-
摘要: 针对重特大交通事故中的行人保护问题, 提出了基于侧面行人特征的实时行人检测预警系统(PDWS)。系统由检测模块和预警模块两部分组成, 其中检测模块使用Haar与HOG特征和AdaBoost与SVM分类器, 通过侧面行人样本库完成行人特征的提取与检测, 同时应用窗口拆分法与快速窗口扫描算法提高检测效率, 得到一个具有高检测率与低误检率的结果。在预警模块融合了行人距离、汽车速度及角速度信息, 判断前方行人存在碰撞的危险。使用系统及算法对城市环境复杂背景下横过街道的行人进行了实车验证。测试结果表明: 对704Pixel×576Pixel图像的检测帧率为13~18帧.s-1, 检测率大于85%, 误检率小于1%, 预警时间小于1s, 实车验证结果达到了车载主动安全系统实时性与准确性的要求。Abstract: A real-time pedestrian detecting and warning system(PDWS) based on the features of pedestrian side was proposed to solve the problem of pedestrian protection in serious traffic accidents. The system consisted of two parts, detecting module and warning module. The feature extraction and detection pedestrian were completed by using side pedestrian sample dataset in detecting module, Haar and HOG features together with AdaBoost and SVM classifiers were applied to complete the feature extraction and detection. Window-dividing method and operator context scanning(OCS) method were used to improve the detecting efficiency, and a result with both high detection rate and low false alarm rate was obtained. The velocity and angular velocity of automobile together with the distance of detected pedestrian were merged in warning module, so the system could judge the collision risk for the pedestrian in the front. The system and the algorithms were tested toward the pedestrian crossing street with a complex urban envirnment in real vehicle. Test result shows that for the images with 704 Pixel×576 Pixel, the frame rate is about 13-18 frame·s-1, the detecting rate is above 85%, the false detecting rate is below 1%, and the warning response time is less than 1 s. The result meets the requirements of on-board active safety system both in accuracy and real-time use.
-
表 1 检测结果
Table 1. Detection result
视频 方法 检测率% 误检率% 帧率/(帧·s-1) 1 HOG+SVM 100.0 1.70 0.75 定位+验证 96.7 0.00 8.30 OCS 93.3 1.70 11.80 OCS+Div2 91.7 0.00 14.60 2 OCS+Div3 80.0 0.00 17.20 HOG+SVM 92.5 3.75 0.75 定位+验证 90.0 0.00 8.60 OCS 86.3 0.00 12.30 3 OCS+Div2 82.5 0.00 14.80 OCS+Div3 86.3 0.00 18.50 HOG+SVM 95.8 0.00 0.75 定位+验证 90.1 4.70 8.70 4 OCS 93.0 4.70 12.10 OCS+Div2 85.2 0.00 14.20 OCS+Div3 73.2 0.00 17.80 HOG+SVM 97.4 4.00 0.75 定位+验证 94.1 0.00 7.90 OCS 88.9 0.00 11.20 OCS+Div2 83.0 2.00 13.20 OCS+Div3 71.2 0.00 16.30 
下载: 导出CSV
表 2 不同距离下的检测率
Table 2. Detection rates at different distances
距离/m 5 10 20 30 40 50 60 检测率/% 78.0 83.0 87.5 80.0 50.0 29.0 < 10.0
下载: 导出CSV
-
[1] 乔维高, 朱西产. 行人与汽车碰撞中头部伤害与保护的研究[J]. 农业机械学报, 2006, 37(9): 29-31. doi: 10.3969/j.issn.1000-1298.2006.09.008QIAO Wei-gao, ZHU Xi-chan. Study of head injury and protection in vehicle-pedestrian collisions[J]. Transactions of the Chinese Society for Agricultural Machinery, 2006, 37(9): 29-31. (in Chinese). doi: 10.3969/j.issn.1000-1298.2006.09.008 [2] 李充, 朱海涛. 行人头部保护试验规程研究及对比[J]. 交通标准化, 2009(17): 75-79. https://www.cnki.com.cn/Article/CJFDTOTAL-JTBH200917022.htmLI Chong, ZHU Hai-tao. Research and comparison on pedes-trian head protection test regulations[J]. Communications Standardization, 2009(17): 75-79. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-JTBH200917022.htm [3] 张轶川, 苗强, 朱西产, 等. 行人头部碰撞子模型适用性的研究[J]. 