摘要
目前,在车辆跟踪识别技术中车辆间相互遮挡问题是一项具有挑战性的任务。就此问题提出了基于Kalman预测跟踪矩阵与形状匹配细分割的跟踪算法。该方法首先基于Kalman滤波对车辆进行粗跟踪配对,然后对发生遮挡的目标跟踪检测区域进行形状匹配的细分割。从实验仿真结果可以看出,该算法能够实现预期的遮挡情况下的视频车辆跟踪的目标。
At present, the vehicle identification technology in vehicle tracking mutual occlusion is a challenging task. This issue was proposed based on Kalman tracking matrix and shape matching predicted fine segmentation and tracking algorithm. In this method, Kalman filtering based on rough tracks matching the vehicle, and then place the target tracking and detection block shape matching region of fine segmentation. It can be seen from the experimental results, the algorithm can achieve the desired occlusion of video vehicle tracking targets.
出处
《电脑开发与应用》
2013年第2期9-12,共4页
Computer Development & Applications
关键词
智能交通
车辆跟踪
形状匹配
intelligent transportation, vehicle tracking, shape matching
