 80 % % NOUN brj-24033 186 81 17.93 17.93 NUM brj-24033 186 82 % % NOUN brj-24033 186 83 openmax openmax ADJ brj-24033 186 84 network network NOUN brj-24033 186 85 19.33 19.33 NUM brj-24033 186 86 % % NOUN brj-24033 186 87 29.50 29.50 NUM brj-24033 186 88 % % NOUN brj-24033 186 89 39.60 39.60 NUM brj-24033 186 90 % % NOUN brj-24033 186 91 41.67 41.67 NUM brj-24033 186 92 % % NOUN brj-24033 186 93 15.31 15.31 NUM brj-24033 186 94 % % NOUN brj-24033 186 95 27.43 27.43 NUM brj-24033 186 96 % % NOUN brj-24033 186 97 37.84 37.84 NUM brj-24033 186 98 % % NOUN brj-24033 186 99 40.23 40.23 NUM brj-24033 186 100 % % NOUN brj-24033 186 101 58.59 58.59 NUM brj-24033 186 102 % % NOUN brj-24033 186 103 58.13 58.13 NUM brj-24033 186 104 % % NOUN brj-24033 186 105 72.79 72.79 NUM brj-24033 186 106 % % NOUN brj-24033 186 107 74.63 74.63 NUM brj-24033 186 108 % % NOUN brj-24033 186 109 osnn osnn NOUN brj-24033 186 110 57.91 57.91 NUM brj-24033 186 111 % % NOUN brj-24033 186 112 77.95 77.95 NUM brj-24033 186 113 % % NOUN brj-24033 186 114 72.62 72.62 NUM brj-24033 186 115 % % NOUN brj-24033 186 116 56.01 56.01 NUM brj-24033 186 117 % % NOUN brj-24033 186 118 57.63 57.63 NUM brj-24033 186 119 % % NOUN brj-24033 186 120 77.44 77.44 NUM brj-24033 186 121 % % NOUN brj-24033 186 122 72.36 72.36 NUM brj-24033 186 123 % % NOUN brj-24033 186 124 55.17 55.17 NUM brj-24033 186 125 % % NOUN brj-24033 186 126 3.10 3.10 NUM brj-24033 186 127 % % NOUN brj-24033 186 128 2.74 2.74 NUM brj-24033 186 129 % % NOUN brj-24033 186 130 9.71 9.71 NUM brj-24033 186 131 % % NOUN brj-24033 186 132 1.23 1.23 NUM brj-24033 186 133 % % NOUN brj-24033 187 1 weightsvdd+libsvm weightsvdd+libsvm PROPN brj-24033 187 2 79.69 79.69 NUM brj-24033 187 3 % % NOUN brj-24033 187 4 77.51 77.51 NUM brj-24033 187 5 % % NOUN brj-24033 187 6 71.33 71.33 NUM brj-24033 187 7 % % NOUN brj-24033 187 8 53.98 53.98 NUM brj-24033 187 9 % % NOUN brj-24033 187 10 77.03 77.03 NUM brj-24033 187 11 % % NOUN brj-24033 187 12 75.87 75.87 NUM brj-24033 187 13 % % NOUN brj-24033 187 14 69.48 69.48 NUM brj-24033 187 15 % % NOUN brj-24033 187 16 51.95 51.95 NUM brj-24033 187 17 % % NOUN brj-24033 187 18 14.52 14.52 NUM brj-24033 187 19 % % NOUN brj-24033 187 20 24.54 24.54 NUM brj-24033 187 21 % % NOUN brj-24033 187 22 19.00 19.00 NUM brj-24033 187 23 % % NOUN brj-24033 187 24 17.58 17.58 NUM brj-24033 187 25 % % NOUN brj-24033 187 26 ours ours ADJ brj-24033 187 27 ( ( PUNCT brj-24033 187 28 nno+dpc nno+dpc NOUN brj-24033 187 29 ) ) PUNCT brj-24033 187 30 98.66 98.66 NUM brj-24033 187 31 % % NOUN brj-24033 187 32 98.17 98.17 NUM brj-24033 187 33 % % NOUN brj-24033 187 34 96.81 96.81 NUM brj-24033 187 35 % % NOUN brj-24033 187 36 94.92 94.92 NUM brj-24033 187 37 % % NOUN brj-24033 187 38 98.64 98.64 NUM brj-24033 187 39 % % NOUN brj-24033 187 40 98.15 98.15 NUM brj-24033 187 41 % % NOUN brj-24033 187 42 96.79 96.79 NUM brj-24033 187 43 % % NOUN brj-24033 187 44 94.87 94.87 NUM brj-24033 187 45 % % NOUN brj-24033 187 46 81.08 81.08 NUM brj-24033 187 47 % % NOUN brj-24033 187 48 88.10 88.10 NUM brj-24033 187 49 % % NOUN brj-24033 187 50 83.93 83.93 NUM brj-24033 187 51 % % NOUN brj-24033 187 52 80.26 80.26 NUM brj-24033 187 53 % % NOUN brj-24033 187 54 table table NOUN brj-24033 187 55 7 7 NUM brj-24033 187 56 . . PUNCT brj-24033 188 1 the the DET brj-24033 188 2 osr osr PROPN brj-24033 188 3 classification classification NOUN brj-24033 188 4 performance performance NOUN brj-24033 188 5 comparisons comparison NOUN brj-24033 188 6 in in ADP brj-24033 188 7 group group NOUN brj-24033 188 8 3 3 NUM brj-24033 188 9 ora ora PROPN brj-24033 188 10 kappa kappa PROPN brj-24033 188 11 coefficient coefficient PROPN brj-24033 188 12 f f X brj-24033 188 13 - - PUNCT brj-24033 188 14 score score NOUN brj-24033 188 15 known know VERB brj-24033 188 16 class class NOUN brj-24033 188 17 number number NOUN brj-24033 188 18 10 10 NUM brj-24033 188 19 15 15 NUM brj-24033 188 20 20 20 NUM brj-24033 188 21 25 25 NUM brj-24033 188 22 10 10 NUM brj-24033 188 23 15 15 NUM brj-24033 188 24 20 20 NUM brj-24033 188 25 25 25 NUM brj-24033 188 26 10 10 NUM brj-24033 188 27 15 15 NUM brj-24033 188 28 20 20 NUM brj-24033 188 29 25 25 NUM brj-24033 188 30 original original ADJ brj-24033 188 31 nno nno NOUN brj-24033 188 32 54.21 54.21 NUM brj-24033 188 33 % % NOUN brj-24033 188 34 48.89 48.89 NUM brj-24033 188 35 % % NOUN brj-24033 188 36 46.33 46.33 NUM brj-24033 188 37 % % NOUN brj-24033 188 38 14.95 14.95 NUM brj-24033 188 39 % % NOUN brj-24033 188 40 53.68 53.68 NUM brj-24033 188 41 % % NOUN brj-24033 188 42 48.32 48.32 NUM brj-24033 188 43 % % NOUN brj-24033 188 44 45.72 45.72 NUM brj-24033 188 45 % % NOUN brj-24033 188 46 13.63 13.63 NUM brj-24033 188 47 % % NOUN brj-24033 188 48 9.33 9.33 NUM brj-24033 188 49 % % NOUN brj-24033 188 50 2.42 2.42 NUM brj-24033 188 51 % % NOUN brj-24033 188 52 6.15 6.15 NUM brj-24033 188 53 % % NOUN brj-24033 188 54 3.69 3.69 NUM brj-24033 188 55 % % NOUN brj-24033 188 56 nno nno NOUN brj-24033 189 1 + + CCONJ brj-24033 189 2 k k X brj-24033 189 3 - - PUNCT brj-24033 189 4 means mean VERB brj-24033 189 5 95.62 95.62 NUM brj-24033 189 6 % % NOUN brj-24033 189 7 58.41 58.41 NUM brj-24033 189 8 % % NOUN brj-24033 189 9 54.25 54.25 NUM brj-24033 189 10 % % NOUN brj-24033 189 11 11.68 11.68 NUM brj-24033 189 12 % % NOUN brj-24033 189 13 95.54 95.54 NUM brj-24033 189 14 % % NOUN brj-24033 189 15 57.94 57.94 NUM brj-24033 189 16 % % NOUN brj-24033 189 17 53.50 53.50 NUM brj-24033 189 18 % % NOUN brj-24033 189 19 10.08 10.08 NUM brj-24033 189 20 % % NOUN brj-24033 189 21 84.34 84.34 NUM brj-24033 189 22 % % NOUN brj-24033 189 23 26.82 26.82 NUM brj-24033 189 24 % % NOUN brj-24033 189 25 30.36 30.36 NUM brj-24033 189 26 % % NOUN brj-24033 189 27 12.87 12.87 NUM brj-24033 189 28 % % NOUN brj-24033 189 29 openmax openmax ADJ brj-24033 189 30 network network NOUN brj-24033 189 31 39.33 39.33 NUM brj-24033 189 32 % % NOUN brj-24033 189 33 49.50 49.50 NUM brj-24033 189 34 % % NOUN brj-24033 189 35 50.00 50.00 NUM brj-24033 189 36 % % NOUN brj-24033 189 37 66.67 66.67 NUM brj-24033 189 38 % % NOUN brj-24033 189 39 37.04 37.04 NUM brj-24033 189 40 % % NOUN brj-24033 189 41 47.83 47.83 NUM brj-24033 189 42 % % NOUN brj-24033 189 43 48.63 48.63 NUM brj-24033 189 44 % % NOUN brj-24033 189 45 65.75 65.75 NUM brj-24033 189 46 % % NOUN brj-24033 189 47 63.44 63.44 NUM brj-24033 189 48 % % NOUN brj-24033 189 49 73.06 73.06 NUM brj-24033 189 50 % % NOUN brj-24033 189 51 74.63 74.63 NUM brj-24033 189 52 % % NOUN brj-24033 189 53 86.77 86.77 NUM brj-24033 189 54 % % NOUN brj-24033 189 55 osnn osnn NOUN brj-24033 189 56 70.71 70.71 NUM brj-24033 189 57 % % NOUN brj-24033 189 58 37.14 37.14 NUM brj-24033 189 59 % % NOUN brj-24033 189 60 52.79 52.79 NUM brj-24033 189 61 % % NOUN brj-24033 189 62 52.17 52.17 NUM brj-24033 189 63 % % NOUN brj-24033 189 64 70.51 70.51 NUM brj-24033 189 65 % % NOUN brj-24033 189 66 36.63 36.63 NUM brj-24033 189 67 % % NOUN brj-24033 189 68 51.88 51.88 NUM brj-24033 189 69 % % NOUN brj-24033 189 70 51.46 51.46 NUM brj-24033 189 71 % % NOUN brj-24033 189 72 4.40 4.40 NUM brj-24033 189 73 % % NOUN brj-24033 189 74 5.71 5.71 NUM brj-24033 189 75 % % NOUN brj-24033 189 76 1.23 1.23 NUM brj-24033 189 77 % % NOUN brj-24033 189 78 1.12 1.12 NUM brj-24033 189 79 % % NOUN brj-24033 189 80 weight weight NOUN brj-24033 189 81 - - PUNCT brj-24033 189 82 svdd+libsvm svdd+libsvm PROPN brj-24033 189 83 93.38 93.38 NUM brj-24033 189 84 % % NOUN brj-24033 189 85 90.00 90.00 NUM brj-24033 189 86 % % NOUN brj-24033 189 87 71.52 71.52 NUM brj-24033 189 88 % % NOUN brj-24033 189 89 39.65 39.65 NUM brj-24033 189 90 % % NOUN brj-24033 189 91 92.05 92.05 NUM brj-24033 189 92 % % NOUN brj-24033 189 93 88.26 88.26 NUM brj-24033 189 94 % % NOUN brj-24033 189 95 68.24 68.24 NUM brj-24033 189 96 % % NOUN brj-24033 189 97 36.81 36.81 NUM brj-24033 189 98 % % NOUN brj-24033 189 99 40.00 40.00 NUM brj-24033 189 100 % % NOUN brj-24033 189 101 47.27 47.27 NUM brj-24033 189 102 % % NOUN brj-24033 189 103 32.84 32.84 NUM brj-24033 189 104 % % NOUN brj-24033 189 105 24.73 24.73 NUM brj-24033 189 106 % % NOUN brj-24033 189 107 ours ours ADJ brj-24033 189 108 ( ( PUNCT brj-24033 189 109 nno+dpc nno+dpc NOUN brj-24033 189 110 ) ) PUNCT brj-24033 189 111 98.32 98.32 NUM brj-24033 189 112 % % NOUN brj-24033 189 113 98.14 98.14 NUM brj-24033 189 114 % % NOUN