Microsoft Word - CET--006.docx


 CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 59, 2017 

A publication of 

 

The Italian Association 
of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Zhuo Yang, Junjie Ba, Jing Pan 
Copyright © 2017, AIDIC Servizi S.r.l. 
ISBN 978-88-95608- 49-5; ISSN 2283-9216 

Dark Adaptation Time Study on Road Tunnel Daytime 

Lighting Based on Visual Performance Method 

Ji Wenga, Feng Dua,b*, Yingkui Hua, Xianyun Caia 

a Chongqing University, Chongqing 400045, China 
b Fujian University of Technology, Fuzhou 351018,  China 

fenghui117@163.com 

Visual adaptation in road tunnel daytime driving is very important to driving safety and should be paid full 

attention. Visual adaptation consists of light-adaptation and dark-adaptation, daytime threshold zone dark-

adaptation problem when driving from outside into the tunnel is more important in tunnel lighting, current 

studies have described significantly distinct in the length of adaptation time. This study simulates the daytime 

lighting environment in the road tunnel threshold zone and records the reaction time and pupil change to many 

test subjects according to visual performance method in the condition of the luminance from light condition 

(ambient exterior luminance 4000 cd·m-2) and to dark condition (interior zone luminance 2cd·m-2), through the 

data analysis the length of dark adaptation time is obtained (19.8s). Study on length of effective dark 

adaptation time can provide length basis for road tunnel threshold zone daytime lighting design. 

1. Introduction 

Visual adaptation in road tunnel daytime driving is very important to driving safety and the accidents happened 

in tunnels emerge in endlessly (Yeung and Wong, 2014; Kircher and Ahlstrom, 2012), so it should be paid full 

attention. The visual adaptation consists of light-adaptation and dark-adaptation, driving outside the tunnel in 

the daytime corresponds to light-adaptation and the time is relatively short while driving into the tunnel in the 

daytime corresponds to dark-adaptation and the time is relatively long. The degree of adaptation is closely 

related to driving safety. The main objective of road tunnel lighting with many changes in light and dark is to 

diminish dark AT (adaptation time) and through installing transitional lighting to meet the demand of the visual 

adaptation (Zhou et al., 2016). 

The length of light and dark AT is related to the ratio of luminance change, absolute value (Blaser and Dudli, 

1993) and individual physiological factors. Generally speaking, the light-adaptation time is about several 

seconds to several minutes while the dark AT is more than ten seconds or half an hour. Kabayama (1967) 

proposed that one needs about ten seconds to restore eyesight in dark-adaptation but about 1 second in light-

adaptation. Zhang and Zhang (1994) did the investigation on vehicle drivers’ dark-adaptation, concluded the 

overall mean dark AT is 34.17±14.25 seconds. Xie Xiuying believed that one can restore ordinary visual work 

in about 1 minute when entering the light environment from the dark. Only after 3~5 minutes the outline of the 

surroundings can be differentiated when entering the dark from the light. Du and Huang, (2013) evaluated the 

visual burden of road tunnel through pupil size changing speed, pointed out that in the middle and long tunnels, 

the dark AT is less than 23 seconds and the light-adaptation time is less than 13 seconds. A. Pena-Gacia 

(2012) pointed out that one needs at least 8 minutes to adapt to the dark environment from the light. Huang 

Yan believed that one needs about 30 to 40 minutes to gain the complete dark-adaptation. This time is a bit 

too long and is meaningless to tunnel lighting design. The stipulations above have great differences, the 

reasons are on one side the adaptation region is different (different luminance change), on the other side the 

research method is also different. Different industries have studied the AT aimed at their specialty, but lack of 

the pertinence of road tunnel lighting. The thesis defines the dark adaptation of achieving road tunnel driving 

visual acknowledgement and ensure driving safety as road tunnel effective dark-adaptation, the dark AT under 

this condition has more research and applicable value to road tunnel driving safety. 

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1759116

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Ji Weng, Feng Du, Yingkui Hu, Xianyun Cai, 2017, Dark adaptation time study on road tunnel daytime lighting 
based on visual performance method, Chemical Engineering Transactions, 59, 691-696  DOI:10.3303/CET1759116   

691

mailto:fenghui117@163.com


Visual performance method have been applied in road tunnel lighting research. Nakamichi and Narisada, 

(1967), and Kabayama (1967) proposed visual performance method, whose experimental contents are just 

measuring the RT (reaction time) data of visual performance. Zhang and Hu, (2017) analyzed the 

comprehensive efficiency of road tunnel lighting based on visual performance method. Zhang and Yang, 

(2009), Zhang and Li, (2013), and Li (2009) studied the influence to road lighting in virtue of RT. Du et al., 

(2007), and Hu et al., (2016) studied the road tunnel lighting in the means of pupil change. 

