Combined blur and RST invariant digital image watermarking using complex moment invariants

In this paper, we present a watermarking system including hiding, detecting and extracting a confidential two-dimension binary watermark, such as a school logo, into a still image, which is robust to geometric distortions and blurring. Watermark embedding is performed in the DCT domain of the original image. The watermark detection is as simple as computing a set of complex moment invariants to illuminate the robustness problem under general geometric image manipulations. In the extraction scheme, we first estimate the geometric distortion factors of the geometrically attacked watermarked image. With these estimated factors, the geometrically attacked image can be transformed back to its original position, size and shape to recover the lost synchronization for watermark extraction. The watermark is then extracted from the corrected image.