自适应分块的多特征融合多目标跟踪

doi:10.3969/j.issn.1001-2400.2017.02.026

摘要/Abstract

摘要：

针对多目标跟踪过程中存在目标遮挡、表观变化以及目标相似的情况，提出一种基于自适应分块的多特征融合粒子滤波多目标跟踪方法．该方法首先根据目标灰度投影进行自适应分块，融合颜色直方图及方向梯度直方图特征描述各子块，并引入加权Bhattacharyya系数计算粒子的子块匹配度;然后利用模糊C均值聚类获得每个目标的粒子群，得到目标最优状态估计; 最后融入粒子空间信息更新子块权重．实验结果表明，该方法在多种复杂情况下，均能准确鲁棒地跟踪多个目标．

关键词: 多目标, 自适应分块, 多特征融合, 方向梯度直方图, 粒子滤波

Abstract:

In order to solve the problems of occlusion, appearance change and similar targets in multi-target tracking, this paper proposes a multi-target tracking method with the particle filter based on the adaptive fragment and multi-feature fusion. First, we divide each target into a few fragments adaptively according to its gray projection and describe each fragment with the color histogram and histogram of oriented gradient(HOG) feature, of which similarity can be obtained by adopting the weighted Bhattacharyya Coefficient. Then, we obtain particle sets of each target by FCM clustering. The optimal state estimation of each target is calculated through particles state in the subgroup. Finally, the weighting factor of each fragment is updated according to the reliability which is calculated by considering particles space information. Experimental results show that the proposed method can track the targets in many complex circumstances accurately and robustly.

Key words: multiple targets, adaptive fragment, multi-feature fusion, histogram of oriented gradient, particle filter

别秀德;刘洪彬;常发亮;彭志勇. 自适应分块的多特征融合多目标跟踪[J]. 西安电子科技大学学报, 2017, 44(2): 151-157.

BIE Xiude;LIU Hongbin;CHANG Faliang;PENG Zhiyong. Multi-target tracking method based on the adaptive fragment and multi-feature fusion[J]. Journal of Xidian University, 2017, 44(2): 151-157.

参考文献

［1］ YI W, MORELANDE M R, KONG L, et al. A Computationally Efficient Particle Filter for Multitarget Tracking Using an Independence Approximation［J］. IEEE Transactions on Signal Processing, 2013, 61(4): 843-856.
［2］ VATAVU A, DANESCU R, NEDEVSCHI S. Stereovision-based Multiple Object Tracking in Traffic Scenarios Using Free-form Obstacle Delimiters and Particle Filters［J］. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(1): 498-511.
［3］李远征, 卢朝阳, 李静. 一种基于多特征融合的视频目标跟踪方法［J］. 西安电子科技大学学报, 2012, 39(4): 1-6.
LI Yuanzheng, LU Zhaoyang, LI Jing. Robust Video Object Tracking Algorithm Based on Multi-feature Fusion［J］. Journal of Xidian University, 2012, 39(4): 1-6.
［4］ DANESCU R, ONIGA F, NEDEVSCHI S, et al. Tracking Multiple Objects Using Particle Filters and Digital Elevation Maps［C］//Proceedings of the IEEE Intelligent Vehicles Symposium. Piscataway: IEEE, 2009: 88-93.
［5］陈龙, 郭宝龙, 孙伟. 基于 FCM 聚类的粒子滤波多目标跟踪算法［J］. 仪器仪表学报, 2011, 32(11): 2536-2542.
CHEN Long, GUO Baolong, SUN Wei. Multi-target Tracking Algorithm Based on FCM and Particle Filter［J］. Chinese Journal of Scientific Instrument, 2011, 32(11): 2536-2542.
［6］ HU W M, LI X, LUO W H, et al. Single and Multiple Object Tracking Using Log-Euclidean Riemannian Subspace and Block-division Appearance Model［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(12): 2420-2440.
［7］ ZHOU R, TENG J. An Improved Resampling Algorithm for Particle Filtering in Small Target Tracking［J］. Journal of Coastal Research, 2015, 73(s1): 600-605.
［8］孙晓燕, 常发亮. 自适应分块多线索融合粒子滤波跟踪方法［J］. 控制与决策, 2014, 29(9): 1678-1682.
SUN Xiaoyan, CHANG Faliang. Adaptive Fragment Multi-cue Fusion Particle Filter Tracking Method［J］. Control and Decision, 2014, 29(9): 1678-1682.
［9］张立朝, 毕笃彦, 查宇飞, 等. 动态特征和静态特征自适应融合的目标跟踪算法［J］. 西安电子科技大学学报, 2015, 42(6): 164-172.
ZHANG Lichao, BI Duyan, ZHA Yufei, et al. Research on Visual Object Tracking by Fusing Dynamic and Static Features［J］. Journal of Xidian University, 2015, 42(6): 164-172.
［10］ DOU J F, LI J X. Robust Visual Tracking Base on Adaptively Multi-feature Fusion and Particle Filter［J］. Optik, 2014, 125(5): 1680-1686.
［11］闫河, 刘婕, 杨德红, 等. 基于特征融合的粒子滤波目标跟踪新方法［J］. 光电子·激光, 2014, 25(10): 1990-1999.
YAN He, LIU Jie, YANG Dehong, et al. A New Particle Filter Object Tracking Algorithm Based on Feature Fusion［J］. Journal of Optoelectronics·Laser, 2014, 25(10): 1990-1999.
［12］ NICKEL K, STIEFELHAGEN R. Fast Audio-visual Multi-Person Tracking for a Humanoid Stereo Camera Head［C］//Proceedings of the 2007 7th IEEE-RAS International Conference on Humanoid Robots. Piscataway: IEEE, 2008: 434-441.
［13］ NICKEL K, STIEFELHAGEN R. Dynamic Integration of Generalized Cues for Person Tracking［C］//Lecture Notes in Computer Science:5305. Heidelberg: Springer Verlag, 2008: 514-526.

