·314· 智能系统学报 第14卷 作主要针对两点：1)应用本文算法在多个基准行 improved detection[J].Advances in neural information 人数据集（如Caltech行人数据集等）上进行实验， processing systems,2014,1:424-432 针对每个数据集的测试结果进行统计分析，优化 [13]ZHANG Shanshan,BENENSON R,SCHIELE B. 本文算法的检测性能：2)继续扩充行人数据集的 Filtered channel features for pedestrian detection[C]//Pro- 数量跟多样性能够进一步的提升算法的检测性能。 ceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.Boston,USA,2015:1751-1760. 参考文献： [14]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Im. agenet classification with deep convolutional neural net- [1]宋婉茹，赵晴晴，陈昌红，等.行人重识别研究综述).智 works[J].Advances in neural information processing sys- 能系统学报，2017,12(6：770-780 tems,2012,25(2):1097-1105 SONG Wanru,ZHAO Qingqing,CHEN Changhong,et al. Survey on pedestrian re-identification research[J].CAAI [15]SIMONYAN K,ZISSERMAN A.Very deep convolu- transactions on intelligent systems,2017,12(6):770-780. tional networks for large-scale image recognition[J].arX- iv:1409.1556.2014 [2]YE Qixiang,LIANG Jixiang,JIAO Jianbin.Pedestrian de- [16]GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich tection in video images via error correcting output code feature hierarchies for accurate object detection and se- classification of manifold subclasses[J].IEEE transactions mantic segmentation[C]//Proceedings of 2014 IEEE Con- on intelligent transportation systems,2012,13(1): 193-202. ference on Computer Vision and Pattern Recognition. [3]LIU Wei,ANGUELOV D,ERHAN D,et al.SSD:single Columbus,USA,2014:580-587. shot multibox detector[C]//Proceedings of 2016 European [17]REN Shaoging,HE Kaiming,GIRSHICK R,et al.Faster Conference on Computer Vision.Cham,Germany,2016: R-CNN:towards real-time object detection with region 21-37. proposal networks[J].IEEE transactions on pattern ana- [4]DALAL N.TRIGGS B.Histograms of oriented gradients lysis and machine intelligence,2017,39(6):1137-1149. for human detection[Cl//IEEE Computer Society Confer- [18]REDMON J,DIVVALA S,GIRSHICK R,et al.You only ence on Computer Vision and Pattern Recognition.San look once:unified,real-time object detection[C]//Pro- Diego,USA,2005:886-893 ceedings of 2016 IEEE Conference on Computer Vision [5]苏松志，李绍滋，陈淑媛，等.行人检测技术综述)电子 and Pattern Recognition.Las Vegas,USA,2016: 学报，2012,40(4)：814-820 779-788. SU Songhi,LI Shaozi,CHEN Shuyuan,et al.A survey on [19]王梦来，李想，陈奇，等.基于CNN的监控视频事件检 pedestrian detection[J].Acta electronica sinica,2012, 测U.自动化学报，2016,42(6)：892-903， 40(4):814-820. WANG Menglai,LI Xiang,CHEN Qi,et al.Surveillance [6]LOWE D G.Distinctive image features from scale-invari- event detection based on CNN[J].Acta automatica sinica. ant keypoints[J].International journal of computer vision, 2016,42(6):892-903 2004,60(2):91-110. [20]HOSANG J,OMRAN M,BENENSON R,et al.Taking a [7]VIOLA P,JONES M.Rapid object detection using a boos- deeper look at pedestrians[Cl//Proceedings of 2015 IEEE ted cascade of simple features[C]//Proceedings of the 2001 Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Conference Computer Vision and Boston,USA,2015:4073-4082. Pattern Recognition.Kauai,USA,2001:511-518 [21]BENENSON R,OMRAN M,HOSANG J,et al.Ten years [8]FERREIRA A J.FIGUEIREDO M A T.Boosting al- of pedestrian detection,what have we learned?[C]//Pro- gorithms:a review of methods,theory,and applications[M]. ceedings of 2014 European Conference on Computer Vis- New York,USA:Springer,2012:35-85. ion.Cham,Germany,2015:613-627. [9]VAPNIK V.The nature of statistical learning theory[M]. [22]吕静，高陈强，杜银和，等.基于双通道特征自适应融合 2nd eds.New York:Springer-Verlag,2000. 的红外行为识别方法).重庆邮电大学学报（自然科学 [10]BREIMAN L.Random forests[J].Machine learning, 版)，2017,293：389-395 2001,45(1)y5-32. LYU Jing,GAO Chengiang,DU Yinhe,et al.Infrared ac- [11]DOLLAR P.APPEL R.BELONGIE S,et al.Fast feature tion recognition method based on adaptive fusion of dual pyramids for object detection[J].IEEE transactions on channel features[J].Journal of Chongqing university of pattern analysis and machine intelligence,2014,36(8): posts and telecommunications (natural science edition), 1532-1545. 2017,29(3:389-395 [12]NAM W.DOLLAR P.HAN J H.Local decorrelation for [23]TIAN Yonglong,LUO Ping,WANG Xiaogang,et al.