CDRWF: Compressed Domain Based Robust Watermarking Framework for Colored Images

As the volume of digital data increases, there is an increasing need for effective compression methods to address storage demands. Concurrently, the importance of robust image watermarking for authentication and ownership verification cannot be overstated. This work tackles the dual challenge of optimizing image compression for storage conservation and implementing strong image watermarking for copyright protection. The suggested approach integrates the K-means clustering compression algorithm to enhance storage efficiency along with a resilient image watermarking technique based on spatial-domain embedding. We introduce a blind robust watermarking approach that uses zero-frequency coefficient alteration independently in the spatial domain instead of using the discrete cosine transformation (DCT) to verify the ownership of colored images. To enhance the robustness of the system, we have incorporated two watermarks into the cover image. This precaution ensures that even if one watermark undergoes deterioration due to attacks, authentication can still be assured by recovering the other watermark. Compared to frequency-domain approaches, our scheme yields better robustness and reduced computing complexity. The average peak signal-to-noise ratio (PSNR) for the test images using our approach is above 39 dB with a compression ratio equal to 5.9978, removing up to 83% of the redundancy of the host image. After comparing our approach with several state-of-the-art methods, its robustness is exposed by the values of normalized correlation coefficient (NCC) close to one and bit error rate (BER) values close to zero. Besides, the scheme is able to embed a total of 8192 watermark bits in the host image of size 512 × 512 × 3. Experimental results affirm the effectiveness of the proposed methodology, marking it as a valuable contribution to the domains of image processing and information security.