Evolutionary Encryption Algorithm Ensuring Privacy in Video Surveillance

Abstract

The widespread use of video surveillance and the development of embedded information processing systems raise the issue of affecting the confidentiality and integrity of video data and even the individuals’ privacy. In order to secure the content of video and protect the public privacy, we propose a new system for detecting and encrypting regions of interest (ROIs). The ROI detection is based on an improved Gaussian mixture model where we use supervisors to control the changes in the image. The regions of interest encryption is performed through the use of an enhanced evolutionary encryption algorithm, exploiting the representation of the RGB color space to get real time system. Results and comparisons presented in this paper show that the proposed scheme can effectively detect and obscure the identity of people to achieve the privacy protection.