Crowd Behaviour Analysis

Abstract: In today's scenario, video surveillance plays an important role. Leading it to be most popular in security applications. Here the question arises what makes it more popular? The answer o this question is its ability to recognize abnormal behavior in a video sequence and further analysis of their behavior has drawn the attention towards the popularity of video surveillance, since it allows filtering out a large number of useless information, which guarantees the high efficiency in the security protection, and save lots of human and material resources. In this paper, we present crowd behavior analysis i.e. this paper provides a video surveillance framework for event recognition in a crowded scene to detect the abnormal human behavior. We proposed a human tracking method based on the Lucas-Kanade optical flow algorithm. This optical method provides the feature for object detection and tracking know as optical flow vector. Keywords: Crowd; Motion detection; Optical flow estimation; Motion vector; Object velocity and direction