brj-24033 189 115 96.47 96.47 NUM brj-24033 189 116 % % NOUN brj-24033 189 117 91.80 91.80 NUM brj-24033 189 118 % % NOUN brj-24033 189 119 98.28 98.28 NUM brj-24033 189 120 % % NOUN brj-24033 189 121 98.08 98.08 NUM brj-24033 189 122 % % NOUN brj-24033 189 123 96.37 96.37 NUM brj-24033 189 124 % % NOUN brj-24033 189 125 91.63 91.63 NUM brj-24033 189 126 % % NOUN brj-24033 189 127 94.38 94.38 NUM brj-24033 189 128 % % NOUN brj-24033 189 129 95.65 95.65 NUM brj-24033 189 130 % % NOUN brj-24033 189 131 92.86 92.86 NUM brj-24033 189 132 % % NOUN brj-24033 189 133 85.15 85.15 NUM brj-24033 189 134 % % NOUN brj-24033 189 135 peer peer NOUN brj-24033 189 136 - - PUNCT brj-24033 189 137 reviewed review VERB brj-24033 189 138 article article NOUN brj-24033 189 139 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 189 140 zhang zhang PROPN brj-24033 189 141 & & CCONJ brj-24033 189 142 zhao zhao PROPN brj-24033 189 143 ( ( PUNCT brj-24033 189 144 2025 2025 NUM brj-24033 189 145 ) ) PUNCT brj-24033 189 146 . . PUNCT brj-24033 190 1 β€œ " PUNCT brj-24033 190 2 wood wood NOUN brj-24033 190 3 image image NOUN brj-24033 190 4 classification classification NOUN brj-24033 190 5 , , PUNCT brj-24033 190 6 ” " PUNCT brj-24033 190 7 bioresources bioresource NOUN brj-24033 190 8 20(1 20(1 NUM brj-24033 190 9 ) ) PUNCT brj-24033 190 10 , , PUNCT brj-24033 190 11 944 944 NUM brj-24033 190 12 - - SYM brj-24033 190 13 955 955 NUM brj-24033 190 14 . . PUNCT brj-24033 191 1 953 953 NUM brj-24033 191 2 the the DET brj-24033 191 3 first first ADJ brj-24033 191 4 part part NOUN brj-24033 191 5 was be AUX brj-24033 191 6 a a DET brj-24033 191 7 one one NUM brj-24033 191 8 - - PUNCT brj-24033 191 9 class class NOUN brj-24033 191 10 classifier classifier NOUN brj-24033 191 11 weight weight NOUN brj-24033 191 12 - - PUNCT brj-24033 191 13 svdd svdd NOUN brj-24033 191 14 ( ( PUNCT brj-24033 191 15 tax tax NOUN brj-24033 191 16 and and CCONJ brj-24033 191 17 duin duin PROPN brj-24033 191 18 2004 2004 NUM brj-24033 191 19 ) ) PUNCT brj-24033 191 20 , , PUNCT brj-24033 191 21 which which PRON brj-24033 191 22 classifies classify VERB brj-24033 191 23 wood wood NOUN brj-24033 191 24 samples sample NOUN brj-24033 191 25 into into ADP brj-24033 191 26 known known ADJ brj-24033 191 27 and and CCONJ brj-24033 191 28 unknown unknown ADJ brj-24033 191 29 categories category NOUN brj-24033 191 30 . . PUNCT brj-24033 192 1 the the DET brj-24033 192 2 second second ADJ brj-24033 192 3 part part NOUN brj-24033 192 4 was be AUX brj-24033 192 5 a a DET brj-24033 192 6 multi multi ADJ brj-24033 192 7 - - ADJ brj-24033 192 8 class class ADJ brj-24033 192 9 classifier classifier NOUN brj-24033 192 10 libsvm libsvm NOUN brj-24033 192 11 , , PUNCT brj-24033 192 12 which which PRON brj-24033 192 13 classifies classify VERB brj-24033 192 14 the the DET brj-24033 192 15 known know VERB brj-24033 192 16 category category NOUN brj-24033 192 17 given give VERB brj-24033 192 18 by by ADP brj-24033 192 19 the the DET brj-24033 192 20 first first ADJ brj-24033 192 21 part part NOUN brj-24033 192 22 into into ADP brj-24033 192 23 specific specific ADJ brj-24033 192 24 wood wood NOUN brj-24033 192 25 species specie NOUN brj-24033 192 26 . . PUNCT brj-24033 193 1 the the DET brj-24033 193 2 optimal optimal ADJ brj-24033 193 3 parameters parameter NOUN brj-24033 193 4 are be AUX brj-24033 193 5 𝑐 𝑐 NOUN brj-24033 193 6 = = SYM brj-24033 193 7 0.04 0.04 NUM brj-24033 193 8 , , PUNCT brj-24033 193 9 𝜎 𝜎 NOUN brj-24033 193 10 = = SYM brj-24033 193 11 0.50 0.50 NUM brj-24033 193 12 for for ADP brj-24033 193 13 the the DET brj-24033 193 14 weightsvdd weightsvdd PROPN brj-24033 193 15 and and CCONJ brj-24033 193 16 𝑐 𝑐 NOUN brj-24033 193 17 = = PROPN brj-24033 193 18 1.00 1.00 NUM brj-24033 193 19 , , PUNCT brj-24033 193 20 𝜎 𝜎 NOUN brj-24033 193 21 = = NOUN brj-24033 193 22 1.05 1.05 NUM brj-24033 193 23 for for ADP brj-24033 193 24 the the DET brj-24033 193 25 libsvm libsvm NOUN brj-24033 193 26 . . PUNCT brj-24033 194 1 in in ADP brj-24033 194 2 the the DET brj-24033 194 3 third third ADJ brj-24033 194 4 osr osr PROPN brj-24033 194 5 classifier classifier NOUN brj-24033 194 6 , , PUNCT brj-24033 194 7 the the DET brj-24033 194 8 k k NOUN brj-24033 194 9 - - PUNCT brj-24033 194 10 means means NOUN brj-24033 194 11 cluster cluster NOUN brj-24033 194 12 analysis analysis NOUN brj-24033 194 13 is be AUX brj-24033 194 14 used use VERB brj-24033 194 15 instead instead ADV brj-24033 194 16 of of ADP brj-24033 194 17 dpc dpc PROPN brj-24033 194 18 automatic automatic ADJ brj-24033 194 19 clustering clustering NOUN brj-24033 194 20 , , PUNCT brj-24033 194 21 and and CCONJ brj-24033 194 22 the the DET brj-24033 194 23 cluster cluster NOUN brj-24033 194 24 number number NOUN brj-24033 194 25 𝐾 𝐾 NOUN brj-24033 194 26 = = NOUN brj-24033 194 27 2 2 NUM brj-24033 194 28 for for ADP brj-24033 194 29 each each DET brj-24033 194 30 known know VERB brj-24033 194 31 class class NOUN brj-24033 194 32 in in ADP brj-24033 194 33 a a DET brj-24033 194 34 nno nno NOUN brj-24033 194 35 classifier classifier NOUN brj-24033 194 36 . . PUNCT brj-24033 195 1 the the DET brj-24033 195 2 osnn osnn NOUN brj-24033 195 3 classifier classifier NOUN brj-24033 195 4 proposed propose VERB brj-24033 195 5 by by ADP brj-24033 195 6 junior junior PROPN brj-24033 195 7 et et PROPN brj-24033 195 8 al al PROPN brj-24033 195 9 . . PROPN brj-24033 195 10 ( ( PUNCT brj-24033 195 11 2017 2017 NUM brj-24033 195 12 ) ) PUNCT brj-24033 195 13 was be AUX brj-24033 195 14 performed perform VERB brj-24033 195 15 for for ADP brj-24033 195 16 comparison comparison NOUN brj-24033 195 17 , , PUNCT brj-24033 195 18 with with ADP brj-24033 195 19 the the DET brj-24033 195 20 threshold threshold NOUN brj-24033 195 21 𝑇𝑅 𝑇𝑅 PROPN brj-24033 195 22 = = PUNCT brj-24033 195 23 0.3 0.3 NUM brj-24033 195 24 . . PUNCT brj-24033 196 1 the the DET brj-24033 196 2 ml ml PROPN brj-24033 196 3 algorithm algorithm NOUN brj-24033 196 4 was be AUX brj-24033 196 5 used use VERB brj-24033 196 6 for for ADP brj-24033 196 7 spectral spectral ADJ brj-24033 196 8 dimension dimension NOUN brj-24033 196 9 reduction reduction NOUN brj-24033 196 10 . . PUNCT brj-24033 197 1 the the DET brj-24033 197 2 final final ADJ brj-24033 197 3 comparative comparative ADJ brj-24033 197 4 osr osr NOUN brj-24033 197 5 classifier classifier NOUN brj-24033 197 6 was be AUX brj-24033 197 7 the the DET brj-24033 197 8 openmax openmax ADJ brj-24033 197 9 network network NOUN brj-24033 197 10 ( ( PUNCT brj-24033 197 11 bendale bendale NOUN brj-24033 197 12 and and CCONJ brj-24033 197 13 boult boult NOUN brj-24033 197 14 2016 2016 NUM brj-24033 197 15 ) ) PUNCT brj-24033 197 16 , , PUNCT brj-24033 197 17 where where SCONJ brj-24033 197 18 a a DET brj-24033 197 19 resnet resnet NOUN brj-24033 197 20 50 50 NUM brj-24033 197 21 was be AUX brj-24033 197 22 used use VERB brj-24033 197 23 as as ADP brj-24033 197 24 the the DET brj-24033 197 25 backbone backbone NOUN brj-24033 197 26 network network NOUN brj-24033 197 27 . . PUNCT brj-24033 198 1 this this DET brj-24033 198 2 openmax openmax ADJ brj-24033 198 3 network network NOUN brj-24033 198 4 is be AUX brj-24033 198 5 mainly mainly ADV brj-24033 198 6 applied apply VERB brj-24033 198 7 in in ADP brj-24033 198 8 image image NOUN brj-24033 198 9 processing processing NOUN brj-24033 198 10 field field NOUN brj-24033 198 11 , , PUNCT brj-24033 198 12 obtaining obtain VERB brj-24033 198 13 the the DET brj-24033 198 14 visible visible ADJ brj-24033 198 15 images image NOUN brj-24033 198 16 of of ADP brj-24033 198 17 wood wood NOUN brj-24033 198 18 cross cross NOUN brj-24033 198 19 sections section NOUN brj-24033 198 20 . . PUNCT brj-24033 199 1 the the DET brj-24033 199 2 original original ADJ brj-24033 199 3 image image NOUN brj-24033 199 4 size size NOUN brj-24033 199 5 was be AUX brj-24033 199 6 1280 1280 NUM brj-24033 199 7 Γ— Γ— NOUN brj-24033 199 8 960 960 NUM brj-24033 199 9 with with ADP brj-24033 199 10 a a DET brj-24033 199 11 magnification magnification NOUN brj-24033 199 12 of of ADP brj-24033 199 13 50 50 NUM brj-24033 199 14 , , PUNCT brj-24033 199 15 and and CCONJ brj-24033 199 16 this this DET brj-24033 199 17 original original ADJ brj-24033 199 18 image image NOUN brj-24033 199 19 was be AUX brj-24033 199 20 cropped crop VERB brj-24033 199 21 into