基于特征自适应选择的鲁棒跟踪算法.pdfVIP

第29 卷 第12 期 控 制 与 决 策 2014 年 12 月 Vol. 29 No. 12 Control and Decision Dec. 2014 文章编号: 1001-0920 (2014) 12-2137-07 DOI: 10.13195/j.kzyjc.2013.1537 基于特征自适应选择的鲁棒跟踪算法 吴成东, 齐苑辰, 陈东岳 (东北大学信息科学与工程学院，沈阳110004) 摘 要: 为了解决真实场景下视频目标的跟踪问题, 提出一种基于特征自适应选择的鲁棒跟踪算法. 首先, 针对在线 AdaBoost 算法特征池特征鲁棒性差的问题, 提出一种基于颜色与金字塔梯度方向直方图特征相结合的特征池构造 方式; 然后, 针对分类器在更新过程中容易受到错误样本影响的问题, 对每帧跟踪结果增加遮挡检测环节以避免漂移 现象的发生. 大量的对比实验表明, 在真实场景下所提出的方法具有更好的效果. 关键词: 目标跟踪；在线学习；自适应更新；遮挡检测 中图分类号: TP391 文献标志码: A Robust object tracking with adaptive feature selection WU Cheng-dong, QI Yuan-chen, CHEN Dong-yue (College of Information Science and Engineering，Northeastern University ，Shenyang 110004，China. Correspondent: QI Yuan-chen，E-mail ：qiyuanchen649@163.com) Abstract: In order to solve the tracking problem of video sequences in real-world scenarios, a robust tracking algorithm based on adaptive feature selection is proposed. Firstly, for the problem that the candidate features of the online AdaBoost algorithm are not robust, a construction mode of the candidate feature pool is proposed, which combines color and pyramid gradient orientation histogram features. Then, for the problem that classifiers are vulnerable to the influence of improper samples during the update, a process of occlusion detection is added at each frame after obtaining the tracking result to avoid the phenomena of drift. Lots of comparison experiments show that the proposed algorithm tracks the object accurately and reliably in realistic videos. Key words: object tracking ；online learning ；adaptive updating ；occlusion detection 0 引引引 言言言 的. 因为本文更关注于对目标观测模型的构建与更新