Efficient spatiotemporal segmentation and video object generation for highway surveillance video

The new MPEG-4 video coding standard enables content-based functionality by introducing the video object. It is based on segmentation and tracking of moving objects. A new procedure is described for spatiotemporal segmentation and video object generation of highway surveillance video. First, the representation model is proposed to describe the relationship between the background and the moving objects. Based on this model, we analyze the spatiotemporal information of some successive frames statistically to recover the background image. Then, the moving objects are detected and the video object planes are extracted by the difference between the video sequence and the recovered background image. Finally, the spatiotemporal regional similarity operator is used for object tracking. Based on the correspondence relationship between the video object planes, the video objects are generated successfully. At the same time, highway monitoring is implemented. The experimental results on an actual highway surveillance video show that our method is quite practical and efficient.