2. Research method 

The thesis adopts visual performance method, simulates and reappears road tunnel driving optical 

circumstance in the lab, finishes the experiment on pupil change and RT. The pupil change is related with 

safety and comfort while the RT is coherent to driving safety. The test are aimed at 13 test subjects, seven 

males (occupy 54%) and six females (occupy 46%), aged between 20 and 30. The naked eye vision are 

meanly above 0.7 (the decimal eye chart), all without color blindness and weakness. 

2.1 Experiments on pupil change 

(1) Experimental instruments: Road tunnel visual performance test instrument, road tunnel ambient exterior 

luminance (L20) simulation light box, interior zone luminance (Lin) simulation light box, LED light controlling box, 

BM-5A luminance meter, iView-X eye tracker, CS-2000 spectral radiation luminance meter etc. The plan and 

elevation of the experiment are shown in Figure 1 and the experimental instruments in Figure 2. 

 
(a) Plan (b) Elevation 

Figure 1: Plan and elevation of the road tunnel visual performance experimental instrument (pupil change) 

 

Figure 2: Experimental instruments 

(2) Experimental parameters 

The luminance of each lighting zone are based on CIE88-2004, the luminance between ambient exterior and 

interior zone is smoothly changed. The value of L20 is typical value of 4000 cd·m
-2, and the Lin is 2 cd·m

-2. The 

final purpose is to find how long the time will be to adapt from the L20 to Lin. The time difference is just the time 

utilized in the strengthened entrance lighting zone. The study adopts same light source, which is LED that with 

high luminous efficiency, good light color and good adjustability, The CCT (abbr. for correlated color 

temperature) of the LED lights is 5000K and the CRI (abbr. for color rendering index) is 75.3. The SPD (abbr. 

for spectral power distribution) of the light is shown in Figure 3.  

692



 

Figure 3: SPD of the experimental LED lights 

(3) Experimental procedure 

The test subjects take part in the experiment in good manner. Firstly, the test subject concentrates on the 

exterior environment, after full adaptation the subject watched the L20 simulation light box (with the luminance 

of 4000 cd·m-2); Secondly, after full adaptation, shut off the L20 simulation light box and at the same time turn 

on the Lin simulation light box, the sighting target (with the size 0.01m×0.01m) will appear in three different 

position (-10, 0, 10) randomly meanwhile. The test subject searches and verifies the sighting target in the 

Lin. Finally, the eye tracker records the data of pupil change and scenario change. 

2.2 Experiments on RT 

(1) The test subjects are the same to the experiment of pupil change. The experimental instruments are also 

similar to last experiment, but without eye tracker. The plan and elevation of the experimental instruments are 

shown in Figure 4. 

(2) Experimental procedure 

The test subjects take part in the experiment in good manner. The first procedure is just like the last 

experiment; Secondly, after full adaptation, turn on the Lin simulation light box and the subject waits for 

different AT (5s, 10s, 15s, 20s, 25s, 30s, 35s, 40s, 45s, 50s), then the sighting target (with the size as above) 

will appear in three different position (the angle as above) randomly meanwhile the RT instrument begins to 

count the time. The test subject searches and verifies the sighting target in the Lin. When the subject finds the 

sighting target, the RT button will be pressed. Finally, the RT instruments record the RT (between releasing 

the sighting target and pressing the button). 

 

 

 

 

 

 

(a) Plan (b) Elevation 

Figure 4: Plan and elevation of the road tunnel visual performance experimental instrument (RT) 

3. Research result 

The road tunnel effective dark AT are verified by synthesizing the data of pupil change and RT. The 

distinguish basis for pupil change is the trend of pupil size. When the trend is stable which means adaptive. 

The distinguish basis for RT is the principle of shorter RT reflects more adaptive. 

The experiments are tested to 13 test subjects. Finally, the results are as follows. 

3.1 Data of pupil change 

There are many data through iView-X eye tracker. A frame occupy 50ms and there are 20 frames in one 

second. The typical data are shown in Figure 5~7. 

