Crime Scene Detection in Video Surveillance

Abstract

Among the real-time applications that are present, one of the challenging task is the tracking of the target in low resolution video as there will be loss of discriminatory information in the visual appearance of the objects that are moving.  Approaches that are already existing are mostly based on the development of a LR (low resolution) video along with the high resolution techniques. But they are the methods that require high costs for competition and this cost will continue to rise while dealing with more events. So through this paper an algorithm is introduced that can detect unusual event without making the type of conversions as above and is most suitable for the development of security systems for ATM where standard low-speed camera is used as they are of low cost.

Keywords: ATMs