45 8 207 8 Journal of South China University of Technology Natural Science Edition Vol 45 No 8 August 207 000-565X20708-0050-07 2 3002202 54004 PERCLOS U49 doi0 3969 /j issn 000-565X 207 08 008 4-7 0 h Stroop 3 Khamis 0 h Liu 2 3 206-09-27 * 4400200367 530825 203M54306 Foundation itemssupported by the Jilin Province Changbai Mountain Scholar Program4400200367 National Natural Science Foundation of China Youth Fund Project530825and China Postdoctoral Science Foundation203M54306 97- E-maillshiwu@ 63 com 984- E-mailwanghonglin0520 @ 26 com
8 5 2 2 Bus hound Fig Test platform 2 2 0 5 5 20 ~ 40 Smart-Eye Pro 5 7 4 LED Bus hound Bus hound5 0 J 20 ~ 40 2 2 2 3 LED 4 2 LED 8-0 LED LED ~ 3 min s Table Information of the driver / / A 20 ~ 40 3 B 20 ~ 30 3 C 30 ~ 40 5 D 20 ~ 40 4 E 20 ~ 40 5 F 30 ~ 40 3 G 20 ~ 40 2 H 20 ~ 40 3 I 20 ~ 40
52 45 2 4 2 LED 3 25 2 Fig 2 Simplified model for determining driving fatigue threshold LED 2 PERCLOS 2 PERCLOS 50 min PERCLOS -2 Percentage of Eyelid Closure O- 0 ~ 20 min 2 ver the Pupil Over Time 80 min LED Bus Hound PERCLOS P70 P80 EM 3 9 600 P70 Bps Smart Eye 70% 4 P80 80% EM 60 Hz 3 PERCLOS 70% LED PERCLOS 2 3 min LED f = t 3 - t 2 t 4 - t Perclos f PERCLOS t 70% t 2 70% 30% t 3 30% 30% t 4 30% 70% 2 2 2 3-4 X Y Z 3 LED 4Bus hound 5 F n = - LED 2
8 53 Z i = - L i = W x * j + W 2 x * 2j + W 3 x * 3j PERCLOS L i 3 2 3 2 3 8 2 2 PERCLOS f = b + 2d f b d 5-7 2 2 X i = x i x i2 x in i = 2 3 n e i e - Y Y 2 Y 3 Y i = y i y i2 y in i = 2 3 2 3 2 x ij - minx ij j y ij = i = x ij - minx ij 2 3j = 2 n max j j 2 E j = - lnn - n p ij lnp ij p ij = Y ij n Y ij i = i = p ij = 0 limp ij lnp ij = 0 3 p ij 0 3 2 3 min PERCLOS E E 2 E 3 W i - E i = i = 2 3 3 - E i 2 2 2 min-max PERCLOS 0 ~ Fig 3 Weighted average of the test 300 ~ 500 ms 0-2 x * ij = x ij - min max - min max min 2 2 3 b 3 3 PERCLOS X X 2 X d = e 3 i - e - 2 槡 n n i = 3 Table 2 2 Threshold value of the test f f 2 f /min f 2 /min 0 22 50 0 350 264
54 45 2 L i = 0 47x * j = 0 4008x * 2j + 0 764x * 3j 44 min 0 239 3 79 50 min 0 239 LED 32 264 min 0 346 2 LED 220 min 0 346 2 2 2 2 0 ~ 0 2 0 ~ 0 2 PERCLOS 3 2 3 3 3 Table 2 2 Threshold values of several tests f f 2 /min /min /min f 0 22 50 0 350 264 4 2 0 232 50 0 349 255 05 3 0 236 40 0 358 245 05 4 0 247 45 0 336 230 85 5 0 258 50 0 348 232 82 6 0 238 30 0 337 95 65 0 239 44 0 346 220 76 8 8 f 2 0 ~ 20 min 4 PERCLOS PERCLOS 2 3 2 44 min 220 min 2
8 55 8 J 20 3903 95-00 06 XU Jian-min SHOU Yan-fang LU Kai Adaptive signal control model based on vehicle emission J Journal of South China University of Technology Natural Science E- dition20 3903 95-00 06 KHAMIS N DEROS B NUAWI M et al Driving fatigue among long distance heavy vehicle drivers in klang valleymalaysia J Applied Mechanics and Materials 204 2 567-573 2LIU Chang SI Zhong-chen PENG Peng-pei et al Detecting driving fatigue of indirect vision driving based on electroencephalogram J Advanced Materials Research 203 765-767228-233 48 02-06 ZHAO Xiao-hua XU Shi-li RONG Xian et al Discriminating Threshold of driving fatigue based on the electroencephalography sample entropy by receiver operating characteristic curve analysis J Journal of Southwest Jiao Tong University 203 48 02-06 2002 72 04-09 4 ERP ZHENG PeiSONG Zheng-heZHOU Yi-ming PER- J 2009 54-4 CLOS-Based recognition algorithms of motor driver fa- SONG Guo-ping ZHANG Kan The ERP study that driving fatigue impact on voluntary attention J Psychological Science 2009 54-4 5 J D 2005 2009 8 7-0 SONG Guo-ping ZHANG Kan Effects of 0h driving fatigue on executive