into ADP brj-24033 199 22 an an DET brj-24033 199 23 960 960 NUM brj-24033 199 24 Γ— Γ— NOUN brj-24033 199 25 960 960 NUM brj-24033 199 26 image image NOUN brj-24033 199 27 , , PUNCT brj-24033 199 28 which which PRON brj-24033 199 29 was be AUX brj-24033 199 30 then then ADV brj-24033 199 31 resized resize VERB brj-24033 199 32 into into ADP brj-24033 199 33 an an DET brj-24033 199 34 224 224 NUM brj-24033 199 35 Γ— Γ— NOUN brj-24033 199 36 224 224 NUM brj-24033 199 37 image image NOUN brj-24033 199 38 . . PUNCT brj-24033 200 1 the the DET brj-24033 200 2 specific specific ADJ brj-24033 200 3 class class NOUN brj-24033 200 4 probability probability NOUN brj-24033 200 5 threshold threshold NOUN brj-24033 200 6 was be AUX brj-24033 200 7 set set VERB brj-24033 200 8 as as ADP brj-24033 200 9 40 40 NUM brj-24033 200 10 % % NOUN brj-24033 200 11 so so SCONJ brj-24033 200 12 that that SCONJ brj-24033 200 13 one one NUM brj-24033 200 14 detected detect VERB brj-24033 200 15 sample sample NOUN brj-24033 200 16 was be AUX brj-24033 200 17 classified classify VERB brj-24033 200 18 as as ADP brj-24033 200 19 an an DET brj-24033 200 20 unknown unknown ADJ brj-24033 200 21 class class NOUN brj-24033 200 22 if if SCONJ brj-24033 200 23 the the DET brj-24033 200 24 largest large ADJ brj-24033 200 25 class class NOUN brj-24033 200 26 membership membership NOUN brj-24033 200 27 probability probability NOUN brj-24033 200 28 was be AUX brj-24033 200 29 less less ADJ brj-24033 200 30 than than ADP brj-24033 200 31 40 40 NUM brj-24033 200 32 % % NOUN brj-24033 200 33 . . PUNCT brj-24033 201 1 the the DET brj-24033 201 2 specific specific ADJ brj-24033 201 3 osr osr NOUN brj-24033 201 4 classification classification NOUN brj-24033 201 5 comparisons comparison NOUN brj-24033 201 6 are be AUX brj-24033 201 7 illustrated illustrate VERB brj-24033 201 8 in in ADP brj-24033 201 9 tables table NOUN brj-24033 201 10 5 5 NUM brj-24033 201 11 to to PART brj-24033 201 12 7 7 NUM brj-24033 201 13 . . PUNCT brj-24033 202 1 the the DET brj-24033 202 2 bold bold ADJ brj-24033 202 3 fonts font NOUN brj-24033 202 4 indicate indicate VERB brj-24033 202 5 the the DET brj-24033 202 6 best good ADJ brj-24033 202 7 classification classification NOUN brj-24033 202 8 accuracy accuracy NOUN brj-24033 202 9 in in ADP brj-24033 202 10 the the DET brj-24033 202 11 relevant relevant ADJ brj-24033 202 12 column column NOUN brj-24033 202 13 . . PUNCT brj-24033 203 1 conclusions conclusion NOUN brj-24033 203 2 1 1 NUM brj-24033 203 3 . . PUNCT brj-24033 204 1 as as ADP brj-24033 204 2 for for ADP brj-24033 204 3 the the DET brj-24033 204 4 spectral spectral ADJ brj-24033 204 5 dimension dimension NOUN brj-24033 204 6 reduction reduction NOUN brj-24033 204 7 , , PUNCT brj-24033 204 8 the the DET brj-24033 204 9 metric metric ADJ brj-24033 204 10 learning learning NOUN brj-24033 204 11 ( ( PUNCT brj-24033 204 12 ml ml NOUN brj-24033 204 13 ) ) PUNCT brj-24033 204 14 algorithm algorithm NOUN brj-24033 204 15 proposed propose VERB brj-24033 204 16 by by ADP brj-24033 204 17 mensink mensink PROPN brj-24033 204 18 et et PROPN brj-24033 204 19 al al PROPN brj-24033 204 20 . . PROPN brj-24033 205 1 ( ( PUNCT brj-24033 205 2 2013 2013 NUM brj-24033 205 3 ) ) PUNCT brj-24033 205 4 seems seem VERB brj-24033 205 5 to to PART brj-24033 205 6 be be AUX brj-24033 205 7 a a DET brj-24033 205 8 proper proper ADJ brj-24033 205 9 choice choice NOUN brj-24033 205 10 . . PUNCT brj-24033 206 1 this this DET brj-24033 206 2 algorithm algorithm NOUN brj-24033 206 3 can can AUX brj-24033 206 4 be be AUX brj-24033 206 5 efficiently efficiently ADV brj-24033 206 6 used use VERB brj-24033 206 7 in in ADP brj-24033 206 8 an an DET brj-24033 206 9 incremental incremental ADJ brj-24033 206 10 learning learning NOUN brj-24033 206 11 classifier classifier NOUN brj-24033 206 12 . . PUNCT brj-24033 207 1 when when SCONJ brj-24033 207 2 some some DET brj-24033 207 3 new new ADJ brj-24033 207 4 classes class NOUN brj-24033 207 5 are be AUX brj-24033 207 6 added add VERB brj-24033 207 7 into into ADP brj-24033 207 8 a a DET brj-24033 207 9 training training NOUN brj-24033 207 10 set set NOUN brj-24033 207 11 of of ADP brj-24033 207 12 the the DET brj-24033 207 13 open open ADJ brj-24033 207 14 set set NOUN brj-24033 207 15 recognition recognition NOUN brj-24033 207 16 ( ( PUNCT brj-24033 207 17 osr osr NOUN brj-24033 207 18 ) ) PUNCT brj-24033 207 19 classifier classifier NOUN brj-24033 207 20 , , PUNCT brj-24033 207 21 the the DET brj-24033 207 22 previous previous ADJ brj-24033 207 23 projection projection NOUN brj-24033 207 24 matrix matrix NOUN brj-24033 207 25 w w NOUN brj-24033 207 26 can can AUX brj-24033 207 27 still still ADV brj-24033 207 28 be be AUX brj-24033 207 29 used use VERB brj-24033 207 30 with with ADP brj-24033 207 31 near near ADV brj-24033 207 32 - - PUNCT brj-24033 207 33 zero zero NUM brj-24033 207 34 errors error NOUN brj-24033 207 35 , , PUNCT brj-24033 207 36 as as SCONJ brj-24033 207 37 pointed point VERB brj-24033 207 38 out out ADP brj-24033 207 39 by by ADP brj-24033 207 40 mensink mensink PROPN brj-24033 207 41 et et PROPN brj-24033 207 42 al al PROPN brj-24033 207 43 . . PROPN brj-24033 208 1 ( ( PUNCT brj-24033 208 2 2013 2013 NUM brj-24033 208 3 ) ) PUNCT brj-24033 208 4 . . PUNCT brj-24033 209 1 however however ADV brj-24033 209 2 , , PUNCT brj-24033 209 3 in in ADP brj-24033 209 4 other other ADJ brj-24033 209 5 feature feature NOUN brj-24033 209 6 dimension dimension NOUN brj-24033 209 7 reduction reduction NOUN brj-24033 209 8 algorithms algorithm NOUN brj-24033 209 9 such such ADJ brj-24033 209 10 as as ADP brj-24033 209 11 principal principal ADJ brj-24033 209 12 component component NOUN brj-24033 209 13 analysis analysis NOUN brj-24033 209 14 ( ( PUNCT brj-24033 209 15 pca pca NOUN brj-24033 209 16 ) ) PUNCT brj-24033 209 17 and and CCONJ brj-24033 209 18 kernel kernel PROPN brj-24033 209 19 pca pca PROPN brj-24033 209 20 , , PUNCT brj-24033 209 21 the the DET brj-24033 209 22 projection projection NOUN brj-24033 209 23 matrix matrix NOUN brj-24033 209 24 w w NOUN brj-24033 209 25 requires require VERB brj-24033 209 26 to to PART brj-24033 209 27 be be AUX brj-24033 209 28 relearned relearn VERB brj-24033 209 29 and and CCONJ brj-24033 209 30 recalculated recalculate VERB brj-24033 209 31 when when SCONJ brj-24033 209 32 new new ADJ brj-24033 209 33 classes class NOUN brj-24033 209 34 are be AUX brj-24033 209 35 added add VERB brj-24033 209 36 . . PUNCT brj-24033 210 1 2 2 X brj-24033 210 2 . . X brj-24033 210 3 by by ADP brj-24033 210 4 osr osr PROPN brj-24033 210 5 experimental experimental ADJ brj-24033 210 6 specific specific ADJ brj-24033 210 7 comparisons comparison NOUN brj-24033 210 8 in in ADP brj-24033 210 9 3 3 NUM brj-24033 210 10 groups group NOUN brj-24033 210 11 , , PUNCT brj-24033 210 12 the the DET brj-24033 210 13 proposed propose VERB brj-24033 210 14 improved improve VERB brj-24033 210 15 nearest near ADJ brj-24033 210 16 non non ADJ brj-24033 210 17 - - ADJ brj-24033 210 18 outlier outlier ADJ brj-24033 210 19 ( ( PUNCT brj-24033 210 20 nno nno NOUN brj-24033 210 21 ) ) PUNCT brj-24033 210 22 classifier classifier NOUN brj-24033 210 23 ( ( PUNCT brj-24033 210 24 nno+dpc nno+dpc NOUN brj-24033 210 25 ) ) PUNCT brj-24033 210 26 version version NOUN brj-24033 210 27 outperformed outperform VERB brj-24033 210 28 the the DET brj-24033 210 29 original original ADJ brj-24033 210 30 nno nno NOUN brj-24033 210 31 classifier classifier NOUN brj-24033 210 32 greatly greatly ADV brj-24033 210 33 . . PUNCT brj-24033 211 1 moreover moreover ADV brj-24033 211 2 , , PUNCT brj-24033 211 3 in in ADP brj-24033 211 4 most most ADJ brj-24033 211 5 cases case NOUN brj-24033 211 6 , , PUNCT brj-24033 211 7 the the DET brj-24033 211 8 improved improve VERB brj-24033 211 9 nno nno NOUN brj-24033 211 10 classifier classifier PROPN brj-24033 211 11 also also ADV brj-24033 211 12 outperformed outperform VERB brj-24033 211 13 five five NUM brj-24033 211 14 other other ADJ brj-24033 211 15 representative representative ADJ brj-24033 211 16 osr osr PROPN brj-24033 211 17 classifiers classifier NOUN brj-24033 211 18 . . PUNCT brj-24033 212 1 the the DET brj-24033 212 2 improved improve VERB brj-24033 212 3 nno nno PROPN brj-24033 212 4 classifier classifier PROPN brj-24033 212 5 ’s ’s PART