Wavelet-based color image watermarking using adaptive entropy casting

An adaptive robust image watermarking technique for color image authentication is proposed. In the proposed approach, the Y channel of a Yuv color host image and a concatenated RGB color-image watermark are decomposed into wavelet coefficients. The contextual entropies of the host wavelet coefficients are computed by a log-sum entropy measurement. Then the watermark wavelet coefficients are embedded into the host wavelet coefficients with larger entropy in the corresponding wavelet subbands. An adaptive casting strategy is utilized to embed the watermark coefficients for fully controlling the imperceptibility of watermarked images and the robustness of watermarks. The experimental results show that the proposed approach provides extra robustness against JPEG-compression and image-processing attacks; moreover, the proposed approach has no need of the original host image to extract watermarks.