首页 > 论文 > 激光与光电子学进展 > 56卷 > 3期(pp:30401--1)

基于相关滤波器的目标抗遮挡算法

Target Anti-Occlusion Algorithm Based on Correlation Filter

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

摘要

相关滤波目标跟踪算法是基于目标模板与待测图像之间的相关性大小来实现目标的定位与跟踪, 核相关滤波器(KCF)的提出更将其推向了新的高度。然而, 通过对KCF算法的深入研究发现, 相关滤波器在抗遮挡性能方面有着严重的不足, 尤其是在目标短暂消失的情况下十分容易出现跟踪丢失的情况。为了解决这个问题, 提出了一种将KCF与前后向误差检测算法相结合的方法, 通过前后向误差算法检测遮挡现象, 并在遮挡发生后及时保留原目标模板, 最后进行小范围的预测并结合原模板重新定位目标位置。实验表明, 此方法能有效解决目标短暂完全消失的遮挡状况, 并在目标重新出现后进行有效的追踪。

Abstract

Correlation filtering target tracking algorithm is based on the correlation between the target template and the image to be tested to realize the location and tracking of the target. Especially, when the kernelized correlation filters (KCF) is proposed which is faster and more accurate, the correlation filtering target tracking algorithm is pushed to a new height. However, with in-depth study of the KCF algorithm, it is found that the correlation filter has some serious shortcomings in the anti-occlusion performance. Especially in the case of a short-term disappearance of the target, it is extremely easy to lost the target. In order to solve this problem, a method combining KCF with the forward and backward error detecting algorithm is proposed. It detects the occlusion phenomenon by the forward-backward error algorithm, and retains the original target template in time after the occlusion occurs. Finally, it re-locates the position of the target by combing the prediction in a small range with the original template. Experimental results show that this method can effectively solve the occlusion condition when the target disappears completely and perform effective tracking after the target reappears.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP3-05

DOI:10.3788/lop56.030401

所属栏目:探测器

基金项目:国家自然科学基金(61502203)、江苏省自然科学基金(BK20150122)、江苏省高等学校自然科学研究面上项目(17KJB520039)

收稿日期:2018-07-06

修改稿日期:2018-07-25

网络出版日期:2018-08-11

作者单位    点击查看

王凯宇:江南大学物联网工程学院, 江苏 无锡 214122
陈志国:江南大学物联网工程学院, 江苏 无锡 214122
傅毅:江南大学物联网工程学院, 江苏 无锡 214122无锡环境科学与工程研究中心, 江苏 无锡 214153

联系人作者:陈志国(427533@qq.com)

【1】Wu X J, Xu T Y, Xu W B. Review of target tracking algorithms in video based on correlation filter[J]. Command Information System and Technology, 2017, 8(3): 1-5.
吴小俊, 徐天阳, 须文波. 基于相关滤波的视频目标跟踪算法综述[J]. 指挥信息系统与技术, 2017, 8(3): 1-5.

【2】Baker S, Matthews I. Lucas-kanade 20 years on: A unifying framework[J]. International Journal of Computer Vision, 2004, 56(3): 221-255.

【3】Nguyen H T, Smeulders A W M. Fast occluded object tracking by a robust appearance filter[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(8): 1099-1104.

【4】Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using mean shift[C]∥Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 2000: 142-149.

【5】Babenko B, Yang M H, Belongie S. Visual tracking with online Multiple Instance Learning[C]∥Computer Vision and Pattern Recognition, 2009: 983-990.

【6】Kalal Z, Matas J, Mikolajczyk K. P-N learning: Bootstrapping binary classifiers by structural constraints[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010: 49-56.

【7】Hare S, Saffari A, Torr P H S. Struck: Structured output tracking with kernels[C]∥International Conference on Computer Vision, 2011: 263-270.

【8】Mei X, Ling H. Robust visual tracking using L1 minimization[J]. IEEE International Conference on computer Vision. Kyoto: IEEE, 2009: 1436-1443.

【9】Bolme D S, Beveridge J R, Draper B A, et al. Visual object tracking using adaptive correlation filters[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010: 2544-2550.

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

【11】Xiang Y, Alahi A, Savarese S. Learning to track: online multi-object tracking by decision making[C]∥IEEE International Conference on Computer Vision (ICCV), 2015: 4705-4713.

【12】Danelljan M, Hger G, Shahbaz Khan F, et al. Accurate scale estimation for robust visual tracking[C]∥Proceedings of the British Machine Vision Conference 2014, British Machine Vision Conference, 2014.

【13】Li Y, Zhu J K. A scale adaptive kernel correlation filter tracker with feature integration[M]∥Cham: Springer International Publishing, 2015: 254-265.

【14】Bertinetto L, Valmadre J, Golodetz S, et al. Staple: complementary learners for real-time tracking[C]∥IEEE Computer Vision and Pattern Recognition, 2016: 1401-1409.

【15】Yan H, Zhang Y, Yang X L, et al. A kernelized correlation filters with occlusion handling[J]. Journal of Optoelectronics·Laser, 2018, 29(6): 647-652.
闫河, 张杨, 杨晓龙, 等. 一种抗遮挡核相关滤波目标跟踪算法[J]. 光电子·激光, 2018, 29(6): 647-652.

【16】Bao X A, Zhan X J, Wang Q, et al. Anti occlusion target tracking algorithm based on KCF and SIFT features[J]. Computer Measurement & Control, 2018, 26(5): 148-152.
包晓安, 詹秀娟, 王强, 等. 基于KCF和SIFT特征的抗遮挡目标跟踪算法[J]. 计算机测量与控制, 2018, 26(5): 148-152.

【17】Lü G J, Mao X, Hu Y J, et al. A method of detecting and tracking occlusion based on comparison of forward and backward errors[J]. Infrared technology, 2016, 38(4): 337-341, 347.
吕高杰, 毛鑫, 胡银记, 等.一种基于前后向误差比较的检测跟踪遮挡方法[J]. 红外技术, 2016, 38(4): 337-341, 347.

【18】Kalal Z, Mikolajczyk K, Matas J. Forward-backward error: automatic detection of tracking failures[C]∥20th International Conference on Pattern Recognition, 2010: 23-26.

【19】Wan L, Bai H L, Dai J. Extended optimal Otsu thresholding method of image segmention[J]. Journal of Harbin Engineering University, 2003, 24(3): 326-329.
万磊, 白洪亮, 戴军. 扩展的Otsu最优阈值图像分割的实现方法[J].哈尔滨工程大学学报, 2003, 24(3): 326-329.

引用该论文

Wang Kaiyu,Chen Zhiguo,Fu Yi. Target Anti-Occlusion Algorithm Based on Correlation Filter[J]. Laser & Optoelectronics Progress, 2019, 56(3): 030401

王凯宇,陈志国,傅毅. 基于相关滤波器的目标抗遮挡算法[J]. 激光与光电子学进展, 2019, 56(3): 030401

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