摘 要：核相关滤波跟踪方法作为一种效果较佳的目标跟踪方法得到了广泛应用。本文介绍了基于前景和基于背景的两类目标检测的特点，以及目标跟踪的相关知识。重点以核相关滤波跟踪方法为主，详细介绍了该方法的特点、算法实现，包括相关滤波、构建岭回归模型、构造循环样本和核函数映射，讨论了其性能优势以及算法设计开发中的数据集。多通道和输入梯度直方图等特征使核相关滤波跟踪方法获得了效率高、速度快、识别能力强、稳定性高的特点。
关键词：核相关滤波；目标跟踪与检测；核函数；岭回归模型
中图分类号：TP391.41 文献标识码：A 文章编号：2096-4706（2019）10-0014-03
Research of Kernel Correlation Filtering Tracking
MENG Xiangrui
（School of Computer Science and Technology，Beijing Institute of Technology，Beijing 100081，China）
Abstract：As a better method of generating target tracking，kernel correlation filter tracking has been widely used. This paper introduces the characteristics of two kinds of target recognition based on foreground and background，and the related knowledge of target tracking. This paper mainly focuses on of kernel correlation filtering tracking. The characteristics and algorithm implementation of this method are introduced in detail，including correlation filtering，ridge regression model，cyclic samples and kernel function mapping. Its performance advantages and data sets in algorithm design and development are discussed. The features of multi-channel and input gradient histogram make the kernel correlation filtering tracking obtain high efficiency，fast speed，strong recognition ability and high stability.
Keywords：kernel correlation filtering；target tracking and detection；kernel function；ridge regression model
参考文献：
[1] Yilmaz，Alper，Javed，Omar，Shah，Mubarak. Object tracking：A survey [J]. ACM Computing Surveys，2006，38（4）：13.
[2] 尹宏鹏，陈波，柴毅，等. 基于视觉的目标检测与跟踪综述 [J]. 自动化学报，2016，42（10）：1466-1489.
[3] João F. Henriques，Caseiro R ，Martins P ，et al. Exploiting the Circulant Structure of Tracking-by-Detection with Kernels [M]//Computer Vision–ECCV 2012. Springer Berlin Heidelberg，2012.
[4] DANELLJAN M，KHAN FS，FELSBERG M，et al. Adaptive Color Attributes for Real-Time Visual Tracking [C]// Computer Vision and Pattern Recognition （CVPR），2014 IEEE Conference on.S.l.：s.n.，2014：1090-1097.
[5] Kaihua Zhang，Lei Zhang，Qingshan Liu，et al. Fast Visual Tracking via Dense Spatio-temporal Context Learning [M]. Cham：Springer International Publishing，2014：127-141.
[6] Henriques J F，Rui C，Martins P，et al. High-Speed Tracking with Kernelized Correlation Filters [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence，2015，37（3）：583-596.
[7] Bernhard Schölkopf. Advances in Kernel Methods-Support Vector Learning [M]// Advances in kernel methods：support vector learning. MIT Press，1999.
[8] HOERL A E，KENNARD R W. Ridge regression：Biased estimation for nonorthogonal problems [J]. Technometrics，2000，42（1）：80-86.
[9] Bolme D S，Beveridge J R，Draper B A，et al. Visual object tracking using adaptive correlation filters [C]// Computer Vision&Pattern Recognition. 2010.
[10] Wu Y，Lim J，Yang M H . Online Object Tracking：A Benchmark [C]// 2013 IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society，2013.
[11] Wu Y，Lim J，Yang M H. Object Tracking Benchmark [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence，2015，37（9）：1834-1848.
作者简介：孟祥瑞（1997.10-），男，汉族，河北张家口人，本科在读，研究方向：图像处理、人工智能。