Investigating approaches for privacy preserving Traffic surveillance

Aiming at the need of personal privacy protection in video surveillance, especially in the traffic surveillance, we discuss in this paper, approaches for preserving privacy for traffic surveillance. We propose three approaches so that the observer can only view a transformed form of surveillance to protect individual privacy. In the first approach, we utilize randomization based techniques using modular arithmetic that first scale the pixel value and then randomize it using prime numbers. Our first approach secure pixel value as well as preserves the privacy of citizens. However the first approach incurs more computational overhead. In our second approach, we use color model theory, i.e. Hue-Saturation-Value (HSV) model to hide person’s facial information to preserve privacy. We show that our second approach is faster than the first approach. However, the second approach is less secure. In the third approach, we apply edge based, i.e. horizontal and vertical edge detection on capture frame to hide person identity. In addition, a third approach is faster than the other two approaches and thus it can be very useful for real time traffic surveillance.