brj-24033 212 6 better well ADJ brj-24033 212 7 classification classification NOUN brj-24033 212 8 performance performance NOUN brj-24033 212 9 comes come VERB brj-24033 212 10 from from ADP brj-24033 212 11 the the DET brj-24033 212 12 classification classification NOUN brj-24033 212 13 strategy strategy NOUN brj-24033 212 14 based base VERB brj-24033 212 15 on on ADP brj-24033 212 16 different different ADJ brj-24033 212 17 clusters cluster NOUN brj-24033 212 18 of of ADP brj-24033 212 19 known know VERB brj-24033 212 20 classes class NOUN brj-24033 212 21 with with ADP brj-24033 212 22 different different ADJ brj-24033 212 23 thresholds threshold NOUN brj-24033 212 24 𝜏 𝜏 X brj-24033 212 25 . . PUNCT brj-24033 213 1 in in ADP brj-24033 213 2 summary summary NOUN brj-24033 213 3 , , PUNCT brj-24033 213 4 the the DET brj-24033 213 5 proposed propose VERB brj-24033 213 6 improved improve VERB brj-24033 213 7 nno nno NOUN brj-24033 213 8 classifier classifier NOUN brj-24033 213 9 performs perform VERB brj-24033 213 10 a a DET brj-24033 213 11 more more ADV brj-24033 213 12 refined refined ADJ brj-24033 213 13 and and CCONJ brj-24033 213 14 specific specific ADJ brj-24033 213 15 classification classification NOUN brj-24033 213 16 in in ADP brj-24033 213 17 an an DET brj-24033 213 18 open open ADJ brj-24033 213 19 set set NOUN brj-24033 213 20 scenario scenario NOUN brj-24033 213 21 . . PUNCT brj-24033 214 1 3 3 X brj-24033 214 2 . . X brj-24033 214 3 in in ADP brj-24033 214 4 each each DET brj-24033 214 5 experimental experimental ADJ brj-24033 214 6 group group NOUN brj-24033 214 7 , , PUNCT brj-24033 214 8 the the DET brj-24033 214 9 proposed propose VERB brj-24033 214 10 improved improve VERB brj-24033 214 11 nno nno PROPN brj-24033 214 12 classification classification NOUN brj-24033 214 13 performance performance NOUN brj-24033 214 14 measures measure NOUN brj-24033 214 15 ( ( PUNCT brj-24033 214 16 i.e. i.e. X brj-24033 214 17 , , PUNCT brj-24033 214 18 ora ora PROPN brj-24033 214 19 , , PUNCT brj-24033 214 20 kappa kappa ADJ brj-24033 214 21 coefficient coefficient NOUN brj-24033 214 22 , , PUNCT brj-24033 214 23 and and CCONJ brj-24033 214 24 f f X brj-24033 214 25 - - PUNCT brj-24033 214 26 score score NOUN brj-24033 214 27 ) ) PUNCT brj-24033 214 28 decrease decrease NOUN brj-24033 214 29 to to ADP brj-24033 214 30 some some DET brj-24033 214 31 extents extent NOUN brj-24033 214 32 , , PUNCT brj-24033 214 33 when when SCONJ brj-24033 214 34 the the DET brj-24033 214 35 known know VERB brj-24033 214 36 class class NOUN brj-24033 214 37 number number NOUN brj-24033 214 38 increases increase NOUN brj-24033 214 39 and and CCONJ brj-24033 214 40 unknown unknown ADJ brj-24033 214 41 class class NOUN brj-24033 214 42 number number NOUN brj-24033 214 43 remains remain VERB brj-24033 214 44 fixed fix VERB brj-24033 214 45 . . PUNCT brj-24033 215 1 this this DET brj-24033 215 2 situation situation NOUN brj-24033 215 3 means mean VERB brj-24033 215 4 that that SCONJ brj-24033 215 5 the the DET brj-24033 215 6 number number NOUN brj-24033 215 7 of of ADP brj-24033 215 8 known know VERB brj-24033 215 9 classes class NOUN brj-24033 215 10 affects affect VERB brj-24033 215 11 the the DET brj-24033 215 12 nno nno NOUN brj-24033 215 13 classification classification NOUN brj-24033 215 14 performance performance NOUN brj-24033 215 15 . . PUNCT brj-24033 216 1 peer peer NOUN brj-24033 216 2 - - PUNCT brj-24033 216 3 reviewed review VERB brj-24033 216 4 article article NOUN brj-24033 216 5 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 216 6 zhang zhang PROPN brj-24033 216 7 & & CCONJ brj-24033 216 8 zhao zhao PROPN brj-24033 216 9 ( ( PUNCT brj-24033 216 10 2025 2025 NUM brj-24033 216 11 ) ) PUNCT brj-24033 216 12 . . PUNCT brj-24033 217 1 β€œ " PUNCT brj-24033 217 2 wood wood NOUN brj-24033 217 3 image image NOUN brj-24033 217 4 classification classification NOUN brj-24033 217 5 , , PUNCT brj-24033 217 6 ” " PUNCT brj-24033 217 7 bioresources bioresource NOUN brj-24033 217 8 20(1 20(1 NUM brj-24033 217 9 ) ) PUNCT brj-24033 217 10 , , PUNCT brj-24033 217 11 944 944 NUM brj-24033 217 12 - - SYM brj-24033 217 13 955 955 NUM brj-24033 217 14 . . PUNCT brj-24033 218 1 954 954 NUM brj-24033 218 2 acknowledgements acknowledgement NOUN brj-24033 218 3 this this DET brj-24033 218 4 research research NOUN brj-24033 218 5 was be AUX brj-24033 218 6 supported support VERB brj-24033 218 7 by by ADP brj-24033 218 8 the the DET brj-24033 218 9 national national ADJ brj-24033 218 10 natural natural PROPN brj-24033 218 11 science science PROPN brj-24033 218 12 foundation foundation PROPN brj-24033 218 13 of of ADP brj-24033 218 14 china china PROPN brj-24033 218 15 ( ( PUNCT brj-24033 218 16 grant grant VERB brj-24033 218 17 number number NOUN brj-24033 218 18 62265001 62265001 NUM brj-24033 218 19 ) ) PUNCT brj-24033 218 20 , , PUNCT brj-24033 218 21 and and CCONJ brj-24033 218 22 the the DET brj-24033 218 23 guangxi guangxi PROPN brj-24033 218 24 university university PROPN brj-24033 218 25 of of ADP brj-24033 218 26 science science NOUN brj-24033 218 27 and and CCONJ brj-24033 218 28 technology technology NOUN brj-24033 218 29 doctoral doctoral ADJ brj-24033 218 30 research research NOUN brj-24033 218 31 fund fund NOUN brj-24033 218 32 ( ( PUNCT brj-24033 218 33 grant grant VERB brj-24033 218 34 number number NOUN brj-24033 218 35 22z07 22z07 NUM brj-24033 218 36 ) ) PUNCT brj-24033 218 37 . . PUNCT brj-24033 219 1 availability availability NOUN brj-24033 219 2 of of ADP brj-24033 219 3 data datum NOUN brj-24033 219 4 and and CCONJ brj-24033 219 5 materials material NOUN brj-24033 219 6 the the DET brj-24033 219 7 wood wood NOUN brj-24033 219 8 spectral spectral ADJ brj-24033 219 9 dataset dataset NOUN brj-24033 219 10 used use VERB brj-24033 219 11 in in ADP brj-24033 219 12 this this DET brj-24033 219 13 work work NOUN brj-24033 219 14 is be AUX brj-24033 219 15 confidential confidential ADJ brj-24033 219 16 , , PUNCT brj-24033 219 17 but but CCONJ brj-24033 219 18 this this DET brj-24033 219 19 dataset dataset NOUN brj-24033 219 20 used use VERB brj-24033 219 21 to to PART brj-24033 219 22 support support VERB brj-24033 219 23 the the DET brj-24033 219 24 findings finding NOUN brj-24033 219 25 of of ADP brj-24033 219 26 this this DET brj-24033 219 27 study study NOUN brj-24033 219 28 is be AUX brj-24033 219 29 available available ADJ brj-24033 219 30 from from ADP brj-24033 219 31 the the DET brj-24033 219 32 corresponding corresponding ADJ brj-24033 219 33 author author NOUN brj-24033 219 34 upon upon SCONJ brj-24033 219 35 request request NOUN brj-24033 219 36 after after SCONJ brj-24033 219 37 this this DET brj-24033 219 38 article article NOUN brj-24033 219 39 is be AUX brj-24033 219 40 accepted accept VERB brj-24033 219 41 and and CCONJ brj-24033 219 42 published publish VERB brj-24033 219 43 online online ADV brj-24033 219 44 . . PUNCT brj-24033 220 1 competing compete VERB brj-24033 220 2 interests interest NOUN brj-24033 220 3 the the DET brj-24033 220 4 authors author NOUN brj-24033 220 5 declare declare VERB brj-24033 220 6 that that SCONJ brj-24033 220 7 they they PRON brj-24033 220 8 have have VERB brj-24033 220 9 no no DET brj-24033 220 10 competing compete VERB brj-24033 220 11 interests interest NOUN brj-24033 220 12 . . PUNCT brj-24033 221 1 author author NOUN brj-24033 221 2 contributions contribution VERB brj-24033 221 3 zhang zhang PROPN brj-24033 221 4 ke ke PROPN brj-24033 221 5 - - PUNCT brj-24033 221 6 xin xin PROPN brj-24033 221 7 makes make VERB brj-24033 221 8 the the DET brj-24033 221 9 wood wood NOUN brj-24033 221 10 species species NOUN brj-24033 221 11 recognition recognition NOUN brj-24033 221 12 experiments experiment NOUN brj-24033 221 13 and and CCONJ brj-24033 221 14 collates collate VERB brj-24033 221 15 the the DET brj-24033 221 16 experimental experimental ADJ brj-24033 221 17 results result NOUN brj-24033 221 18 . . PUNCT brj-24033 222 1 zhao zhao PROPN brj-24033 222 2 peng peng PROPN brj-24033 222 3 proposes propose VERB brj-24033 222 4 the the DET brj-24033 222 5 research research NOUN brj-24033 222 6 idea idea NOUN brj-24033 222 7 and and CCONJ brj-24033 222 8 the the DET brj-24033 222 9 experimental experimental ADJ brj-24033 222 10 framework framework NOUN brj-24033 222 11 , , PUNCT brj-24033 222 12 writing