首页 > 论文 > 激光与光电子学进展 > 54卷 > 9期(pp:91004--1)

基于HSV颜色特征和贡献度重构的行人跟踪

Pedestrian Tracking Based on HSV Color Features and Reconstruction by Contributions

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

在光照、背景变化、遮挡、噪声、快速运动等复杂环境下, 准确地实现行人跟踪一直是富有挑战性的任务。针对这些问题, 提出基于HSV颜色特征和贡献度重构的行人跟踪算法。在粒子滤波的框架内, 从HSV空间提取目标的混合颜色特征生成目标模板集, 依据不同区域对跟踪结果的影响对区域进行贡献度分配, 并将其引入到一个自适应的正则化模型中, 将具有最小重构误差的区域判定为待跟踪目标。为了增强算法的稳健性, 跟踪过程中对模板进行实时更新。在OTB 100个序列上进行测试, 本文算法得到跟踪结果的平均中心误差和跟踪成功率两项指标分别为0.6624 pixel和0.4153, 优于同类其他算法。实验结果表明, 该算法能够在复杂的视频场景中实现对行人的连续跟踪, 且稳健性较好, 有利于在实际系统中的实现。

Abstract

It is a challenging task to track pedestrian accurately in complicated environment such as illumination, background variation, occlusion, noise and fast motion. Aiming at these problems, the tracing algorithm based on HSV color features and reconstruction by contributions is proposed. The proposed algorithm extracts the mixed color features of target in HSV space to generate the target template set within the particle filter framework. According to the influence of different regions on the tracking results, the contribution of the region is distributed. And it is introduced into the adaptive regularization model, and the region with the minimum reconstruction error is determined as the target to be tracked. In order to be more robust, the templates are updated in real time during the tracking progress. The average center error of tracking results and tracking success rate of 100 sequences tested in OTB are 0.6624 pixel and 0.4153, respectively, and the proposed algorithm has better performance than others. Experimental results show that the proposed algorithm can realize the continuous tracking for pedestrian in complex video scenes and is beneficial to be realized in the practice system with better robustness.

投稿润色
补充资料

中图分类号:TP391

DOI:10.3788/lop54.091004

所属栏目:图像处理

基金项目:国家自然科学基金(61303104,61203238,11178017, 61473015)、北京市自然科学基金(4132014, 4162017)、北京市优秀人才资助项目(2016000020124G088)

收稿日期:2017-02-14

修改稿日期:2017-03-16

网络出版日期:--

作者单位    点击查看

刘梦飞:首都师范大学信息工程学院, 北京 100048
付小雁:首都师范大学信息工程学院, 北京 100048首都师范大学电子系统可靠性技术北京市重点实验室, 北京 100048
尚媛园:首都师范大学信息工程学院, 北京 100048成像技术北京市高精尖创新中心, 北京 100048
丁辉:首都师范大学信息工程学院, 北京 100048首都师范大学高可靠嵌入式系统技术北京市工程技术研究中心, 北京 100048

联系人作者:刘梦飞(lmf_getbetter@sina.com)

备注:刘梦飞(1994-), 女, 硕士研究生, 主要从事目标跟踪与计算机视觉方面的研究。

【1】Avidan S. Ensemble tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(2): 261-271.

【2】Zhao Z Y, Collins R T. Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking[C]. Proceedings of the 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2008: 1-8.

【3】Babenko B, Yang M, Belongie S. Visual tracking with online multiple instance learning[C]. Proceedings of the 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2009: 983-990.

【4】Williams O, Blake A, Cipolla R. Sparse Bayesian learning for efficient visual tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(8): 1292-1304.

【5】Zhang Yueqiang, Su Ang, Liu Haibo, et al. Three dimensional rigid object pose tracking and optimization based on multilevel line representation and M-estimation[J]. Acta Optica Sinica, 2015, 35(1): 0115003.
张跃强, 苏 昂, 刘海波, 等. 基于多级直线表述和M-估计的三维目标位姿跟踪优化算法[J]. 光学学报, 2015, 35(1): 0115003.

【6】Zhao Y J, Zhang B, Zhang X L. Mean shift blob tracking with target model adaptive update[C]. Proceedings of Chinese Control Conference, Nanjing, China, 2014: 4831-4835.

【7】Ross D, Lim J, Lin R, et al. Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2008, 77(1): 125-141.

【8】Liu Wei, Zhao Wenjie, Li Cheng. An online learning visual tracking method based on compressive sensing[J]. Acta Optica Sinica, 2015, 35(9): 0915001.
刘 威, 赵文杰, 李 成. 一种基于压缩感知的在线学习跟踪算法[J]. 光学学报, 2015, 35(9): 0915001.

【9】Mei X, Ling H B. Robust visual tracking using L1 minimization[C]. International Conference of Computer Vision, 2009: 1436-1433.

【10】Mei X, Ling H B, Wu Y, et al. Minimum error bounded efficient L1 tracker with occlusion detection[J]. IEEE Transactions on image processing, 2013, 22(7): 2661-2675.

【11】Zhang K H, Zhang L, Yang M H. Real-timecompressive tracking[C]. European Conference on Computer Version, 2012: 864-877.

【12】Bao C G, Wu Y. Real timerobust tracker using accelerate proximal gradient approach[C]. Proceedings of the 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2012: 1830-1837.

【13】Yang X, Wang M, Zhang L M, et al. An efficient tracking system by orthogonalized templates[J]. IEEE Transactions on Industrial Electronics, 2016, 63(5): 3187-3197.

【14】Henriques J F, Caseiro R, Martins P. High-speed tracking with kernelized correlation filters[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(3): 583-596.

【15】Mei X, Ling H B, Jacobs D W. Sparse representation of cast shadows via L1 regularized least squares[C]. International Conference of Computer Vision, 2009: 583-590.

【16】Kaneko T, Hori O. Feature selection for reliable tracking using template matching[C]. Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003: 796-802.

【17】Matthews I, Ishikawa T, Baker S. The template update problem[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(6): 810-815.

【18】Kim S J, Koh K, Lustig M, et al. A interior-point method for large-scale L1 regularized least squares[J]. IEEE Journal on Selected Topics in Signal Processing, 2007, 1(4): 606-617.

【19】Li Shuangshuang, Zhao Gaopeng, Wang Jianyu. Distractor-aware object tracking based on multi-feature and scale-adaptive[J]. Acta Optica Sinica, 2017, 37(5): 0515005.
李双双, 赵高鹏, 王建宇. 基于特征融合和尺度自适应的干扰感知目标跟踪[J]. 光学学报, 2017, 37(5): 0515005.

【20】Villegas M, Paredes R, Juan A, et al. Face verification on color images using local features[C]. Proceedings of the 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2008: 1-6.

【21】Mehta R, Zhu. R. Blue or red? Exploring the effect of color on cognitive task performances[J]. Science, 2009, 323(5918): 1226-1229.

【22】Wu Y, Lim J, Yang M H. Online object tracking a benchmark[C]. Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2013: 2411-2418.

引用该论文

Liu Mengfei,Fu Xiaoyan,Shang Yuanyuan,Ding Hui. Pedestrian Tracking Based on HSV Color Features and Reconstruction by Contributions[J]. Laser & Optoelectronics Progress, 2017, 54(9): 091004

刘梦飞,付小雁,尚媛园,丁辉. 基于HSV颜色特征和贡献度重构的行人跟踪[J]. 激光与光电子学进展, 2017, 54(9): 091004

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF