基于角特征和Mean-shift的车辆跟踪方法  

Vehicle Tracking Method Based on Corner Feature and Mean-shift

作　　者：何得平[1] 朱光喜[1] 赵广州[1] 

机构地区：[1]华中科技大学武汉光电国家实验室,武汉430074

出　　处：《计算机工程》2010年第6期196-197,200,共3页Computer Engineering

摘　　要：提出一种基于角特征点和Mean-shift的车辆跟踪方法,对不断改变尺寸的车辆目标进行有效跟踪,利用角特征点把模板目标构造成若干个同尺寸的子窗口,每个子窗口单独地按Mean-shift算法进行跟踪,并寻找其相应的中心位置,通过关联过程确定车辆目标在整个视场中的运行轨迹。实验结果表明,该方法简单、可靠,可以稳定地跟踪车辆目标。A vehicle tracking method based on corner feature and Mean-shift is presented for efficient tracking object that continuously changes in size. The same in size＇s kernel windows of model object are built based on corner features. The Mean-shift algorithm is carried out one by one. According to the tracking result, the object＇s final position is estimated. Experimental results show this method is simple, reliable, and can track the vehicle stably.

关 键 词：车辆跟踪 Mean—shift算法 跟踪窗口 角特征点 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]