In Figure 5, the horizontal axis means time with the unit frame, while vertical axis means the diameter of pupil 

with the unit mm (millimeter), the blue data stand for X-directional pupil size and the red data stand for the Y-

directional pupil size. The law of the two are quite similar. For analyzing the data more conveniently, 

smoothing the data of Y-directional pupil size, the results are shown in Figure 6.  

693



  

Figure 5: Pupil change data of a test subject 

 

Figure 6: Pupil change data of a test subject after smoothing 

From Figure 5~6, the pupil size is changing rapidly with the different adaptation. When human eyes adapt to 

some extent, that is, when the lights become weak that leads to the pupil size increasing to some extent, the 

data shows some law. When the luminance from 4000 cd·m-2 to 2 cd·m-2, the pupil size gradually enlarged 

from 801 frames (with the Y-directional diameter 3.2726mm) to 1201 frame (with the Y-directional diameter 

5.8609mm). It can be seen that the total time is 400 frames (20.0 seconds) to the overall adaptation procedure 

and accomplished the adaptation task. In order to see the procedure more clearly part of the data is enlarged 

to Figure 6. 

 

Figure 7: Pupil change data of a test subject after cutting out 

From Figure 7 the procedure can be seen more clearly. From 4000 cd·m-2 to 2 cd·m-2, the pupil size gradually 

enlarged from 159 frame (with the Y-directional diameter 3.2726mm) to 559 frame (with the Y-directional 

diameter 5.8609mm). The total adaptation procedure cost 20.0 seconds. 

In the same way 8 groups of effective data are achieved through trimming. The other data lack of some useful 

data or the instruments meet some problems. The useful data are shown in table 1. 

Table 1: Dark AT of the test subjects from the pupil change 

Test subjects (1) (2) (3) (4) (5) (6) (7) (8) 

gender male male female female female female male male 
age (a) 24 21 21 30 26 30 22 26 
Dark AT (s) 19.9 20.2 22.1 19.2 18.7 18.8 19.1 18.8 

 

The statistical results to the data of effective AT are shown in table 2. 

Table 2: Effective dark AT statistics 

Mean value (s) 
Standard deviation 
(s) 

Observation data 
Confidence coefficient 
(95%) 

19.6 1.16 8 0.97 

694



The statistics shows that the length of effective mean dark AT is 19.6 seconds. 

3.2 RT data 

The shorter RT means safer driving behavior and the suitable AT is corresponding to this shortest RT. So the 

shortest RT is the basis here. 

Select the data of a test subject and listed in table 3. Here the data of RT and AT are fitted and the result 

shows with the increase of AT the RT firstly decrease and then increase. The fitting result shows they 

conforms best to quadratic, the R2 (multiple correlative coefficient) is 0.7067. Shown in Figure 8. 

Table 3: The AT and RT of a test subject 

AT (s) 5 10 15 20 25 30 40 

RT (ms) 458 443 432 302 347 364 457 

 

Figure 8: The relationship between RT and AT of the subject 

From Figure 8, the shortest RT of this subject is 302ms, and the corresponding AT is 20s. Eleven groups of 

data among thirteen subjects are effective. After analyzing the data, the results are shown in table 4. 

Table 4: The effective dark AT data corresponding to the RT of all subjects 

Test subjects (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) 

Shortest RT(ms) 384 410 293 275 416 361 309 286 302 375 481 
Effective dark AT(s) 25 10 25 15 30 35 25 30 20 40 15 

 

Through the analysis, the effective dark-adaptation (some of the data in table 4 are the same) time and 

shortest RT data are shown in table 5. 

Table 5: The data of effective dark AT and shortest RT of all subjects 

Effective dark AT(s) 10 15 20 25 30 35 40 

Shortest RT(ms) 420 386 295 310 350 360 375 

 

Fit the relationship of dark AT and shortest RT, the relationship conforms to quadratic the best, the R2 is 

0.7601. Shown in Figure 9. When the shortest RT is 20s the effective AT is 20.0s. 

 

Figure 9: Relationship between effective AT and shortest RT 

4. Conclusion 

The thesis introduced visual performance method, utilized the parameters of pupil change and RT, and 

finished the experiment on dark AT. Synthesize the two results, the conclusions are drawn. 

(1) Although complete adaptation is important, for the road tunnel driving, the AT in complete adaptation is too 

long and lack of actual meaning. Through the current studies, human eyes can even accomplish visual work 

695



while driving into the tunnel without complete adaptation. This thesis defines this time as effective dark AT, 

which is the start point of this study. 

