基于颜色块重检的自适应抗遮挡目标跟踪算法

doi:10.3778/j.issn.1002-8331.2101-0313

摘要/Abstract

摘要： 为了解决目标跟踪过程中目标遇到遮挡物容易跟丢的问题，提出了基于颜色块重检的自适应抗遮挡目标跟踪算法。通过计算获取初始帧目标的颜色块，当发生遮挡时基于该颜色块信息对丢失帧进行颜色分割、形态学处理得到候选颜色块，并对候选颜色块进行匹配和定位，最终定位到实际目标。相比于LCT+的支持向量机检测器，基于颜色块的自适应目标重检测实现了对全局图像的目标重检测，有效避免了目标丢失的情况。在OTB50和OTB100上对所提算法的跟踪性能进行了评估，结果表明相比于LCT+和其他的主流跟踪算法，所提算法具有较好的抗遮挡性能。

关键词: 颜色块, 抗遮挡, 目标跟踪, 重检测, 颜色分割, 区域生长

Abstract: In order to solve the problem that the target is easy to be lost when encountering occlusions in the target tracking process, an adaptive anti-occlusion target tracking algorithm based on color block re-detection is proposed. Firstly, the color block information of the target intheinitial frame is obtained by calculation. When occlusion occurs, the candidate color block is obtained by color segmentation and morphological processing of the lost frame based on that color block information. Then, the candidate color block is matched and located, and finally the actual target is located.Compared with LCT+ support vector machine detector, the adaptive target re-detection of lost frames based on color blocks realizes the target re-detection of global image, effectively avoiding target loss. The tracking performance of the algorithm is evaluated on the OTB50 and OTB100. Experimental results show that the proposed algorithm has better performance in comparison with LCT+ and other mainstream tracking algorithms.

Key words: color block, anti-occlusion, target tracking, re-detection, color segmentation, region growing

陈富健, 谢维信. 基于颜色块重检的自适应抗遮挡目标跟踪算法[J]. 计算机工程与应用, 2022, 58(17): 189-198.

CHEN Fujian, XIE Weixin. Adaptive Anti-Occlusion Target Tracking Algorithm Based on Color Block Re-Detection[J]. Computer Engineering and Applications, 2022, 58(17): 189-198.

参考文献

[1] KALAL Z，MIKOLAJCZYK K，MATAS J.Tracking-learning-detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2011，34（7）：1409-1422.
[2] MA C，YANG X，ZHANG C，et al.Long-term correlation tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2015：5388-5396.
[3] MA C，HUANG J B，YANG X，et al.Adaptive correlation filters with long-term and short-term memory for object tracking[J].International Journal of Computer Vision，2018，126（8）：771-796.
[4] ZHAO J，XIAO G，ZHANG X，et al.An improved long-term correlation tracking method with occlusion handling[J].Chinese Optics Letters，2019，17（3）：36-41.
[5] HU Y，XIAO M，ZHANG K，et al.Aerial infrared target tracking in complex background based on combined tracking and detecting[J].Mathematical Problems in Engineering，2019：1-17.
[6] REDMON J，FARHADI A.Yolov3：an incremental improvement[J].arXiv：1804.02767，2018.
[7] WANG X，ZHANG K，LI S，et al.An optimal long-term aerial infrared object tracking algorithm with re-detection[J].IEEE Access，2019，7：114320-114333.
[8] LI Y，ZHU J，HOI S C H.Reliable patch trackers：robust visual tracking by exploiting reliable patches[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2015：353-361.
[9] 王任华，王本璇，孔军，等.全局与局部分块联合的目标跟踪算法[J].南京理工大学学报，2020，44（4）：462-470.
WANG R H，WANG B X，KONG J，et al.Target tracking algorithm via local-based and global-based collaborative modeling[J].Journal of Nanjing University of Science and Technology，2020，44（4）：462-470.
[10] DAI K，ZHANG Y，WANG D，et al.High-performance long-term tracking with meta-updater[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020：6298-6307.
[11] HENRIQUES J F，CASEIRO R，MARTINS P，et al.Exploiting the circulant structure of tracking-by-detection with kernels[C]//European Conference on Computer Vision.Berlin，Heidelberg：Springer，2012：702-715.
[12] POSSEGGER H，MAUTHNERT，BISCHOF H.In defense of color-based model-free tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2015：2113-2120.
[13] BERTINETTO L，VALMADRE J，GOLODETZ S，et al.Staple：complementary learners for real-time tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：1401-1409.
[14] WU Y，LIM J，YANG M H.Object tracking benchmark[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，37（9）：1834-1848.
[15] DANELLJAN M，H?GER G，KHAN F S，et al.Discrimi-native scale space tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2016，39（8）：1561-1575.
[16] DANELLJAN M，H?GER G，KHAN F，et al.Accurate scale estimation for robust visual tracking[C]//British Machine Vision Conference，Nottingham，September 1-5，2014.
[17] HU H，MA B，SHEN J，et al.Manifold regularized correlation object tracking[J].IEEE Transactions on Neural Networks and Learning Systems，2017，29（5）：1786-1795.
[18] 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，2014，37（3）：583-596.