汽车工程, 2010, 32(3): 223-227. https://www.cnki.com.cn/Article/CJFDTOTAL-QCGC201003011.htmZHANG Yi-chuan, MIAO Qiang, ZHU Xi-chan, et al. A study on applicability of sub-model for pedestrian head impact[J]. Automotive Engineering, 2010, 32(3): 223-227. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-QCGC201003011.htm [4] 潘光亮, 章兰珠, 陈碧峰, 等. 基于行人头部保护碰撞的试验设备设计[J]. 华东理工大学学报: 自然科学版, 2011, 37(2): 244-248. doi: 10.3969/j.issn.1006-3080.2011.02.021PAN Guang-liang, ZHANG Lan-zhu, CHEN Bi-feng, et al. Design of experimental facility based on pedestrian protection head impact[J]. Journal of East China University of Science and Technology: Natural Science Editon, 2011, 37(2): 244-248. (in Chinese). doi: 10.3969/j.issn.1006-3080.2011.02.021 [5] 赵正, 陈超卓, 吴沈荣. 车体前端造型及材料对于行人腿部伤害指标的影响[J]. 汽车安全与节能学报, 2010, 1(4): 297-306. doi: 10.3969/j.issn.1676-8484.2010.04.008ZHAO Zheng, CHEN Chao-zhuo, WU Shen-rong. Influence of the vehicle front sculpt and bumper foam on pedestrian leg injury index[J]. Journal of Automotive Safety and Energy, 2010, 1(4): 297-306. (in Chinese). doi: 10.3969/j.issn.1676-8484.2010.04.008 [6] 肖宏伟, 陆善彬, 武栎楠. 基于行人腿部保护的保险杠造型优化设计[J]. 汽车技术, 2010(10): 30-33. https://www.cnki.com.cn/Article/CJFDTOTAL-QCJS201010010.htmXIAO Hong-wei, LU Shan-bin, WU Li-nan. Bumper stylingoptimization design based on pedestrian leg protection[J]. Automobile Technology, 2010(10): 30-33. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-QCJS201010010.htm [7] 李向荣, 吕恒绪, 张向磊, 等. 立姿假人实车碰撞试验的研究[J]. 汽车工程, 2011, 33(1): 11-14, 37. https://www.cnki.com.cn/Article/CJFDTOTAL-QCGC201101005.htmLI Xiang-rong, LU Heng-xu, ZHANG Xiang-lei, et al. A study on real vehicle-standing pedestrian dummy impact test[J]. Automotive Engineering, 2011, 33(1): 11-14, 37. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-QCGC201101005.htm [8] 李向荣, 王凯, 刘志新, 等. 汽车发动机罩儿童行人保护安全性仿真研究[J]. 汽车工程, 2010, 32(1): 56-59. https://www.cnki.com.cn/Article/CJFDTOTAL-QCGC201001012.htmLI Xiang-rong, WANG Kai, LIU Zhi-xin, et al. A simulation study on vehicle bonnet safety for child pedestrian protection[J]. Automotive Engineering, 2010, 32(1): 56-59. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-QCGC201001012.htm [9] LIORCA D F, SOTELO M A, PARRA I, et al. An experi-mental study on pitch compensation in pedestrian-protection systems for collision avoidance and mitigation[J]. IEEE Transactions on Intelligent Transportation Systems, 2009, 10(3): 469-474. [10] LIORCA D F, MILANES V, ALONSO I P, et al. Autono-mous pedestrian collision avoidance using a fuzzy steering controller[J]. IEEE Transactions on Intelligent Transportation Systems, 2011, 12(2): 390-401. [11] DAVID K, FLACH A. Car-2-X and pedestrian safety: innovative collision avoidance system[J]. IEEE Vehicular Technology Magazine, 2010, 5(1): 70-76. [12] GIDEL S, CHECCHIN P, BLANC C, et al. Pedestrian detection and tracking in an urban environment using a multi-layer laser scanner[J]. IEEE Transactions on Intelligent Transportation Systems, 2010, 11(3): 579-588. [13] GAVRILA D. Sensor-based pedestrian protection[J]. IEEE Intelligent Systems, 2001, 16(6): 77-81. [14] GERONIMO D, LOPEZ A M, SAPPA A D, et al. Survey of pedestrian detection for advanced