A framework for privacy assurance in a public video-surveillance system

Video surveillance systems can be found today almost in any public areas and, on one hand, they are used to ensure security. On the other hand, most of the time there are no security threats. In this situation, the system is actually violating our right to privacy without assuring any security. This paper presents a framework for privacy assurance in a public video-surveillance system. A new processing block is introduced: the privacy enforcer. The privacy enforcer automatically detects in real-time faces of people in a video recording and ensures privacy by scrambling the pixels position around the head. The scrambling is done by a three-step circular pixel shift algorithm. The scrambling is reversible in order to recover the original recording if a security threat is detected.