一种相关滤波中的模型自适应更新技术
An Adaptive Model Updating Technology in Correlation Filtering
摘要: 针对单一更新策略难以适应复杂多变的目标跟踪环境的问题，本文在ECO (高效卷积算子)算法基础之上，提出了一种基于旁瓣比的自适应策略更新跟踪算法。该算法通过对特征响应图中峰值旁瓣比的分析，实现了跟踪模型更新策略的自适应调整。同时使用OTB-100数据集对改进后的跟踪算法进行测试，经测试后的实验结果表明，相比原来的ECO算法，该算法可提升遮挡等复杂环境下目标跟踪的精度和鲁棒性。
Abstract: Aiming at the problem that a single update strategy is difficult to adapt to the complex and changeable target tracking environment, an adaptive strategy update tracking algorithm based on side lobe ratio is proposed based on ECO (efficient convolution operators) algorithm. By analyzing the peak to side lobe ratio in the characteristic response graph, the algorithm realizes the adaptive adjustment of the tracking model update strategy. At the same time, the OTB-100 data set is used to test the improved tracking algorithm. The experimental results show that compared with the original ECO algorithm, the algorithm can improve the accuracy and robustness of target tracking in complex environments such as occlusion.
文章引用：江山, 王录涛, 王奇, 文成江. 一种相关滤波中的模型自适应更新技术[J]. 计算机科学与应用, 2022, 12(2): 436-447. https://doi.org/10.12677/CSA.2022.122044

参考文献

[1] 张微, 康宝生. 相关滤波目标跟踪进展综述[J]. 中国图象图形学报, 2017, 22(8): 1017-1033.
[2] 刘航. 复杂背景下基于相关滤波的目标跟踪算法研究[D]: [硕士学位论文]. 湘潭: 湘潭大学, 2020.
[3] 葛宝义, 左宪章, 胡永江. 视觉目标跟踪方法研究综述[J]. 中国图象图形学报, 2018, 23(8): 1091-1107.
[4] 许倩倩. 基于卷积神经网络多特征融合的相关滤波目标跟踪算法研究[D]: [硕士学位论文]. 杭州: 浙江工业大学, 2020.
[5] Lu, H.C., Li, P.X. and Wang, D. (2018) Visual Object Tracking: A Survey. Pattern Recognition and Artificial Intelligence, 31, 61-76.
[6] Danelljan, M., Robinson, A., Shahbaz Khan, F. and Felsberg, M. (2016) Beyond Correlation Filters: Learn-ing Continuous Convolution Operators for Visual Tracking. In: Leibe, B., Matas, J., Sebe, N. and Welling, M., Eds., ECCV 2016, Lecture Notes in Computer Science, Vol. 9909, Springer, Cham. [Google Scholar] [CrossRef] 
[7] Danelljan, M., Bhat, G., Shahbaz Khan, F., et al. (2017) ECO: Efficient Convolution Operators for Tracking. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 6638-6646. [Google Scholar] [CrossRef] 
[8] Real, E., Shlens, J., Mazzocchi, S., et al. (2017) Youtube-Bounding Boxes: A Large High-Precision Human-Annotated Data Set for Object Detection in Video. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 5296-5305. [Google Scholar] [CrossRef] 
[9] 谢悦. 基于相关滤波的抗遮挡目标跟踪算法研究[D]: [硕士学位论文]. 长沙: 湖南大学, 2020.
[10] 张思贤, 杨艺, 张猛, 米鹏博. 高效的多特征自适应相关滤波跟踪器[J/OL]. 系统仿真学报, 1-11. http://kns.cnki.net/kcms/detail/11.3092.v.20210916.1113.006.html, 2021-12-03.
[11] 张飞, 马时平, 张立朝, 何林远, 仇祝令, 韩永赛. 空间可靠性和相关滤波器联合学习的跟踪算法[J]. 西安电子科技大学学报, 2021, 48(5): 167-177. [Google Scholar] [CrossRef] 
[12] 武哲纬, 周世杰, 刘启和. 运算资源受限环境下的目标跟踪算法综述[J]. 计算机工程与应用, 2021, 57(21): 24-40.
[13] 王明斌. 基于相关滤波的视觉目标跟踪算法研究[D]: [硕士学位论文]. 北京: 北京交通大学, 2020.[CrossRef] 
[14] 黄凯奇, 任伟强, 谭铁牛. 图像物体分类与检测算法综述[J]. 计算机学报, 2014, 37(6): 1225-1240.
[15] 刘巧元, 王玉茹, 张金玲, 殷明浩. 基于相关滤波器的视频跟踪方法研究进展[J]. 自动化学报, 2019, 45(2): 265-275. [Google Scholar] [CrossRef] 
[16] 朱明敏, 胡茂海. 基于相关滤波器的长时视觉目标跟踪方法[J]. 计算机应用, 2017, 37(5): 1466-1470.
[17] 李胜杰. 面向实时鲁棒目标跟踪的深度粒子相关滤波算法研究[D]: [硕士学位论文]. 北京: 北京邮电大学, 2020.[CrossRef] 
[18] Zhang, J.M., Ma, S.G. and Sclaroff, S. (2014) MEEM: Robust Tracking via Multiple Experts Using Entropy Minimization. 2014 European Conference on Computer Vision, Zurich, 6-12 September 2014, 188-203. [Google Scholar] [CrossRef] 
[19] Ma, C., Huang, J.B., Yang, X.K., et al. (2015) Hierarchical Convolutional Features for Visual Tracking. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 7-13 December 2015, 3074-3082. [Google Scholar] [CrossRef] 
[20] Bertinetto, L., Valmadre, J., Golodetz, S., et al. (2016) Staple: Com-plementary Learners for Real-Time Tracking. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 1401-1409. [Google Scholar] [CrossRef] 
[21] Li, Y., Zhu, J.K., Hoi, S.C.H., et al. (2017) Robust Estimation of Similarity Transformation for Visual Object Tracking.