Abstract:

In this paper, a geometrically robust image content authentication algorithm based on the Polar Complex Exponential Transform is presented. First of all, the important visual image contents are enhanced by the response values of SUSAN detector, and the PCET moments of the enhanced image are calculated to get the content authentication features. Then with deterministic adaptive quantization the authentication bits are obtained from the quantified authentication features. Finally, the authentication bits are embedded in the PCETs amplitude of the original image with quantization index modulation method. When verifying the image, the watermark extracted from a test image are compared with the the authentication bits of tested image to determine whether the image has been tampered. Experimental results show that this algorithm has the following advantages: it is robust to geometry distortion of rotation and scale; it can resist the processes such as compression, enhance, filter and noise; and it is sensitivity to small tampering.

Key words: Geometric invariant Image content authentication, Digital watermark, Geometric distortion, Polar Complex Exponential Transform