自适应预测压缩跟踪.pdfVIP

第 43 卷第 12 期 光电工程 Vol.43, No.12 2016 年 12 月 Opto-Electronic Engineering Dec, 2016 文章编号：1003-501X(2016)12-0092-07 自适应预测压缩跟踪 吴腾飞，江艳霞，刘子渊，仲思凯 ( 上海理工大学 光电信息与计算机工程学院自动化系，上海 200093 ) 摘要：为解决压缩跟踪过快引入跟踪误差和样本采集方面的缺陷，首先，引入基于预测向量的采样搜索策略，通 过前两帧跟踪到的目标位置预测后一帧目标的运动方位，并采用扇形区域采样方式缩小有效样本的范围；其次， 根据前后两帧跟踪到的目标的对照来判断复杂背景或遮挡的发生，利用 Bhattacharyya 系数自适应地改变分类器参 数更新系数。实验证明，这些策略避免了因压缩跟踪缺陷导致的跟踪失败，改进后的算法比原算法具有更好的鲁 棒性和时效性。 关键词：压缩跟踪；预测向量；自适应；贝叶斯分类 中图分类号：TP391.41 文献标志码：A doi：10.3969/j.issn.1003-501X.2016.12.015 Adaptive and Predictive Compressive Tracking WU Tengfei ，JIANG Yanxia ，LIU Ziyuan ，ZHONG Sikai ( Department of Automation, School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China ) Abstract : To deal with the defects of Compressive Tracking (CT) in the tracking error and the sample collection, firstly, the predictive vector was introduced to search samples that can direct motion of the target. Then the fan-shaped sampling areas reduced the amount of computation greatly. Furthermore, we could determine complex background or occlusion through comparison of the neighboring target images, and then update the classifier parameters automatically by applying the Bhattacharyya coefficient. Experiment shows that these improvements can avoid the failure of compressive tracking and the adaptive predictive compressive tracking (VACT) is better than the original algorithm (CT) in robustness and speed. Key words : compressive tracking; predictive vector; self-adaptive; Bayes discrim