(2) The study adopted visual performance, the parameters are respectively pupil change and RT. Through the 

two parameters the mean effective dark AT is 19.8s when the luminance is typically from (L20) 4000 cd·m
-2 to 

(Lin) 2 cd·m
-2. 

(3) According to the principle of effective dark AT corresponds to tunnel travel length, the length of the 

strengthened entrance lighting zone is 440m. Through the analysis to CIE and the specifications of many 

countries, the conclusion is that in order to achieve driving safety utmost, the length of CIE and the stipulations 

of other countries should be prolonged. 

(4) The constitution of eyes and visual fatigue caused the great difference to the current study and the 

commonly thinking. 

(5) This study aimed at the luminance change from ambient exterior to interior zone, the luminance values are 

typical 4000 cd·m-2 to 2 cd·m-2 and the travel speed is 80km·h-1. The future study will aim at more luminance 

change and travel speed, to perfect and replenish the effective dark AT research. 

Acknowledgements 

The authors wish to thank for the support of NSFC (51278507) and (CYB16011) to give us the chance to 

continue the study. 

Reference  

Blaser P., Dudli H., 1993, Tunnel lighting: Method of calculating luminance of access zone L20, Lighting 

Research and Technology, 25(1), 25-30.  

Du Z., Huang F., 2013, Light and dark adaptation time based on pupil area variation at entrance and exit 

areas of highway tunnel, Journal of highway and transportation research and development, 8 (1), 73-77, 

DOI: 10.1061/JHTRCQ.0000366. 

Du Z., Pan X., Guo X., 2007, Experimental studies of visual adaptation on driving through freeway tunnel’s 

entrance and exit, Journal of Harbin Institute of Technology, 39(12), 1998-2001. 

Hu Y., Chen Z., Zhang Q., Weng J., Huang K., Lin Y., 2016, Tunnel threshold zone luminance determined 

method based on drivers’ visual adaptation, Journal of Civil, Architectural & Environmental Engineering, 

38(4), 20-26. 

JTJ 026.1-1999: Specification for design of ventilation and lighting of highway tunnel. Beijing: Ministry of 

communications of PRC, 1999. 

Kabayama H., 1967, Study on adaptive illumination for sudden change of brightness, Journal of the 

Illuminating Engineering Institute of Japan, 47(10), 9-12, DOI: 10.2150/jieij1917.47.10_488. 

Kircher K., Ahlstrom C., 2012, The impact of tunnel design and lighting on the performance of attentive and 

visually distracted drivers, Accident Analysis and Prevention. 47, 153–161, DOI: 

10.1016/j.aap.2012.01.019. 

Li Y., 2009, Study on the influence of road lighting to traffic safety based on photo-biological effect; Chong 

Qing: Chongqing University. 

Nakamichi F., Narisada K., 1967, Experiment on the Visibility of the tunnel lighting, Journal of the Illuminating 

Engineering Institute of Japan, 47(9), 4-12, DOI: 10.2150/jieij1917.51.10_566. 

Peña-García A., 2012, Computational optimization of semi-transparent tension structures for the use of solar 

light in road tunnels, Tunneling and Underground Space Technology, 32, 127–131, DOI: 

10.1016/j.tust.2012.06.004. 

Yeung J.S., Wong Y.D., 2014, The effect of road tunnel environment on car following behaviour, Accident 

Analysis and Prevention, 70, 100-109, DOI: 10.1016/j.aap.2014.03.014. 

Zhang J., Zhang X., 1994, Investigation and study of automobile drivers’ dark adaptation function, Chinese 

Journal of Industrial Medicine, 2(7), 99-100. 

Zhang Q., Li Y., 2013, The applicability of different color temperature LED light sources in road lighting, China 

Illuminating Engineering Journal, 24(5), 70-77. 

Zhang Q., Yang C., 2009, A research of new light source for road lighting based on vision function method, 

Journal of Tongji University (natural science, 37(6), 781-785. 

Zhang X., Hu J., 2017, The comprehensive efficiency analysis of tunnel lighting base on visual performance, 

Advances in Mechanical Engineering. 9(4), 1-9. 

Zhou Y., Wang S.L., Wang J., 2016, Vibration response of an existing tunnel to adjacent blasting construction, 

Chemical Engineering Transactions, 55, 31-36, DOI: 10.3303/CET1655006. 

696