基于分块颜色直方图的MS 跟踪算法

中国民航大学学报

基于分块颜色直方图的MS 跟踪算法

张红颖，王赛男，胡文博

（中国民航大学电子信息与自动化学院，天津300300）

收稿日期:2018-01-18 修回日期:2018-03-02 出版日期:2019-02-25 发布日期:2019-04-16
作者简介:张红颖（1978—），女，天津人，副教授，博士，研究方向为图像工程与计算机视觉.
基金资助:
国家自然科学基金项目（U1533203）

MS tracking algorithm based on block color histogram

ZHANG Hongying, WANG Sainan, HU Wenbo

（College of Electronic Information and Automation, CAUC, Tianjin 300300, China）

Received:2018-01-18 Revised:2018-03-02 Online:2019-02-25 Published:2019-04-16

摘要/Abstract

摘要： 针对复杂场景下的目标遮挡问题，提出了一种基于分块颜色直方图的均值漂移（MS）跟踪算法。为了得到较强的局部特征表达能力，改进算法采用分块颜色直方图，将目标的整体区域颜色特征划分成几个局部区域的颜色特征，有效提高了算法的抗遮挡能力。为了准确地跟踪目标，提出了一种分块颜色直方图的置信图生成方式，使MS 算法在置信图迭代中寻找最终目标区域。实验证明，该算法在目标发生遮挡时，仍能持续准确地实现目标跟踪，相比于传统MS 跟踪算法性能更优。

关键词: 分块颜色直方图, 置信图, MS, 目标跟踪

Abstract: Aiming at the problem of object occlusion in complex scenes, an improved tracking algorithm based on block color histogram and MS is proposed. First, in order to obtain strong ability to express local features, block color histogram is utilized. The target is divided into several regions to color features. Then a confidence map generating method of the block color histogram is proposed, helping MS algorithm search the best target area in the iteration of confidence map. Experimental results prove that the proposed algorithm can keep tracking the target continuously and accurately even though the target is occluded, which improves the robustness of the algorithm compared with traditional MS tracking algorithm.

Key words: block color histogram, confidence map, MS, object tracking

中图分类号:

TP391

张红颖，王赛男，胡文博. 基于分块颜色直方图的MS 跟踪算法[J]. 中国民航大学学报.

ZHANG Hongying, WANG Sainan, HU Wenbo. MS tracking algorithm based on block color histogram[J]. .

参考文献 11

[1]	高文, 朱明, 贺柏根, 等. 目标跟踪技术综述[J]. 中国光学, 2014,7(3): 365-375.
[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]	COMANICIU D, RAMESH V, MEER P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence,2003, 25(5): 564-575.
[4]	张红颖, 胡正. 融合局部三值数量和色度信息的均值漂移跟踪[J].电子与信息学报, 2014, 36(3): 624-630.
[5]	LI XIAOHE, ZHANG TAIYI, SHEN XIAODONG, et al. Object tracking using an adaptive Kalman filter combined with mean shift[J]. Optical Engineering, 2010, 49(2): 020503-1-020503-3.
[6]	孙中森, 孙俊喜, 宋建中, 等. 一种抗遮挡的运动目标跟踪算法[J].光学精密工程, 2007, 15(2): 267-271.
[7]	王长有, 刘皓, 张海强, 等. 基于灰色预测和Mean-Shift 的抗遮挡跟踪算法[J]. 控制工程, 2017, 24(7): 1323-1328.
[8]	DALAL N, TRIGGS B. Histograms of Oriented Gradients for Human Detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR忆05). Washington, DC, USA, 2005,1: 886-893.
[9]	张雷, 于凤芹. 基于置信图特性的改进时空上下文目标跟踪[J].计算机工程, 2016, 42(8): 277-281, 288.
[10]	肖鹏, 段洣毅, 赵琪, 等. 基于置信图自适应融合的视觉目标跟踪[J]. 无线电工程, 2013, 43(9): 20-23.
[11]	WU YI, LIM JONGWOO, YANG MINGHSUAN. Online Object Tracking: A Benchmark[C]//2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland, OR, USA, 2013: 2411-2418.