采用多边形位置校正的时空正则化相关滤波跟踪  

Polygon position correction based spatial-temporal regularization correlation filter tracking

作　　者：徐子钦 王涛 高赟[1] 张晋[2] Xu Ziqin;Wang Tao;Gao Yun;Zhang Jin(School of Information Science&Engineering,Yunnan University,Kunming 650504,China;Kunming Institute of Physical,Kunming 650223,China)

机构地区：[1]云南大学信息学院,昆明650504 [2]昆明物理研究所,昆明650223

出　　处：《计算机应用研究》2021年第9期2871-2875,共5页Application Research of Computers

基　　金：国家自然科学基金青年项目(61802337);国家自然科学基金地区项目(11663007)。

摘　　要：针对时空正则化相关滤波跟踪算法在背景混乱、光照变化等挑战场景下跟踪结果与正确目标位置总是存在偏离的问题,提出了采用多边形位置校正的时空正则化相关滤波跟踪算法,以提高跟踪结果的准确性和鲁棒性。该算法采用峰值噪声比对当前跟踪结果的置信度进行判别,对置信度判别结果为不可信的情况采用多边形位置校正方法对当前跟踪结果进行修正。实验表明,该算法在成功率曲线下面积和精度两种评价标准下都有较大的提升,尤其是在背景混乱和光照变化场景下都具有更好的跟踪效果,具有一定的应用价值。Aiming at the problem that the tracking result of spatial-temporal regularization correlation filter tracking always deviates from the correct target position in challenging scenarios such as background clutter and illumination variation,this paper proposed polygon position correction based spatial-temporal regularization correlation filter tracking to improve the accurate and robust of tracking results.The algorithm used the peak-to-noise ratio to discriminate the confidence of the current tracking result and used the polygon position correction method to correct the current tracking result when the confidence determination result was unreliable.The experiment shows that this algorithm has a great improvement in precision and the area under the success rate curve,especially in the background clutter and illumination variation has better tracking results,and has a certain application price.

关 键 词：目标跟踪 相关滤波 置信度判别 多边形位置校正 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]