write VERB brj-24033 222 13 the the DET brj-24033 222 14 whole whole ADJ brj-24033 222 15 manuscript manuscript NOUN brj-24033 222 16 . . PUNCT brj-24033 223 1 all all DET brj-24033 223 2 authors author NOUN brj-24033 223 3 read read VERB brj-24033 223 4 and and CCONJ brj-24033 223 5 approve approve VERB brj-24033 223 6 the the DET brj-24033 223 7 final final ADJ brj-24033 223 8 manuscript manuscript NOUN brj-24033 223 9 . . PUNCT brj-24033 224 1 references reference NOUN brj-24033 224 2 cited cite VERB brj-24033 224 3 alhayani alhayani PROPN brj-24033 224 4 , , PUNCT brj-24033 224 5 b. b. PROPN brj-24033 224 6 , , PUNCT brj-24033 224 7 and and CCONJ brj-24033 224 8 ilhan ilhan PROPN brj-24033 224 9 , , PUNCT brj-24033 224 10 h. h. PROPN brj-24033 224 11 ( ( PUNCT brj-24033 224 12 2017 2017 NUM brj-24033 224 13 ) ) PUNCT brj-24033 224 14 . . PUNCT brj-24033 225 1 β€œ " PUNCT brj-24033 225 2 hyper hyper ADJ brj-24033 225 3 - - ADJ brj-24033 225 4 spectral spectral ADJ brj-24033 225 5 image image NOUN brj-24033 225 6 classification classification NOUN brj-24033 225 7 using use VERB brj-24033 225 8 dimensionality dimensionality NOUN brj-24033 225 9 reduction reduction NOUN brj-24033 225 10 techniques technique NOUN brj-24033 225 11 , , PUNCT brj-24033 225 12 ” " PUNCT brj-24033 225 13 international international ADJ brj-24033 225 14 journal journal NOUN brj-24033 225 15 of of ADP brj-24033 225 16 innovative innovative ADJ brj-24033 225 17 research research NOUN brj-24033 225 18 in in ADP brj-24033 225 19 electrical electrical ADJ brj-24033 225 20 , , PUNCT brj-24033 225 21 electronics electronic NOUN brj-24033 225 22 , , PUNCT brj-24033 225 23 instrumentation instrumentation NOUN brj-24033 225 24 and and CCONJ brj-24033 225 25 control control NOUN brj-24033 225 26 engineering engineering NOUN brj-24033 225 27 5(4 5(4 NOUN brj-24033 225 28 ) ) PUNCT brj-24033 225 29 , , PUNCT brj-24033 225 30 71 71 NUM brj-24033 225 31 - - SYM brj-24033 225 32 74 74 NUM brj-24033 225 33 . . PUNCT brj-24033 226 1 doi doi NOUN brj-24033 226 2 : : PUNCT brj-24033 226 3 10.17148 10.17148 NUM brj-24033 226 4 / / SYM brj-24033 226 5 ijireeice.2017.5414 ijireeice.2017.5414 NOUN brj-24033 226 6 belkin belkin NOUN brj-24033 226 7 , , PUNCT brj-24033 226 8 m. m. NOUN brj-24033 226 9 , , PUNCT brj-24033 226 10 and and CCONJ brj-24033 226 11 niyogi niyogi ADV brj-24033 226 12 , , PUNCT brj-24033 226 13 p. p. NOUN brj-24033 226 14 ( ( PUNCT brj-24033 226 15 2003 2003 NUM brj-24033 226 16 ) ) PUNCT brj-24033 226 17 . . PUNCT brj-24033 227 1 β€œ " PUNCT brj-24033 227 2 laplacian laplacian ADJ brj-24033 227 3 eigenmaps eigenmap NOUN brj-24033 227 4 for for ADP brj-24033 227 5 dimensionality dimensionality NOUN brj-24033 227 6 reduction reduction NOUN brj-24033 227 7 and and CCONJ brj-24033 227 8 data datum NOUN brj-24033 227 9 representation representation NOUN brj-24033 227 10 , , PUNCT brj-24033 227 11 ” " PUNCT brj-24033 227 12 neural neural ADJ brj-24033 227 13 computation computation NOUN brj-24033 227 14 15(6 15(6 NUM brj-24033 227 15 ) ) PUNCT brj-24033 227 16 , , PUNCT brj-24033 227 17 1373 1373 NUM brj-24033 227 18 - - SYM brj-24033 227 19 1396 1396 NUM brj-24033 227 20 . . PUNCT brj-24033 228 1 doi doi NOUN brj-24033 228 2 : : PUNCT brj-24033 228 3 10.1162/089976603321780317 10.1162/089976603321780317 NUM brj-24033 228 4 bendale bendale NOUN brj-24033 228 5 , , PUNCT brj-24033 228 6 a. a. NOUN brj-24033 228 7 , , PUNCT brj-24033 228 8 and and CCONJ brj-24033 228 9 boult boult NOUN brj-24033 228 10 , , PUNCT brj-24033 228 11 t. t. PROPN brj-24033 228 12 e. e. PROPN brj-24033 228 13 ( ( PUNCT brj-24033 228 14 2015 2015 NUM brj-24033 228 15 ) ) PUNCT brj-24033 228 16 . . PUNCT brj-24033 229 1 β€œ " PUNCT brj-24033 229 2 towards towards ADP brj-24033 229 3 open open ADJ brj-24033 229 4 world world NOUN brj-24033 229 5 recognition recognition NOUN brj-24033 229 6 , , PUNCT brj-24033 229 7 ” " PUNCT brj-24033 229 8 in in ADP brj-24033 229 9 : : PUNCT brj-24033 229 10 proceedings proceeding NOUN brj-24033 229 11 of of ADP brj-24033 229 12 ieee ieee NOUN brj-24033 229 13 computer computer NOUN brj-24033 229 14 vision vision NOUN brj-24033 229 15 and and CCONJ brj-24033 229 16 pattern pattern NOUN brj-24033 229 17 recognition recognition NOUN brj-24033 229 18 , , PUNCT brj-24033 229 19 boston boston PROPN brj-24033 229 20 , , PUNCT brj-24033 229 21 usa usa PROPN brj-24033 229 22 , , PUNCT brj-24033 229 23 pp.1893 pp.1893 PROPN brj-24033 229 24 - - PUNCT brj-24033 229 25 1902 1902 NUM brj-24033 229 26 . . PUNCT brj-24033 230 1 bendale bendale PROPN brj-24033 230 2 , , PUNCT brj-24033 230 3 a. a. NOUN brj-24033 230 4 , , PUNCT brj-24033 230 5 and and CCONJ brj-24033 230 6 boult boult NOUN brj-24033 230 7 , , PUNCT brj-24033 230 8 t. t. PROPN brj-24033 230 9 e. e. PROPN brj-24033 230 10 ( ( PUNCT brj-24033 230 11 2016 2016 NUM brj-24033 230 12 ) ) PUNCT brj-24033 230 13 . . PUNCT brj-24033 231 1 β€œ " PUNCT brj-24033 231 2 towards towards ADP brj-24033 231 3 open open ADJ brj-24033 231 4 set set VERB brj-24033 231 5 deep deep ADJ brj-24033 231 6 networks network NOUN brj-24033 231 7 , , PUNCT brj-24033 231 8 ” " PUNCT brj-24033 231 9 in in ADP brj-24033 231 10 : : PUNCT brj-24033 231 11 proceedings proceeding NOUN brj-24033 231 12 of of ADP brj-24033 231 13 ieee ieee NOUN brj-24033 231 14 computer computer NOUN brj-24033 231 15 vision vision NOUN brj-24033 231 16 and and CCONJ brj-24033 231 17 pattern pattern NOUN brj-24033 231 18 recognition recognition NOUN brj-24033 231 19 , , PUNCT brj-24033 231 20 las las PROPN brj-24033 231 21 vegas vegas PROPN brj-24033 231 22 , , PUNCT brj-24033 231 23 usa usa PROPN brj-24033 231 24 , , PUNCT brj-24033 231 25 pp.1563 pp.1563 PROPN brj-24033 231 26 - - PUNCT brj-24033 231 27 1572 1572 NUM brj-24033 231 28 . . PUNCT brj-24033 232 1 geng geng PROPN brj-24033 232 2 , , PUNCT brj-24033 232 3 c. c. PROPN brj-24033 232 4 x. x. PROPN brj-24033 232 5 , , PUNCT brj-24033 232 6 huang huang PROPN brj-24033 232 7 , , PUNCT brj-24033 232 8 s. s. PROPN brj-24033 232 9 j. j. PROPN brj-24033 232 10 , , PUNCT brj-24033 232 11 and and CCONJ brj-24033 232 12 chen chen PROPN brj-24033 232 13 , , PUNCT brj-24033 232 14 s. s. PROPN brj-24033 232 15 c. c. PROPN brj-24033 232 16 ( ( PUNCT brj-24033 232 17 2021 2021 NUM brj-24033 232 18 ) ) PUNCT brj-24033 232 19 . . PUNCT brj-24033 233 1 β€œ " PUNCT brj-24033 233 2 recent recent ADJ brj-24033 233 3 advances advance NOUN brj-24033 233 4 in in ADP brj-24033 233 5 open open ADJ brj-24033 233 6 set set NOUN brj-24033 233 7 recognition recognition NOUN brj-24033 233 8 : : PUNCT brj-24033 233 9 a a DET brj-24033 233 10 survey survey NOUN brj-24033 233 11 , , PUNCT brj-24033 233 12 ” " PUNCT brj-24033 233 13 ieee ieee NOUN brj-24033 233 14 trans trans PROPN brj-24033 233 15 pattern pattern NOUN brj-24033 233 16 analysis analysis NOUN brj-24033 233 17 and and CCONJ brj-24033 233 18 machine machine NOUN brj-24033 233 19 intelligence intelligence NOUN brj-24033 233 20 43(10 43(10 PROPN brj-24033 233 21 ) ) PUNCT brj-24033 233 22 , , PUNCT brj-24033 233 23 3614 3614 NUM brj-24033 233 24 - - SYM brj-24033 233 25 3631 3631 NUM brj-24033 233 26 . . PUNCT brj-24033 234 1 doi doi NOUN brj-24033 234 2 : : PUNCT brj-24033 234 3 10.1109 10.1109 NUM brj-24033 234 4 / / SYM brj-24033 234 5 tpami.2020.2981604 tpami.2020.2981604 VERB brj-24033 234 6 jain jain NOUN brj-24033 234 7 , , PUNCT brj-24033 234 8 l. l. PROPN brj-24033 234 9 p. p. PROPN brj-24033 234 10 , , PUNCT brj-24033 234 11 scheirer scheirer ADV brj-24033 234 12 , , PUNCT brj-24033 234 13 w. w. PROPN brj-24033 234 14 j. j. PROPN brj-24033 234 15 , , PUNCT brj-24033 234 16 and and CCONJ brj-24033 234 17 boult boult NOUN brj-24033 234 18 , , PUNCT brj-24033 234 19 t. t. PROPN brj-24033 234 20 e. e. PROPN brj-24033 234 21 ( ( PUNCT brj-24033 234 22 2014 2014 NUM brj-24033 234 23 ) ) PUNCT brj-24033 234 24 . . PUNCT brj-24033 235 1 β€œ " PUNCT brj-24033 235 2 multi multi ADJ brj-24033 235 3 - - ADJ brj-24033 235 4 class class ADJ brj-24033 235 5 open open ADJ brj-24033 235 6 set set NOUN brj-24033 235 7 recognition recognition NOUN brj-24033 235 8 using use VERB brj-24033 235 9 probability probability NOUN brj-24033 235 10 of of ADP brj-24033 235 11 inclusion inclusion NOUN brj-24033 235 12 , , PUNCT brj-24033 235 13 ” " PUNCT brj-24033 235 14 in in ADP brj-24033 235 15 : : PUNCT brj-24033 235 16 proceedings proceeding NOUN brj-24033 235 17 of of ADP brj-24033 235 18 european european ADJ brj-24033 235 19 conference conference NOUN brj-24033 235 20 of of ADP brj-24033 235 21 computer computer NOUN brj-24033 235 22 vision vision NOUN brj-24033 235 23 , , PUNCT brj-24033 235 24 zurich zurich PROPN brj-24033 235 25 , , PUNCT brj-24033 235 26 switzerland switzerland PROPN brj-24033 235 27 , , PUNCT brj-24033 235 28 pp pp ADJ brj-24033 235 29 . . PUNCT brj-24033 236 1 393 393 NUM brj-24033 236 2 - - SYM brj-24033 236 3 409 409 NUM brj-24033 236 4 . . PUNCT brj-24033 237 1 junior junior PROPN brj-24033 237 2 , , PUNCT brj-24033 237 3 p. p. PROPN brj-24033 237 4 r. r. PROPN brj-24033 237 5 m. m. PROPN brj-24033 237 6 , , PUNCT brj-24033 237 7 souza souza PROPN brj-24033 237 8 , , PUNCT brj-24033 237 9 r. r. PROPN brj-24033 237 10 m. m. PROPN brj-24033 237 11 d. d. PROPN brj-24033 237 12 , , PUNCT brj-24033 237 13 werneck werneck PROPN brj-24033 237 14 , , PUNCT brj-24033 237 15 r. r. PROPN brj-24033 237 16 d. d. PROPN brj-24033 237 17 o. o. PROPN brj-24033 237 18 , , PUNCT brj-24033 237 19 stein stein PROPN brj-24033 237 20 , , PUNCT brj-24033 237 21 b. b. PROPN brj-24033 238 1 v. v. PROPN brj-24033 238 2 , , PUNCT brj-24033 238 3 pazinato pazinato PROPN brj-24033 238 4 , , PUNCT brj-24033 238 5 d. d. PROPN brj-24033 238 6 v. v. PROPN brj-24033 238 7 , , PUNCT brj-24033 238 8 almeida almeida PROPN brj-24033 238 9 , , PUNCT brj-24033 238 10 w. w. PROPN brj-24033 238 11 r. r. PROPN brj-24033 238 12 , , PUNCT brj-24033 238 13 penatti penatti PROPN brj-24033 238 14 , , PUNCT brj-24033 238 15 o. o. PROPN brj-24033 238 16 a. a. PROPN brj-24033 238 17 , , PUNCT brj-24033 238 18 torres torre NOUN brj-24033 238 19 , , PUNCT brj-24033 238 20 r. r. PROPN brj-24033 238 21 d. d. PROPN brj-24033 238 22 , , PUNCT brj-24033 238 23 and and CCONJ brj-24033 238 24 rocha rocha PROPN brj-24033 238 25 , , PUNCT brj-24033 238 26 a. a. NOUN brj-24033 238 27 ( ( PUNCT brj-24033 238 28 2017 2017 NUM brj-24033 238 29 ) ) PUNCT brj-24033 238 30 . . PUNCT brj-24033 239 1 β€œ " PUNCT brj-24033 239 2 nearest near ADJ brj-24033 239 3 neighbors neighbor NOUN brj-24033 239 4 distance distance NOUN brj-24033 239 5 ratio ratio NOUN brj-24033 239 6 open open ADJ brj-24033 239 7 set set NOUN brj-24033 239 8 classifier classifier NOUN brj-24033 239 9 , , PUNCT brj-24033 239 10 ” " PUNCT brj-24033 239 11 machine machine NOUN brj-24033 239 12 learning learn VERB brj-24033 239 13 106(3 106(3 NUM brj-24033 239 14 ) ) PUNCT brj-24033 239 15 , , PUNCT brj-24033 239 16 359 359 NUM brj-24033 239 17 - - SYM brj-24033 239 18 386 386 NUM brj-24033 239 19 . . PUNCT brj-24033 239 20 doi doi NOUN brj-24033 239 21 : : PUNCT brj-24033 239 22 10.1007 10.1007 NUM brj-24033 239 23 / / SYM brj-24033 239 24 s10994 s10994 VERB brj-24033 239 25 - - PUNCT brj-24033 239 26 016 016 NUM brj-24033 239 27 - - PUNCT brj-24033 239 28 5610 5610 NUM brj-24033 239 29 - - SYM brj-24033 239 30 8 8 NUM brj-24033 239 31 ma ma PROPN brj-24033 239 32 , , PUNCT brj-24033 239 33 t. t. PROPN brj-24033 239 34 , , PUNCT brj-24033 239 35 inagaki inagaki PROPN brj-24033 239 36 , , PUNCT brj-24033 239 37 t. t. PROPN brj-24033 239 38 , , PUNCT brj-24033 239 39 and and CCONJ brj-24033 239 40 tsuchikawa tsuchikawa PROPN brj-24033 239 41 , , PUNCT brj-24033 239 42 s. s. PROPN brj-24033 239 43 ( ( PUNCT brj-24033 239 44 2021 2021 NUM brj-24033 239 45 ) ) PUNCT brj-24033 239 46 . . PUNCT brj-24033 240 1 β€œ " PUNCT brj-24033 240 2 demonstration demonstration NOUN brj-24033 240 3 of of ADP brj-24033 240 4 the the DET brj-24033 240 5 applicability applicability NOUN brj-24033 240 6 of of ADP brj-24033 240 7 visible visible ADJ brj-24033 240 8 and and CCONJ brj-24033 240 9 near near ADV brj-24033 240 10 - - PUNCT brj-24033 240 11 infrared infrare VERB brj-24033 240 12 spatially spatially ADV brj-24033 240 13 resolved resolve VERB brj-24033 240 14 spectroscopy spectroscopy NOUN brj-24033 240 15 for for ADP brj-24033 240 16 rapid rapid ADJ brj-24033 240 17 and and CCONJ brj-24033 240 18 nondestructive nondestructive ADJ brj-24033 240 19 wood wood NOUN brj-24033 240 20 classification classification NOUN brj-24033 240 21 , , PUNCT brj-24033 240 22 ” " PUNCT brj-24033 240 23 holzforschung holzforschung PROPN brj-24033 240 24 75(5 75(5 NUM brj-24033 240 25 ) ) PUNCT brj-24033 240 26 , , PUNCT brj-24033 240 27 419 419 NUM brj-24033 240 28 - - SYM brj-24033 240 29 427 427 NUM brj-24033 240 30 . . PUNCT brj-24033 241 1 doi doi NOUN brj-24033 241 2 : : PUNCT brj-24033 241 3 10.1515 10.1515 NUM brj-24033 241 4 / / SYM brj-24033 241 5 hf-2020 hf-2020 PROPN brj-24033 241 6 - - PUNCT brj-24033 241 7 0074 0074 NUM brj-24033 241 8 mensink mensink NOUN brj-24033 241 9 , , PUNCT brj-24033 241 10 t. t. PROPN brj-24033 241 11 , , PUNCT brj-24033 241 12 verbeek verbeek NOUN brj-24033 241 13 , , PUNCT brj-24033 241 14 j. j. PROPN brj-24033 241 15 , , PUNCT brj-24033 241 16 perronnin perronnin PROPN brj-24033 241 17 , , PUNCT brj-24033 241 18 f. f. PROPN brj-24033 241 19 , , PUNCT brj-24033 241 20 and and CCONJ brj-24033 241 21 csurka csurka PROPN brj-24033 241 22 , , PUNCT brj-24033 241 23 g. g. PROPN brj-24033 241 24 ( ( PUNCT brj-24033 241 25 2013 2013 NUM brj-24033 241 26 ) ) PUNCT brj-24033 241 27 . . PUNCT brj-24033 242 1 β€œ " PUNCT brj-24033 242 2 distance distance NOUN brj-24033 242 3 - - PUNCT brj-24033 242 4 based base VERB brj-24033 242 5 image image NOUN brj-24033 242 6 peer peer NOUN brj-24033 242 7 - - PUNCT brj-24033 242 8 reviewed review VERB brj-24033 242 9 article article NOUN brj-24033 242 10 bioresources.cnr.ncsu.edu bioresources.cnr.ncsu.edu X brj-24033 242 11 zhang zhang PROPN brj-24033 242 12 & & CCONJ brj-24033 242 13 zhao zhao PROPN brj-24033 242 14 ( ( PUNCT brj-24033 242 15 2025 2025 NUM brj-24033 242 16 ) ) PUNCT brj-24033 242 17 . . PUNCT brj-24033 243 1 β€œ " PUNCT brj-24033 243 2 wood wood NOUN brj-24033 243 3 image image NOUN brj-24033 243 4 classification classification NOUN brj-24033 243 5 , , PUNCT brj-24033 243 6 ” " PUNCT brj-24033 243 7 bioresources bioresource NOUN brj-24033 243 8 20(1 20(1 NUM brj-24033 243 9 ) ) PUNCT brj-24033 243 10 , , PUNCT brj-24033 243 11 944 944 NUM brj-24033 243 12 - - SYM brj-24033 243 13 955 955 NUM brj-24033 243 14 . . PUNCT brj-24033 244 1 955 955 NUM brj-24033 244 2 classification classification NOUN brj-24033 244 3 : : PUNCT brj-24033 244 4 generalizing generalize VERB brj-24033 244 5 to to ADP brj-24033 244 6 new new ADJ brj-24033 244 7 classes class NOUN brj-24033 244 8 at at ADP brj-24033 244 9 near near ADV brj-24033 244 10 - - PUNCT brj-24033 244 11 zero zero NUM brj-24033 244 12 cost cost NOUN brj-24033 244 13 , , PUNCT brj-24033 244 14 ” " PUNCT brj-24033 244 15 ieee ieee NOUN brj-24033 244 16 trans trans PROPN brj-24033 244 17 pattern pattern NOUN brj-24033 244 18 analysis analysis NOUN brj-24033 244 19 and and CCONJ brj-24033 244 20 machine machine NOUN brj-24033 244 21 intelligence intelligence NOUN brj-24033 244 22 35(11 35(11 NUM brj-24033 244 23 ) ) PUNCT brj-24033 244 24 , , PUNCT brj-24033 244 25 2624 2624 NUM brj-24033 244 26 - - SYM brj-24033 244 27 2637 2637 NUM brj-24033 244 28 . . PUNCT brj-24033 245 1 doi doi NOUN brj-24033 245 2 : : PUNCT brj-24033 245 3 10.1109 10.1109 NUM brj-24033 245 4 / / SYM brj-24033 245 5 tpami.2013.83 tpami.2013.83 NOUN brj-24033 245 6 mignotte mignotte NOUN brj-24033 245 7 , , PUNCT brj-24033 245 8 m. m. NOUN brj-24033 245 9 ( ( PUNCT brj-24033 245 10 2011 2011 NUM brj-24033 245 11 ) ) PUNCT brj-24033 245 12 . . PUNCT brj-24033 246 1 β€œ " PUNCT brj-24033 246 2 mds mds NOUN brj-24033 246 3 - - PUNCT brj-24033 246 4 based base VERB brj-24033 246 5 multiresolution multiresolution NOUN brj-24033 246 6 nonlinear nonlinear ADJ brj-24033 246 7 dimensionality dimensionality NOUN brj-24033 246 8 reduction reduction NOUN brj-24033 246 9 model model NOUN brj-24033 246 10 for for ADP brj-24033 246 11 color color NOUN brj-24033 246 12 image image NOUN brj-24033 246 13 segmentation segmentation NOUN brj-24033 246 14 , , PUNCT brj-24033 246 15 ” " PUNCT brj-24033 246 16 ieee ieee NOUN brj-24033 246 17 transactions transaction NOUN brj-24033 246 18 on on ADP brj-24033 246 19 neural neural ADJ brj-24033 246 20 networks network NOUN brj-24033 246 21 22(3 22(3 NOUN brj-24033 246 22 ) ) PUNCT brj-24033 246 23 , , PUNCT brj-24033 