摘要
为了解决传统目标跟踪算法在目标发生旋转、变形等情况下容易丢失的缺陷,提出了首先使用Harris角点算法提取几个目标角点特征,光流法进行匹配点配准,最后采用重心算法确定目标脱靶量的跟踪方法.利用了角点提取算法具有运算速度快且能代表目标形状特征的特点以及光流配准具有匹配精度高的优点.实验结果表明,该方法具有较高的跟踪精度,对于目标旋转、变形、遮挡都具有很高的鲁棒性.
In order to solve target lost shortcomings of the traditional target tracking system because of target＇s rotation, transformation etc, a kind of new tracking method is proposed. This method uses Harris corner detection algorithm to draw several typical target corner features and then uses optic flow algorithm to match the corresponding corner point. In the end, the target displacement is computed by bary centre algorithm. The proposed method uses the property of corner extracting algorithm＇s high speed and representing target＇s shape and optic flow match＇s high resolution. It is proved by test that this method arrives at high tracking precision and has the robust feature to target＇s rotation, transformation, shelter,etc.
出处
《电子器件》
CAS
2007年第4期1397-1399,共3页
Chinese Journal of Electron Devices
关键词
角点检测
光流法
目标跟踪
点配准
corner detection algorithm
optic flow
target tracking
point match