作主要针对两点：1）应用本文算法在多个基准行 人数据集（如 Caltech 行人数据集等）上进行实验， 针对每个数据集的测试结果进行统计分析，优化 本文算法的检测性能；2）继续扩充行人数据集的 数量跟多样性能够进一步的提升算法的检测性能。 参考文献： 宋婉茹, 赵晴晴, 陈昌红, 等. 行人重识别研究综述[J]. 智 能系统学报, 2017, 12(6): 770–780. SONG Wanru, ZHAO Qingqing, CHEN Changhong, et al. Survey on pedestrian re-identification research[J]. CAAI transactions on intelligent systems, 2017, 12(6): 770–780. [1] YE Qixiang, LIANG Jixiang, JIAO Jianbin. Pedestrian de￾tection in video images via error correcting output code classification of manifold subclasses[J]. IEEE transactions on intelligent transportation systems, 2012, 13(1): 193–202. [2] LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proceedings of 2016 European Conference on Computer Vision. Cham, Germany, 2016: 21–37. [3] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//IEEE Computer Society Confer￾ence on Computer Vision and Pattern Recognition. San Diego, USA, 2005: 886–893 [4] 苏松志, 李绍滋, 陈淑媛, 等. 行人检测技术综述[J]. 电子 学报, 2012, 40(4): 814–820. SU Songhi, LI Shaozi, CHEN Shuyuan, et al. A survey on pedestrian detection[J]. Acta electronica sinica, 2012, 40(4): 814–820. [5] LOWE D G. Distinctive image features from scale-invari￾ant keypoints[J]. International journal of computer vision, 2004, 60(2): 91–110. [6] VIOLA P, JONES M. Rapid object detection using a boos￾ted cascade of simple features[C]//Proceedings of the 2001 IEEE Computer Society Conference Computer Vision and Pattern Recognition. Kauai, USA, 2001: 511–518. [7] FERREIRA A J, FIGUEIREDO M A T. Boosting al￾gorithms: a review of methods, theory, and applications[M]. New York, USA: Springer, 2012: 35–85. [8] VAPNIK V. The nature of statistical learning theory[M]. 2nd eds. New York: Springer-Verlag, 2000. [9] BREIMAN L. Random forests[J]. Machine learning, 2001, 45(1): 5–32. [10] DOLLÁR P, APPEL R, BELONGIE S, et al. Fast feature pyramids for object detection[J]. IEEE transactions on pattern analysis and machine intelligence, 2014, 36(8): 1532–1545. [11] [12] NAM W, DOLLÁR P, HAN J H. Local decorrelation for improved detection[J]. Advances in neural information processing systems, 2014, 1: 424–432. ZHANG Shanshan, BENENSON R, SCHIELE B. Filtered channel features for pedestrian detection[C]//Pro￾ceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA, 2015: 1751–1760. [13] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Im￾agenet classification with deep convolutional neural net￾works[J]. Advances in neural information processing sys￾tems, 2012, 25(2): 1097–1105. [14] SIMONYAN K, ZISSERMAN A. Very deep convolu￾tional networks for large-scale image recognition[J]. arX￾iv: 1409.1556, 2014. [15] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and se￾mantic segmentation[C]//Proceedings of 2014 IEEE Con￾ference on Computer Vision and Pattern Recognition. Columbus, USA, 2014: 580–587. [16] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern ana￾lysis and machine intelligence, 2017, 39(6): 1137–1149. [17] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Pro￾ceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA, 2016: 779–788. [18] 王梦来, 李想, 陈奇, 等. 基于 CNN 的监控视频事件检 测[J]. 自动化学报, 2016, 42(6): 892–903. WANG Menglai, LI Xiang, CHEN Qi, et al. Surveillance event detection based on CNN[J]. Acta automatica sinica, 2016, 42(6): 892–903. [19] HOSANG J, OMRAN M, BENENSON R, et al. Taking a deeper look at pedestrians[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA, 2015: 4073–4082. [20] BENENSON R, OMRAN M, HOSANG J, et al. Ten years of pedestrian detection, what have we learned?[C]//Pro￾ceedings of 2014 European Conference on Computer Vis￾ion. Cham, Germany, 2015: 613–627. [21] 吕静, 高陈强, 杜银和, 等. 基于双通道特征自适应融合 的红外行为识别方法[J]. 重庆邮电大学学报(自然科学 版), 2017, 29(3): 389–395. LYU Jing, GAO Chenqiang, DU Yinhe, et al. Infrared ac￾tion recognition method based on adaptive fusion of dual channel features[J]. Journal of Chongqing university of posts and telecommunications (natural science edition), 2017, 29(3): 389–395. [22] [23] TIAN Yonglong, LUO Ping, WANG Xiaogang, et al. ·314· 智 能 系 统 学 报 第 14 卷