Content Based Image Watermarking in the Ridgelet Domain

A novel robust watermark embedding and extracting algorithm in ridgelet domain is proposed. Since the ridgelet transform (RT) can efficiently represent image with linear singularities and has directional sensitivity, the image is first partitioned into small pieces. Firstly these small pieces are classified to different characteristic categories (with weak texture, strong texture) according to the statistics properties of columns coefficients in RT domain which have the directional energetic property. To improve the robustness of processing like noise attack, the middle frequencies of RT subband are selected. And eventually the watermarks are embedded in the most important energetic directions of the pieces with strong texture which are less sensitively to humanpsilas vision. The strength of embedding watermarking is adaptively determined by the local piece, which ensures invisibilities of watermark. Experimental results show that the proposed watermarked scheme is robust to noise, cut and other intensive attacks.