Detection of Criminal Activities/Criminal through CCTV by analysing live footage for melee, mob formation, body language of suspect etc

Abstract : In this paper, we describe a surveillance program which is to be designed that can automatically detect the gestures or signs of aggression and brutality on real time. A single CCTV human operator can handle very limited set of operations, so as the number of CCTV camera increases the need of the human intervention also increases which can also cause human errors so automation for certain detection operations is necessary. Basically, our proposed system consists of 2 main modules which are capable of detecting the actions of objectional objects and humans in the frame for example gun and knife. Here in this project, we propose algorithms which are able to make the people attentive about (1) Presence of a any hazardous act, the danger is detected when the objectional object appears in the frame with the presence of human. (2) An abnormal activity of human when they be handling the weapons or acts of assaults, In this project, we focus to allow the real time application, we completely focus on reducing the number of negative alarms.
 EXISTING SYSTEM :
 ? The existing system of the CCTV surveillance is done by the human operator and also the automated system of CCTV surveillance is not that accurate for making decision and give responses as per the bizarre situation. ? Our main objective is to develop an intelligent Monitoring Model system and can take over the existing passive CCTV surveillance system which proves to be inefficient and ineffective as, the number of CCTV increase the number of human operator or human intervention to observe the system also increases.
 DISADVANTAGE :
 ? The passive system is used only in the military bases but not metro stations, airports and mall etc. It is a complex architecture. In passive system there is not individual safety level considered. ? Basically, the video in passive system is captured continuously which can lead to privacy issue. Also, the system does not have efficient alarm system that responses in return of any act of abuse or bullying. ? There is nor blurring out of the sensitive information in present system.
 PROPOSED SYSTEM :
 • Our current model consists of 2 detection functions, one for detecting crimes which are done when there are little movements detected (e.g., Robbery or people armed with weapons) and another one for detecting crimes that committed with the heavy or large movements. • To avoid the negative/false alarms detection, we have executed a system where the already stored detections are all cleared and only the recently detected. • In our particular solution, there is no such social credit system that keeps permanent log of all the activity, So in this there is no need of continuous monitoring and assessment.
 ADVANTAGE :
 ? The major advantage of the project includes efficiency, fast to access and uniqueness. CCTV surveillance is thought to be very simple and at the same time challenging, raising doubts about its efficacy and utility. ? CCTVs vastly increase the efficiency of police units in the field in terms of providing precise technical intelligence to solve the crime and also act as a deterrent. ? This is also seen as a part of evidence-based policing since it can provide the means to test the efficacy of police deployment and surveillance to control crime.

We have more than 145000 Documents , PPT and Research Papers

Have a question ?

Mail us : info@nibode.com