A OUTLINE OF SUSPICIOUS AND VIOLENT ACTIVITY DETECTION OF HUMANS
D Karthiga, M. Selvam · Journal of Emerging Technologies and Innovative Research · 2020
: Crimes like stealing, violence against folks, injury to property, etc., became quite common in a very society, that a significant concern. The normal police work systems act like post mortem tools within the sense that they'll be used for the investigation to detect the person behind the theft, but it is only after the crime has already occurred. During this chapter, we tend to propose a technique for mechanically detection the suspicious or violent activities of an individual from the police work video. We tend to train the SVM classifier with the HOG options extracted from the video frames of two types: frames showing no violent activities and people showing violent activities like kicking, pushing, punching, etc. In the testing phase, the frames from the surveillance video square measure browse and processed so as to classify them as violent or traditional frames. If the frames classified as violent frames square measure detected, associate alarm is raised to alert the controller. It will be wont to keep track of the time period that an individual is found loitering at a place being monitored. If the time exceeds a predefined threshold, the alarm is raised to alert regarding any potential suspicious activity in order that it will be checked on time.