Crowd movement segmentation using velocity field histogram curve
Wei Li, Jiuhong Ruan, Hua-An Zhao · 2012
This paper presents a fast, accurate and novel approach for the problem of flow segmentation in dense and very dense moving crowds. First, optical flow method is used to remove background noise of scenes. Second, the angle information of foreground velocity field is turned into gray level image and histogram curve is employed to find out extreme points. Finally, the updated minimum points can be utilized to segment foreground crowd into different flows. Compared to other methods, the proposed approach is efficient and fast, the processing speed is 17 frames per second.