Published Paper Details:

Abnormal detection refers to as a result of unusual or exceptional striking activity that may reveal a person or group of people to danger in particular context. There has been an increase in the number of offensive or disruptive activities recently. Because of this, security has been given the utmost importance. If a single person is watching a live CCTV screen more than 20 minutes it can leads to loss 95% of his/her eye vision. Human monitoring of video is very laborious task and requires a work force and their constant attention. To automate this process here we are creating a computer vision-basedsurveillance system to notify users. Video surveillance system plays an important role in keeping indoors and outdoors environment secure. Various surveillance methods were introduced for security purposes. The current advancement in computer vision and machine learning has outstanding role in enabling such intelligent framework. Different algorithms that are specially designed for building smart vision framework seek to understand and build correct semantic inference from observe motion cause by target. This paper addresses an overview most recent research algorithm that focuses on abnormal behavior detection in surveillance application.In this paper we have used YOLO technique and TensorFlow's MoViNettechnique to detect these abnormal activities.