Video Surveillance using Deep Learning - A Review

Abstract

Video surveillance is a rapidly growing industry. Video surveillance, more commonly called CCTV (closed-circuit television), is an industry that is more than 30 years old and one that has had its share of technology changes.Video Surveillance has become an indispensable component for ensuring public safety in the modern world. Sophisticated video object tracking techniques specially designed for surveillance applications are of increasing importance for analyzing and understanding numerous surveillance videos in an effective manner. A large majority of video surveillance applications are concerned with monitoring activities within structured environments, such as indoor environments, surrounding areas of buildings, highways, traffic junctions, etc., the structures of which are often static and known to the surveillance personnel. One important characteristic of moving objects in these applications is that the motions of objects are constrained by the structure of the environment under surveillance. Therefore, it is beneficial and essential to explore the impacts of the environments upon the object motions, and integrate them into object tracking for improved performances.The problem of video surveillance has been well studied which has been adapted for several issues. The behavior of any human can be monitored through video surveillance. There are number of approaches available for the video surveillance and behavior analysis. The previous methods uses background models, object tracking for the problem of behavior analysis.Pervasive usage of video surveillance is rapidly increasing in developed countries. Continuous security threats to public safety demand use of such systems. Contemporary video surveillance systems offer advanced functionalities which threaten the privacy of those recorded in the video. There is a need to balance the usage of video surveillance against its negative impact on privacy.