246 24 447 447 NUM brj-24033 246 25 - - SYM brj-24033 246 26 460 460 NUM brj-24033 246 27 . . PUNCT brj-24033 247 1 doi doi NOUN brj-24033 247 2 : : PUNCT brj-24033 247 3 10.1109 10.1109 NUM brj-24033 247 4 / / SYM brj-24033 247 5 tnn.2010.2101614 tnn.2010.2101614 NUM brj-24033 247 6 park park NOUN brj-24033 247 7 , , PUNCT brj-24033 247 8 s. s. PROPN brj-24033 247 9 y. y. PROPN brj-24033 247 10 , , PUNCT brj-24033 247 11 kim kim PROPN brj-24033 247 12 , , PUNCT brj-24033 247 13 j. j. PROPN brj-24033 247 14 h. h. PROPN brj-24033 247 15 , , PUNCT brj-24033 247 16 kim kim PROPN brj-24033 247 17 , , PUNCT brj-24033 247 18 j. j. PROPN brj-24033 247 19 c. c. PROPN brj-24033 247 20 , , PUNCT brj-24033 247 21 yang yang PROPN brj-24033 247 22 , , PUNCT brj-24033 247 23 s. s. PROPN brj-24033 247 24 y. y. PROPN brj-24033 247 25 , , PUNCT brj-24033 247 26 and and CCONJ brj-24033 247 27 choi choi NOUN brj-24033 247 28 , , PUNCT brj-24033 247 29 i. i. PROPN brj-24033 247 30 g. g. PROPN brj-24033 247 31 ( ( PUNCT brj-24033 247 32 2021 2021 NUM brj-24033 247 33 ) ) PUNCT brj-24033 247 34 . . PUNCT brj-24033 248 1 β€œ " PUNCT brj-24033 248 2 classification classification NOUN brj-24033 248 3 of of ADP brj-24033 248 4 softwoods softwood NOUN brj-24033 248 5 using use VERB brj-24033 248 6 wood wood NOUN brj-24033 248 7 extract extract NOUN brj-24033 248 8 information information NOUN brj-24033 248 9 and and CCONJ brj-24033 248 10 near near ADP brj-24033 248 11 infrared infrared ADJ brj-24033 248 12 spectroscopy spectroscopy NOUN brj-24033 248 13 , , PUNCT brj-24033 248 14 ” " PUNCT brj-24033 248 15 bioresources bioresource NOUN brj-24033 248 16 16(3 16(3 NUM brj-24033 248 17 ) ) PUNCT brj-24033 248 18 , , PUNCT brj-24033 248 19 75 75 NUM brj-24033 248 20 - - SYM brj-24033 248 21 80 80 NUM brj-24033 248 22 . . PUNCT brj-24033 249 1 doi doi NOUN brj-24033 249 2 : : PUNCT brj-24033 249 3 10.15376 10.15376 NUM brj-24033 249 4 / / SYM brj-24033 249 5 biores.16.3.5301 biores.16.3.5301 PROPN brj-24033 249 6 - - PUNCT brj-24033 249 7 5312 5312 NUM brj-24033 249 8 reddy reddy NOUN brj-24033 249 9 , , PUNCT brj-24033 249 10 g. g. PROPN brj-24033 249 11 t. t. PROPN brj-24033 249 12 , , PUNCT brj-24033 249 13 reddy reddy PROPN brj-24033 249 14 , , PUNCT brj-24033 249 15 m. m. NOUN brj-24033 249 16 p. p. PROPN brj-24033 249 17 k. k. PROPN brj-24033 249 18 , , PUNCT brj-24033 249 19 lakshmanna lakshmanna PROPN brj-24033 249 20 , , PUNCT brj-24033 249 21 k. k. PROPN brj-24033 249 22 , , PUNCT brj-24033 249 23 kaluri kaluri PROPN brj-24033 249 24 , , PUNCT brj-24033 249 25 r. r. PROPN brj-24033 249 26 , , PUNCT brj-24033 249 27 rajput rajput PROPN brj-24033 249 28 , , PUNCT brj-24033 249 29 d. d. PROPN brj-24033 249 30 s. s. PROPN brj-24033 249 31 , , PUNCT brj-24033 249 32 and and CCONJ brj-24033 249 33 srivastava srivastava PROPN brj-24033 249 34 , , PUNCT brj-24033 249 35 g. g. PROPN brj-24033 249 36 ( ( PUNCT brj-24033 249 37 2020 2020 NUM brj-24033 249 38 ) ) PUNCT brj-24033 249 39 . . PUNCT brj-24033 250 1 β€œ " PUNCT brj-24033 250 2 analysis analysis NOUN brj-24033 250 3 of of ADP brj-24033 250 4 dimensionality dimensionality NOUN brj-24033 250 5 reduction reduction NOUN brj-24033 250 6 techniques technique NOUN brj-24033 250 7 on on ADP brj-24033 250 8 big big ADJ brj-24033 250 9 data datum NOUN brj-24033 250 10 , , PUNCT brj-24033 250 11 ” " PUNCT brj-24033 250 12 ieee ieee NOUN brj-24033 250 13 access access NOUN brj-24033 250 14 8 8 NUM brj-24033 250 15 , , PUNCT brj-24033 250 16 54776 54776 NUM brj-24033 250 17 - - SYM brj-24033 250 18 54788 54788 NUM brj-24033 250 19 . . PUNCT brj-24033 251 1 doi doi NOUN brj-24033 251 2 : : PUNCT brj-24033 251 3 10.1109 10.1109 NUM brj-24033 251 4 / / SYM brj-24033 251 5 access.2020.2980942 access.2020.2980942 NUM brj-24033 251 6 rodriguez rodriguez NOUN brj-24033 251 7 , , PUNCT brj-24033 251 8 a. a. NOUN brj-24033 251 9 , , PUNCT brj-24033 251 10 and and CCONJ brj-24033 251 11 laio laio NOUN brj-24033 251 12 , , PUNCT brj-24033 251 13 a. a. NOUN brj-24033 251 14 ( ( PUNCT brj-24033 251 15 2014 2014 NUM brj-24033 251 16 ) ) PUNCT brj-24033 251 17 . . PUNCT brj-24033 252 1 β€œ " PUNCT brj-24033 252 2 clustering cluster VERB brj-24033 252 3 by by ADP brj-24033 252 4 fast fast ADJ brj-24033 252 5 search search NOUN brj-24033 252 6 and and CCONJ brj-24033 252 7 find find NOUN brj-24033 252 8 of of ADP brj-24033 252 9 density density NOUN brj-24033 252 10 peaks peak NOUN brj-24033 252 11 , , PUNCT brj-24033 252 12 ” " PUNCT brj-24033 252 13 science science NOUN brj-24033 252 14 344(6191 344(6191 PROPN brj-24033 252 15 ) ) PUNCT brj-24033 252 16 , , PUNCT brj-24033 252 17 1492 1492 NUM brj-24033 252 18 - - SYM brj-24033 252 19 1496 1496 NUM brj-24033 252 20 . . PUNCT brj-24033 253 1 doi doi NOUN brj-24033 253 2 : : PUNCT brj-24033 253 3 10.1126 10.1126 NUM brj-24033 253 4 / / SYM brj-24033 253 5 science.1242072 science.1242072 SYM brj-24033 253 6 scheirer scheirer ADV brj-24033 253 7 , , PUNCT brj-24033 253 8 w. w. PROPN brj-24033 253 9 j. j. PROPN brj-24033 253 10 , , PUNCT brj-24033 253 11 jain jain PROPN brj-24033 253 12 , , PUNCT brj-24033 253 13 l. l. PROPN brj-24033 253 14 p. p. PROPN brj-24033 253 15 , , PUNCT brj-24033 253 16 and and CCONJ brj-24033 253 17 boult boult NOUN brj-24033 253 18 , , PUNCT brj-24033 253 19 t. t. PROPN brj-24033 253 20 e. e. PROPN brj-24033 253 21 ( ( PUNCT brj-24033 253 22 2014 2014 NUM brj-24033 253 23 ) ) PUNCT brj-24033 253 24 . . PUNCT brj-24033 254 1 β€œ " PUNCT brj-24033 254 2 probability probability NOUN brj-24033 254 3 models model NOUN brj-24033 254 4 for for ADP brj-24033 254 5 open open ADJ brj-24033 254 6 set set NOUN brj-24033 254 7 recognition recognition NOUN brj-24033 254 8 , , PUNCT brj-24033 254 9 ” " PUNCT brj-24033 254 10 ieee ieee NOUN brj-24033 254 11 trans trans PROPN brj-24033 254 12 pattern pattern NOUN brj-24033 254 13 analysis analysis NOUN brj-24033 254 14 and and CCONJ brj-24033 254 15 machine machine NOUN brj-24033 254 16 intelligence intelligence NOUN brj-24033 254 17 36(11 36(11 PROPN brj-24033 254 18 ) ) PUNCT brj-24033 254 19 , , PUNCT brj-24033 254 20 23172324 23172324 NUM brj-24033 254 21 . . PUNCT brj-24033 255 1 doi doi NOUN brj-24033 255 2 : : PUNCT brj-24033 255 3 10.1109 10.1109 NUM brj-24033 255 4 / / SYM brj-24033 255 5 tpami.2014.2321392 tpami.2014.2321392 PROPN brj-24033 255 6 scheirer scheirer ADV brj-24033 255 7 , , PUNCT brj-24033 255 8 w. w. PROPN brj-24033 255 9 j. j. PROPN brj-24033 255 10 , , PUNCT brj-24033 255 11 rocha rocha PROPN brj-24033 255 12 , , PUNCT brj-24033 255 13 a. a. PROPN brj-24033 255 14 d. d. PROPN brj-24033 255 15 r. r. PROPN brj-24033 255 16 , , PUNCT brj-24033 255 17 sapkota sapkota NOUN brj-24033 255 18 , , PUNCT brj-24033 255 19 a. a. NOUN brj-24033 255 20 , , PUNCT brj-24033 255 21 and and CCONJ brj-24033 255 22 boult boult NOUN brj-24033 255 23 , , PUNCT brj-24033 255 24 t. t. PROPN brj-24033 255 25 e. e. PROPN brj-24033 255 26 ( ( PUNCT brj-24033 255 27 2013 2013 NUM brj-24033 255 28 ) ) PUNCT brj-24033 255 29 . . PUNCT brj-24033 256 1 β€œ " PUNCT brj-24033 256 2 toward toward ADP brj-24033 256 3 open open ADJ brj-24033 256 4 set set NOUN brj-24033 256 5 recognition recognition NOUN brj-24033 256 6 , , PUNCT brj-24033 256 7 ” " PUNCT brj-24033 256 8 ieee ieee NOUN brj-24033 256 9 trans trans PROPN brj-24033 256 10 pattern pattern NOUN brj-24033 256 11 analysis analysis NOUN brj-24033 256 12 and and CCONJ brj-24033 256 13 machine machine NOUN brj-24033 256 14 intelligence intelligence NOUN brj-24033 256 15 35(7 35(7 NUM brj-24033 256 16 ) ) PUNCT brj-24033 256 17 , , PUNCT brj-24033 256 18 17571772 17571772 NUM brj-24033 256 19 . . PUNCT brj-24033 257 1 doi doi NOUN brj-24033 257 2 : : PUNCT brj-24033 257 3 10.1109/ 10.1109/ NUM brj-24033 257 4 tpami.2012.256 tpami.2012.256 PROPN brj-24033 257 5 tax tax NOUN brj-24033 257 6 , , PUNCT brj-24033 257 7 d. d. PROPN brj-24033 257 8 m. m. PROPN brj-24033 257 9 j. j. PROPN brj-24033 257 10 , , PUNCT brj-24033 257 11 and and CCONJ brj-24033 257 12 duin duin PROPN brj-24033 257 13 , , PUNCT brj-24033 257 14 r. r. PROPN brj-24033 257 15 p. p. PROPN brj-24033 257 16 w. w. PROPN brj-24033 257 17 ( ( PUNCT brj-24033 257 18 2004 2004 NUM brj-24033 257 19 ) ) PUNCT brj-24033 257 20 . . PUNCT brj-24033 258 1 β€œ " PUNCT brj-24033 258 2 support support VERB brj-24033 258 3 vector