function J Ergonomics 2009 8 7-0 SONG Guo-ping ZHANG Kan Effects of driving fatigue on visual involuntary attentionan ERPs study J Ergonomics 2009 52-4 7 ERP LIN Wei-wei KNN query technology of mobile terminals J 200 335 067-069 SONG Guo-ping ZHANG Kan The ERP study that driving fatigue impact on visual attention J Psychological Science 200 335 067-069 9 J 20 390-8 BIE Yi-mingWANG Dian-hai ZHAO Ying-ying Multiple-phase bus signal priority strategy for arterial coordination intersection J Journal of South China University of Technology Natural Science Edition 20 39 0-8 0 MFD J 3 ROM 202 40 38-46 J 203 ZHU Lin YU Lei SONG Guo-hua MFD-based investigation into macroscopic traffic status of urban networks and its influencing factors J Journal of South China University of Technology Natural Science Edition 202 40 38-46 PERCLOS J tigue J Journal of China Agricultural University 2002 72 04-09 2 3 D 2006 4 J 2003 54 0-3 6 WANG Bin Study status of reaction time and influnce ERP J 2009 52-4 factors J Journal of Capital College of Physical Education 2003 54 0-3 5 Hadoop J 202 40 52-58 in highway network J Journal of South China University of Technology Natural Science Edition 202 40 52-58 6
56 45 J 202 40-2 YANG Jun-mei YU HuaWEI Gang Independent component analysis and its application to signal processing J Journal of South China University of Technology Natural Science Edition202 40-2 7 DANG Wen-hui ZHANG De-jiang SUN Ru-jiang Improved threshold selection method in the application J of 20 393 0-06 LI Shi-wu WANG Lin-hong GUO Dong et al Adaptive signal control model based on vehicle emission J Journal of South China University of TechnologyNatural Science Edition20 393 0-06 8 J 202 0 4-3 the network monitoring J Science & Technology Vision 202 04-3 Driving Fatigue Quantization Based on Entropy Weight Method LI Shi-wu YIN Yan-na 2 WANG Lin-hong XU Yi School of TransportationJilin UniversityChangchun 30022JilinChina2 College of Automobile and Transportation EngineeringGuilin University of Aerospace TechnologyGuilin 54004GuangxiChina AbstractIn order to obtain the judging threshold of the driving fatigue objectively and accuratelythe driver's eye movement datareaction time and execution time under the sober and mental fatigue states are collected from a driving simulator The driver's eye movement datareaction time and execution time during continuous driving are used to respectively describe the driver's gaze characteristicsreaction ability and executive ability Thena relationship model is constructed by adopting the entropy weight method to obtain the weighted average of the three kinds of data In the modelthe three kinds of data and the quantitative values of the driving fatigue are respectively taken as the independent and dependent variables In order to improve the objectivity of the driving fatigue thresholdthe variable threshold determination method is adopted to select the thresholds of the optimal fatigue at the first fatigue threshold and the secondary fatigue threshold Moreoverthe change rules of the driver fatigue under the continuous driving condition are revealedand the quantitative method of the driving fatigue is evaluated Experimental results show that the proposed method can effectively improve the precision of the driving fatigue quantizationand it has a broad application prospect in the driving safety field based on the prevention of driving fatigue Key wordsautomobile driverspercentage of eyelid closure over the pupil over timereaction abilityexecutive abilityfatigue quantization