卡尔曼滤波在跟踪运动目标上的应用

在多关节机器人跟踪运动目标过程中,图像处理的主要内容包括图像的识别和目标轨迹的判定,需要在整个视觉空间进行搜索和匹配。这是一个典型的全局搜索问题,耗时较长、实时性较差,难以实现运动目标的快速跟踪。针对多关节机器人跟踪运动目标过程中存在的彩色图像数据量大、处理时间长,从而影响实时性的问题,采用线性卡尔曼滤波将全局搜索转化为局部搜索,减少搜索空间、加快匹配速度。通过实验可知,匹配时间由原来的157 ms降低到20 ms。实验结果证明了该方法的有效性。

Application of Kalman Filter in the Object Tracking

The image process mainly includes the shape recognition and the determination of the object tracking which demands to search and match on the whole vision space during the multi-joint robot to track the moving object.It is a type of global search problem which demands much time and it is difficult to accomplish the quick track.The difficulty of Object Tracking in the multi-joint robot is how to speed up the colorful digital image sequence model matching process which usually has huge information and takes a long time.In order to solve this problem,this paper takes the measure of liner Kalman filter to transfer the global searching scope to the local scope,thus reducing the scope space during model matching.Finally,the paper gives the experiment result,which proves this method is an effective one.