1. Xinjiang Technical Institute of Physics & Chemistry, Chinese Academy of Science, Urumq Xinjiang 830011, China;
2. School of Civil Engineering and Transportation, South China University of Technology, Guangzhou Guangdong 510640, China;
3. School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin Heilongjiang 150090, China
A Pedestrian Detection Method Based on Hirarchical Tree Cascade Classsification at Nighttime
ZHANG Rong-hui1, ZHOU Jia-li2, YOU Feng2, ZHOU Xi1, PEI Yu-long3
1. Xinjiang Technical Institute of Physics & Chemistry, Chinese Academy of Science, Urumq Xinjiang 830011, China;
2. School of Civil Engineering and Transportation, South China University of Technology, Guangzhou Guangdong 510640, China;
3. School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin Heilongjiang 150090, China
摘要Illumination for pedestrian detection at nighttime is weak, and detection is easily affected through variations in illumination. Thus, a bicharacteristic method of pedestrian detection at nighttime based on hierarchical tree cascade classification is presented according to "coarse-to-fine" principle. The proposed method consists of two stages of cascade classifiers. Coarse cascade classifiers are constructed in complete binary tree architecture. These classifiers use Haar-like features for the rapid identification of candidate pedestrian areas. By contrast, fine cascade classifiers have a parallel structure. Edge let features are used for detection along three parts: the head-shoulder, trunk, and leg parts of candidate pedestrian areas. Bayesian decision-making is adopted to achieve pedestrian target detection and a comprehensive analysis of the detection results from these three parts. Experiments show that the proposed method has high accuracy, ideal real-time performance, and strong reliability. Research works, such as the present study, can serve as reference for vehicle safety driving technology.
Abstract:Illumination for pedestrian detection at nighttime is weak, and detection is easily affected through variations in illumination. Thus, a bicharacteristic method of pedestrian detection at nighttime based on hierarchical tree cascade classification is presented according to "coarse-to-fine" principle. The proposed method consists of two stages of cascade classifiers. Coarse cascade classifiers are constructed in complete binary tree architecture. These classifiers use Haar-like features for the rapid identification of candidate pedestrian areas. By contrast, fine cascade classifiers have a parallel structure. Edge let features are used for detection along three parts: the head-shoulder, trunk, and leg parts of candidate pedestrian areas. Bayesian decision-making is adopted to achieve pedestrian target detection and a comprehensive analysis of the detection results from these three parts. Experiments show that the proposed method has high accuracy, ideal real-time performance, and strong reliability. Research works, such as the present study, can serve as reference for vehicle safety driving technology.
张荣辉, 周佳利, 游峰, 周喜, 裴玉龙. 基于分级树状级联分类的双特征夜间行人检测[J]. Journal of Highway and Transportation Research and Development, 2015, 9(2): 77-83.
ZHANG Rong-hui, ZHOU Jia-li, YOU Feng, ZHOU Xi, PEI Yu-long. A Pedestrian Detection Method Based on Hirarchical Tree Cascade Classsification at Nighttime. Journal of Highway and Transportation Research and Development, 2015, 9(2): 77-83.
[1] GERONIMO D, LOPEZ M, SAPPA A D. Survey of Pedestrian Detection for Advanced Driver Assistance Systems[J]. IEEE Transactions Pattern Analysis Machine Intelligence, 2010, 32(7):1239-1258.
[2] TANG Yi. Study on Video-based Detection and Tracking Method of Pedestrian in ITS[D]. Guangzhou:South China University of Technology, 2010. (in Chinese).
[3] GUO Lie,GAO Long,ZHAO Zong-yan.Pedestrian Detection and Tracking Based on Automotive Vision[J]. Journal of Southwest Jiaotong University, 2012, 47(01):19-25. (in Chinese)
[4] GE Jun-feng, LUO Yu-pin, TEI Gyo-mei. Real-time Pedestrian Detection and Tracking at Nighttime for Driver-assistance Systems[J]. IEEE Transactions on Intelligent Transportation Systems, 2009, 10(2):283-298.
[5] CHENG Ru-zhong, ZHAO Yong, Wong Chup-Chung,et al. Real-time Pedestrian Detecting and Warning System[J].Journal of Traffic and Transportation Engineering, 2012,12(05):110-118. (in Chinese)
[6] QU Yong-yu, LIU Qing, GUO Jian-ming. HOG and Color Based Pedestrian Detection[J].Journal of Wuhan University of Technology,2011, 33(4):134-138. (in Chinese)
[7] CHEN Yu-xi, YU Le-bing. Night-time Pedestrian Detection by Infrared-visual Video Fusion[J].Computer Engineering and Applications,2009, 45(24):213-215. (in Chinese)
[8] ZHANG Yang, LIU Wei-Ming, WU Yi-hu. A Fast Pedestrian Detection Method for Vehicle Auxiliary Driving System[J].Journal of Highway and Research and Development 2013,30(11):132-138.
[9] PARK Ki-yeong, HWANG Sun-young. An Improved Haar-like Feature for Efficient Object Detection[J].Pattern Recognition Letters,2014,42(1):148-153.
[10] YANG Sung-min,JO Kang-hyun.HOG Based Pedestrian Detection and Behavior Pattern Recognition for Traffic Signal Control[J]. Journal of Institute of Control, Robotics, and Systems,2013,19(11):1017-1021.
[11] ZHANG Yang, LIU Wei-ming. A Novel Pedestrian Detection Method based on Cost-Sensitive Support Vector Machine and Chaotic Particle Swarm Optimization with Tmutation[J]. Przeglad Elektrotechnicany, 2012, 88(1):22-25.
[12] PANGOP L N, CHAUSSE F, CORNOU S, et al. Adaptive Bayesian Combination of Features from Laser Scanner and Camera for Pedestrian Detection[C]//6th IFAC Symposium on Intelligent Autonomous Vehicles(IAV 200707),Toulouse:[s.n.], 2007.
[13] HUANG Chung-hsien, WU Yi-ta, SHIH Ming-Yu. Unsupervised Pedestrian Re-identification for Loitering Detection[C]//3rd Pacific Rim Symposium on Image and Video Technology(PSIVT2009),Tokyo:[s.n.],2009:771-783.
[14] XU YW, CAO XB. An Efficient Classifier Ensemble-based Approach for Pedestrian Detection[J]. Journal of IEEE Transactions on Systems,Man,and Cybernetics,2011,41(1):107-117.
[15] YOU Feng, ZHANG Rong-hui, WANG Hai-wei, et al. Dynamic Model and Tracking Control of Tractor Semi-trailer Vehicle with Underactuated System[J]. Journal of Jilin University:Engineering and Technology Edition, 2014, 44(5):1296-1302. (in Chinese)
[16] TAIANA M,NASCIMENTO J C,BERNARDINO A. An Improved Labelling for the INRIA Person Data Set for Pedestrian Detection[C]//6th Iberian Conference on Pattern Recognition and Image Analysis. Madeira, Portugal:Springer-Verlay Berlin and Heidelberg GmbH & Co.K, 2013.
[17] MOHANP C, POGGIO T. Example-Based Object Detection in Images by Components[J]. IEEE Transaction On Pattern Analysis and Machine Intelligence (S0162-8828), 2001, 23(4):349-361.
[1]
周星宇, 李红梅, 郑伟皓, 唐智慧, 杨丽君. 基于交互式BP-UKF模型的短时交通流预测方法[J]. Journal of Highway and Transportation Research and Development, 2019, 13(2): 56-64.
[2]
王东柱, 陈艳艳, 马建明, C. Michael WALTON. 车联网环境下的高速公路合流区协调控制方法及效果评价[J]. Journal of Highway and Transportation Research and Development, 2017, 11(4): 90-97.
[3]
戢晓峰, 张玲, 冯川. 多源出行信息影响下的路径选择行为仿真研究[J]. Journal of Highway and Transportation Research and Development, 2017, 11(1): 77-83.
[4]
王雁, 穆春阳, 马行. 基于Zernike不变矩与SVM交通标志的识别[J]. Journal of Highway and Transportation Research and Development, 2016, 10(3): 85-89.
[5]
高剑, 张凡, 孙玲. 公路资产与交通安全的关系研究(II)[J]. Journal of Highway and Transportation Research and Development, 2015, 9(4): 91-96.
[6]
张邻, 黄选伟, 吴伟明. 机非混行驾驶行为博弈分析[J]. Journal of Highway and Transportation Research and Development, 2015, 9(3): 85-93.
[7]
庞明宝, 蔡章辉, 杨敏, 张莎莎. 高速公路协调控制信号施加策略[J]. Journal of Highway and Transportation Research and Development, 2015, 9(1): 79-87.
[8]
宋向辉, 李亚檬, 赵佳海, 王新科. 一种面向安全辅助预警的车辆相对位置判定方法[J]. Journal of Highway and Transportation Research and Development, 2014, 8(2): 88-92.
[9]
王东柱, 杨琪. 欧洲合作智能交通系统发展现状及相关标准分析[J]. Journal of Highway and Transportation Research and Development, 2014, 8(2): 81-87.
[10]
高剑, 张建栋, 罗梅村. 公路资产与交通安全的关系研究(I)[J]. Journal of Highway and Transportation Research and Development, 2014, 8(2): 99-104.
[11]
杨文丽, 杨蕴, 刘鸿伟. 基于出行者需求快速聚类的公路交通路径规划研究[J]. Journal of Highway and Transportation Research and Development, 2014, 8(1): 84-87.
[12]
王东柱, 陈艳艳. 基于浮动车停车点数据交叉口车辆排队长度计算方法[J]. Journal of Highway and Transportation Research and Development, 2013, 7(3): 84-89.
[13]
庞明宝, 任莎莎, 王彦虎, 陈培. 基于减法聚类的高速公路主线可变速度FNN混沌控制[J]. Journal of Highway and Transportation Research and Development, 2013, 7(3): 90-98.
[14]
龚, 李苏剑. 基于CPS理论的城市交通控制与诱导融合框架[J]. Journal of Highway and Transportation Research and Development, 2013, 7(1): 82-89.