Computer Vision to Improve Security Surveillance through the Identification of Digital Patterns

Abstract

The need to have good security, either in the streets, at home or at workplaces, cannot be overemphasized. Due to its significance, security experts continue to improve the mechanism and the tools used to manage security issues. The revolution in computing and information technology has significantly affected how people deal with security. Computer vision application has been developed for security purposes, especially, by improving surveillance systems. Computer vision can manage face detection, motion detection, person identification, tracking, access control, and interpretation of movement. Most of these tasks can be used to improve security surveillance. Computer integrated systems can be used to identify strange behavior and aid security management. Forensic science involves the analysis of images to establish patterns. As a result, strategies aimed at utilizing computer vision focus on the improvement of image computation power of computer systems. Surveillance processes based on the use of cameras involve different phases: environmental design, discovery of movements, analysis and description of behaviors, organizing objects in motion, tracing as well as discovery of individuals. This paper seeks to analyze how computer systems can be trained to identify digital patterns in order to help in surveillance processes including tracking of strange behavior and crime.