Key Generation and Verification for Image Authentication

We present a novel approach using distributed source coding for image authentication. The key idea is to provide a Slepian?Wolf encoded quantized image projection as image authentication data. This can be correctly decoded with the help of an authentic image using as side information. Distributed source coding provides the desired robustness against legitimate variations while detecting illegitimate modification. The decoder incorporating expectation maximization algorithms can authenticate images which have undergone contrast, brightness, and affine warping adjustments. We introduce the image authentication system using distributed source coding from transmitter side to receiver side. We formulate image authentication problem as a hypothesis testing problem. The original image projection is quantized and encoded using chaos based coding, a form of distributed source coding. By correctly choosing the size of the chaos based bit stream, it can be decoded using the legitimate image as side information.