A novel robust blind watermarking algorithm based on wavelet and Fractional Fourier Transform

To enhance the robustness of the watermark, this paper proposed an algorithm based on Fractional Fourier Transform. First we analyze the reasonable order of fractional operator for Direct Discrete Fourier Transform method, and then we abstract the feature-vector of the host image, and register to Copyright Agencies to get the watermark Key. In the watermark detection, the Key and the Feature-vector abstracted from attacked watermarked image were used to detect the watermark. Experiments show that method based on FRFT is robust to local distortion, and the local rotation distortion angle is greater than 120⁰.