[1]	张宇鹏,吴自力,陈鸣,张璐璐. 面向交叉微服务链的任务调度优化[J]. 西安电子科技大学学报, 2021, 48(6): 32-39.
[2]	杜文占,余志勇,杨剑,姜海滨. 改进NSGA-II算法及监测天线部署优化研究[J]. 西安电子科技大学学报, 2021, 48(5): 239-248.
[3]	刘天宇,王翥. 一种多样性控制的多目标粒子群算法[J]. 西安电子科技大学学报, 2021, 48(3): 106-114.
[4]	杨五四,陈莉,王毅,张茂省. 一种适应度排序的高维多目标粒子群优化算法[J]. 西安电子科技大学学报, 2021, 48(3): 78-84.
[5]	张逸宸,水鹏朗. 一种改进的泊松点过程概率多假设跟踪方法[J]. 西安电子科技大学学报, 2021, 48(2): 27-34.
[6]	王明杰,姬红兵,刘龙. 噪声野值下的学生t分布混合CPHD滤波[J]. 西安电子科技大学学报, 2019, 46(5): 134-141.
[7]	李翠芸,陈东伟,石仁政. 自适应目标新生δ广义标签多伯努利滤波算法[J]. 西安电子科技大学学报, 2019, 46(2): 12-16.
[8]	王燕. 改进多目标进化算法的云工作流调度[J]. 西安电子科技大学学报, 2019, 46(1): 130-136.
[9]	王强;郭立新;刘忠玉. 粗糙海面与上方多目标瞬态电磁散射混合算法[J]. 西安电子科技大学学报, 2018, 45(5): 57-63.
[10]	韩亚敏;柴争义;李亚伦;朱思峰. 长尾群组推荐的免疫多目标优化实现[J]. 西安电子科技大学学报, 2018, 45(3): 109-116.
[11]	王丹青;李萍. 利用GEK代理模型的天线快速多目标优化设计[J]. 西安电子科技大学学报, 2018, 45(2): 129-134+165.
[12]	刘润邦;朱志宇. 万有引力优化的粒子滤波算法[J]. 西安电子科技大学学报, 2018, 45(2): 141-147.
[13]	梁付新;刘洪彬;常发亮. 多特征融合匹配的霍夫森林多目标跟踪[J]. 西安电子科技大学学报, 2018, 45(1): 129-134+161.
[14]	贺磊南;卢虎;姬正洲. WiFi/PDR室内分层三维定位技术[J]. 西安电子科技大学学报, 2017, 44(4): 167-173+166.
[15]	李翠芸;王精毅;姬红兵;王荣. 模型参数未知时的CPHD多目标跟踪方法[J]. 西安电子科技大学学报, 2017, 44(2): 37-41+184.