【作者】 丁卫； 龚振邦；

【Author】 DING Wei,GONG Zhen-bang (College of Mechatronics Engineering and Automation,Shanghai University,Shanghai 200072,China)

【机构】 上海大学机电工程与自动化学院；

【摘要】 提出一种基于动态聚类的地面移动目标低空跟踪的方法。该方法根据目标跟踪的实际情况选择分类特征,确定聚类准则函数,采用动态聚类方法,依次对像平面上的象素进行子窗口聚类,然后对子窗口内的象素进行运动区域聚类,最后对运动区域的象素进行目标聚类,从而比较精确地获得被跟踪目标的重心位置,并将其作为下一帧的聚类中心。由此,目标跟踪的过程变成一个动态聚类的过程。该方法是一个非监督学习方法,简单实用,通过仿真实验证明了该方法的有效性。更多还原

【Abstract】 A method of low altitude tracking ground moving object based on dynamic clustering is proposed. The characteristics of classification are chosen and the clustering rule functions are constructed according to the actual instance of object tracking.By dynamic clustering,some pixels in the image plane clustered in a sub-window,some pixels in the sub-window clustered into a motion region and the tracked object are divided from the motion region in the end.The center of gravity of the tracked object is the center of clustering in the next frame.The process of object tracking is a process of dynamic clustering.The method is an unsupervised learning means.The simulation results show that the method is effective.更多还原