vector NOUN brj-24033 258 4 data datum NOUN brj-24033 258 5 description description NOUN brj-24033 258 6 , , PUNCT brj-24033 258 7 ” " PUNCT brj-24033 258 8 machine machine NOUN brj-24033 258 9 learning learn VERB brj-24033 258 10 54(1 54(1 NUM brj-24033 258 11 ) ) PUNCT brj-24033 258 12 , , PUNCT brj-24033 258 13 45 45 NUM brj-24033 258 14 - - SYM brj-24033 258 15 66 66 NUM brj-24033 258 16 . . PUNCT brj-24033 258 17 tuncer tuncer NOUN brj-24033 258 18 , , PUNCT brj-24033 258 19 f. f. PROPN brj-24033 258 20 d. d. PROPN brj-24033 258 21 , , PUNCT brj-24033 258 22 dogu dogu PROPN brj-24033 258 23 , , PUNCT brj-24033 258 24 d. d. PROPN brj-24033 258 25 , , PUNCT brj-24033 258 26 and and CCONJ brj-24033 258 27 akdeniz akdeniz NOUN brj-24033 258 28 , , PUNCT brj-24033 258 29 e. e. PROPN brj-24033 258 30 ( ( PUNCT brj-24033 258 31 2021 2021 NUM brj-24033 258 32 ) ) PUNCT brj-24033 258 33 . . PUNCT brj-24033 259 1 β€œ " PUNCT brj-24033 259 2 efficiency efficiency NOUN brj-24033 259 3 of of ADP brj-24033 259 4 preprocessing preprocesse VERB brj-24033 259 5 methods method NOUN brj-24033 259 6 for for ADP brj-24033 259 7 discrimination discrimination NOUN brj-24033 259 8 of of ADP brj-24033 259 9 anatomically anatomically ADV brj-24033 259 10 similar similar ADJ brj-24033 259 11 pine pine ADJ brj-24033 259 12 species specie NOUN brj-24033 259 13 by by ADP brj-24033 259 14 nir nir ADJ brj-24033 259 15 spectroscopy spectroscopy NOUN brj-24033 259 16 , , PUNCT brj-24033 259 17 ” " PUNCT brj-24033 259 18 wood wood NOUN brj-24033 259 19 material material NOUN brj-24033 259 20 science science NOUN brj-24033 259 21 & & CCONJ brj-24033 259 22 engineering engineering PROPN brj-24033 259 23 1 1 NUM brj-24033 259 24 - - SYM brj-24033 259 25 10 10 NUM brj-24033 259 26 . . PUNCT brj-24033 260 1 doi doi NOUN brj-24033 260 2 : : PUNCT brj-24033 260 3 10.1080/17480272.2021.2012821 10.1080/17480272.2021.2012821 NUM brj-24033 260 4 yu yu PROPN brj-24033 260 5 , , PUNCT brj-24033 260 6 z. z. PROPN brj-24033 260 7 , , PUNCT brj-24033 260 8 qin qin PROPN brj-24033 260 9 , , PUNCT brj-24033 260 10 l. l. PROPN brj-24033 260 11 , , PUNCT brj-24033 260 12 chen chen PROPN brj-24033 260 13 , , PUNCT brj-24033 260 14 y. y. PROPN brj-24033 260 15 , , PUNCT brj-24033 260 16 and and CCONJ brj-24033 260 17 parmar parmar PROPN brj-24033 260 18 , , PUNCT brj-24033 260 19 m. m. PROPN brj-24033 260 20 d. d. PROPN brj-24033 260 21 ( ( PUNCT brj-24033 260 22 2020 2020 NUM brj-24033 260 23 ) ) PUNCT brj-24033 260 24 . . PUNCT brj-24033 261 1 β€œ " PUNCT brj-24033 261 2 stock stock NOUN brj-24033 261 3 price price NOUN brj-24033 261 4 forecasting forecasting NOUN brj-24033 261 5 based base VERB brj-24033 261 6 on on ADP brj-24033 261 7 lle lle PROPN brj-24033 261 8 - - PUNCT brj-24033 261 9 bp bp PROPN brj-24033 261 10 neural neural ADJ brj-24033 261 11 network network NOUN brj-24033 261 12 model model NOUN brj-24033 261 13 , , PUNCT brj-24033 261 14 ” " PUNCT brj-24033 261 15 physica physica NOUN brj-24033 261 16 a a DET brj-24033 261 17 : : PUNCT brj-24033 261 18 statistical statistical ADJ brj-24033 261 19 mechanics mechanic NOUN brj-24033 261 20 and and CCONJ brj-24033 261 21 its its PRON brj-24033 261 22 applications application NOUN brj-24033 261 23 553 553 NUM brj-24033 261 24 , , PUNCT brj-24033 261 25 124197 124197 NUM brj-24033 261 26 . . PUNCT brj-24033 262 1 doi doi NOUN brj-24033 262 2 : : PUNCT brj-24033 262 3 10.1016 10.1016 NUM brj-24033 262 4 / / SYM brj-24033 262 5 j.physa.2020.124197 j.physa.2020.124197 PROPN brj-24033 262 6 zhan zhan PROPN brj-24033 262 7 , , PUNCT brj-24033 262 8 w. w. PROPN brj-24033 262 9 , , PUNCT brj-24033 262 10 chen chen PROPN brj-24033 262 11 , , PUNCT brj-24033 262 12 b. b. PROPN brj-24033 262 13 , , PUNCT brj-24033 262 14 wu wu PROPN brj-24033 262 15 , , PUNCT brj-24033 262 16 x. x. PROPN brj-24033 262 17 , , PUNCT brj-24033 262 18 yang yang PROPN brj-24033 262 19 , , PUNCT brj-24033 262 20 z. z. PROPN brj-24033 262 21 , , PUNCT brj-24033 262 22 lin lin PROPN brj-24033 262 23 , , PUNCT brj-24033 262 24 c. c. PROPN brj-24033 262 25 , , PUNCT brj-24033 262 26 lin lin PROPN brj-24033 262 27 , , PUNCT brj-24033 262 28 j. j. PROPN brj-24033 262 29 , , PUNCT brj-24033 262 30 and and CCONJ brj-24033 262 31 guan guan PROPN brj-24033 262 32 , , PUNCT brj-24033 262 33 x. x. NOUN brj-24033 262 34 ( ( PUNCT brj-24033 262 35 2023 2023 NUM brj-24033 262 36 ) ) PUNCT brj-24033 262 37 . . PUNCT brj-24033 263 1 β€œ " PUNCT brj-24033 263 2 wood wood NOUN brj-24033 263 3 identification identification NOUN brj-24033 263 4 of of ADP brj-24033 263 5 cyclobalanopsis cyclobalanopsis NOUN brj-24033 263 6 ( ( PUNCT brj-24033 263 7 endl endl NOUN brj-24033 263 8 . . PUNCT brj-24033 263 9 ) ) PUNCT brj-24033 264 1 oerst oerst NOUN brj-24033 264 2 based base VERB brj-24033 264 3 on on ADP brj-24033 264 4 microscopic microscopic ADJ brj-24033 264 5 features feature NOUN brj-24033 264 6 and and CCONJ brj-24033 264 7 ctgan ctgan VERB brj-24033 264 8 - - PUNCT brj-24033 264 9 enhanced enhance VERB brj-24033 264 10 explainable explainable ADJ brj-24033 264 11 machine machine NOUN brj-24033 264 12 learning learning NOUN brj-24033 264 13 models model NOUN brj-24033 264 14 , , PUNCT brj-24033 264 15 ” " PUNCT brj-24033 264 16 frontiers frontier NOUN brj-24033 264 17 in in ADP brj-24033 264 18 plant plant NOUN brj-24033 264 19 science science NOUN brj-24033 264 20 . . PUNCT brj-24033 265 1 14 14 NUM brj-24033 265 2 , , PUNCT brj-24033 265 3 article article NOUN brj-24033 265 4 1203836 1203836 NUM brj-24033 265 5 . . PUNCT brj-24033 266 1 doi:10.3389 doi:10.3389 PROPN brj-24033 266 2 / / SYM brj-24033 266 3 fpls.2023.1203836 fpls.2023.1203836 PROPN brj-24033 266 4 zhang zhang PROPN brj-24033 266 5 , , PUNCT brj-24033 266 6 h. h. PROPN brj-24033 266 7 , , PUNCT brj-24033 266 8 and and CCONJ brj-24033 266 9 patel patel PROPN brj-24033 266 10 , , PUNCT brj-24033 266 11 v. v. ADP brj-24033 266 12 m. m. NOUN brj-24033 266 13 ( ( PUNCT brj-24033 266 14 2017 2017 NUM brj-24033 266 15 ) ) PUNCT brj-24033 266 16 . . PUNCT brj-24033 267 1 β€œ " PUNCT brj-24033 267 2 sparse sparse ADJ brj-24033 267 3 representation representation NOUN brj-24033 267 4 - - PUNCT brj-24033 267 5 based base VERB brj-24033 267 6 open open ADJ brj-24033 267 7 set set NOUN brj-24033 267 8 recognition recognition NOUN brj-24033 267 9 , , PUNCT brj-24033 267 10 ” " PUNCT brj-24033 267 11 ieee ieee NOUN brj-24033 267 12 trans trans PROPN brj-24033 267 13 pattern pattern NOUN brj-24033 267 14 analysis analysis NOUN brj-24033 267 15 and and CCONJ brj-24033 267 16 machine machine NOUN brj-24033 267 17 intelligence intelligence NOUN brj-24033 267 18 39(8 39(8 NUM brj-24033 267 19 ) ) PUNCT brj-24033 267 20 , , PUNCT brj-24033 267 21 1690 1690 NUM brj-24033 267 22 - - SYM brj-24033 267 23 1696 1696 NUM brj-24033 267 24 . . PUNCT brj-24033 268 1 doi doi NOUN brj-24033 268 2 : : PUNCT brj-24033 268 3 10.1109 10.1109 NUM brj-24033 268 4 / / SYM brj-24033 268 5 tpami.2016.2613924 tpami.2016.2613924 NOUN brj-24033 268 6 article article NOUN brj-24033 268 7 submitted submit VERB brj-24033 268 8 : : PUNCT brj-24033 268 9 september september PROPN brj-24033 268 10 30 30 NUM brj-24033 268 11 , , PUNCT brj-24033 268 12 2024 2024 NUM brj-24033 268 13 ; ; PUNCT brj-24033 268 14 peer peer NOUN brj-24033 268 15 review review NOUN brj-24033 268 16 completed complete VERB brj-24033 268 17 : : PUNCT brj-24033 268 18 october october PROPN brj-24033 268 19 26 26 NUM brj-24033 268 20 , , PUNCT brj-24033 268 21 2024 2024 NUM brj-24033 268 22 ; ; PUNCT brj-24033 268 23 revised revise VERB brj-24033 268 24 version version NOUN brj-24033 268 25 received receive VERB brj-24033 268 26 and and CCONJ brj-24033 268 27 accepted accept VERB brj-24033 268 28 : : PUNCT brj-24033 268 29 november november PROPN brj-24033 268 30 1 1 NUM brj-24033 268 31 , , PUNCT brj-24033 268 32 2024 2024 NUM brj-24033 268 33 ; ; PUNCT brj-24033 268 34 published publish VERB brj-24033 268 35 : : PUNCT brj-24033 268 36 november november PROPN brj-24033 268 37 27 27 NUM brj-24033 268 38 , , PUNCT brj-24033 268 39 2024 2024 NUM brj-24033 268 40 . . PUNCT brj-24033 269 1 doi doi NOUN brj-24033 269 2 : : PUNCT brj-24033 269 3 10.15376 10.15376 NUM brj-24033 269 4 / / SYM brj-24033 269 5 biores.20.1.944 biores.20.1.944 NOUN brj-24033 269 6 - - SYM brj-24033 269 7 955 955 NUM