driver assistance systems[J]. IEEE Transactions on Pattern Analysis and Machine Intelli-gence, 2010, 32(7): 1239-1258. [15] VIOLA P, JONES M J, SNOW D. Detecting pedestrians usingpatterns of motion and appearance[C]∥IEEE. Proceedings of Ninth IEEE International Conference on Computer Vision. Nice: IEEE, 2003: 734-741. [16] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]∥IEEE. Proceedings of the 2005IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego: IEEE, 2005: 886-893. [17] ZHU Qiang, AVIDAN S, YEH M C, et al. Fast human detection using a cascade of histograms of oriented gradients[C]∥IEEE. Proceedings of the 2006IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2006: 1491-1498. [18] WANG C C R, LIEN J J J. AdaBoost learning for human detection based on histograms of oriented gradients[C]∥IEEE. Proceedings of the 8th Asian Conference on Computer Vision. Toyota: IEEE, 2007: 885-895. [19] JIA Hui-xing, ZHANG Yu-jin. Fast human detection by boosting histograms of oriented gradients[C]∥IEEE. Fourth International Conference on Image and Graphics. Chengdu: IEEE, 2007: 683-688. [20] WANG Zhen-rui, JIA Yu-lan, HUANG Hua, et al. Pedestrian detection using boosted HOG features[C]∥IEEE. Proceedings of the 11th International IEEE Conference on Intelligent Transportation Systems. Beijing: IEEE, 2008: 1155-1160. [21] GIANNOUKOS I, ANAGNOSTOPOULOS C, LOUMOS V, et al. Operator context scanning to support high segmenta-tion rates for real time license plate recognition[J]. Pattern Recognition, 2010, 43(11): 3866-3878. [22] CUI Xin-yi, LIU Ya-zhou, SHAN Shi-guang, et al. 3DHaar-like features for pedestrian detection[C]∥IEEE. Proceedings of the 2007IEEE International Conference on Multimedia and expo. Beijing: IEEE, 2007: 1263-1266. [23] LIU Ya-zhou, CHEN Xi-lin, YAO Hong-xun. Contour-motion feature(CMF): a space-time approach for robust pedestrian detection[J]. Pattern Recognition Letters, 2009, 30(2): 148-156. [24] DOLLAR P, WOJEK C, SCHIELE B, et al. Pedestrian detection: a benchmark[C]∥IEEE. Proceedings of the 2009 IEEE Computer Socient Conference on Computer Vision and Pattern Recognition. Miami: IEEE, 2009: 304-311. [25] RUSSELL B C, TORRALBA A, MURPHY K P, et al. LabelMe: a database and web-based tool for image annotation[J]. Interna-tional Journal of Computer Vision, 2008, 77(1/2/3): 157-173. [26] WOJEK C, WALK S, SCHIELE B. Multi-cue onboard pedestrian detection[C]∥IEEE. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Miami: IEEE, 2009: 794-801. [27] VIOLA P, JONES M. Rapid object detection using a boosted cascade of simple features[C]∥IEEE. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Kauai: IEEE, 2001: 511-518. [28] FAN R K, CHANG K W, HSIEH C J, et al. LIBLINEAR: a library for large linear classification[J]. Journal of Machine Learning Research, 2008(9): 1871-1874. [29] CHENG Ru-zhong, ZHAO Yong, WONG Chup-chung, et al. An on-board pedestrian detection and warning system with features of side pedestrian[C]∥SPIE. Proceeding of Visual Information Processing and Communication III. Burlingame: SPIE, 2012: 1-10. -
点击查看大图
图(18) / 表(2)
计量
- 文章访问数: 1084
- HTML全文浏览量: 247
- PDF下载量: 2210
- 被引次数: 0
