Abnormal Crowd Behavior Detection Using Image Processing

Crowd Management and Abnormal crowd behavior detection is an important research topic in computer vision. There are many instances in which thousands of people are killed due stampedes, panic etc. Abnormal Crowd Behavior Detection is not only needed research topic but also the direct help to saving these lives. However, there is not much of research done on the prediction and detection of such Abnormal Behavior Detection. Our system will be able to detect abnormal crowd behavior using Motion History Image and Optical Flow technique. Optical flow is used to predict the magnitude and direction of velocity which is later used with a predefined threshold to classify the video.