A QDCT- and SVD-based color image watermarking scheme using an optimized encrypted binary computer-generated hologram

Based on quaternion discrete cosine transform (QDCT) and singular value decomposition (SVD), a novel color image watermarking scheme using an optimized encrypted binary computer-generated hologram (CGH) as the watermark is presented. A Fibonacci transform-based binary CGH technique that uses particle swarm optimization (PSO) algorithm to improve the reconstructed image quality has been proposed to generate a hologram of a watermark. By use of QDCT, the host color image is treated holistically as a vector field. The component of the host color image after QDCT, which is suitable for embedding watermark, is analyzed. With SVD, the mark CGH is inserted into the host color image by modifying the U matrix. To embed the binary CGH, multiple embedding strength factors are used in this study. To withstand the geometric attacks, based on Zernike moment algorithm and invariant centroid, an improved geometric distortion detection algorithm is proposed. Considering the significant color information, this algorithm is performed on the color invariance model. To obtain better imperceptibility and robustness, PSO is used for finding the proper multiple embedding strength factors. Compared with traditional geometric estimation method, the improved algorithm can estimate the transform parameters of the geometric distorted color image more accurately. Experimental results show that the proposed color image watermarking is not only secure and invisible, but also robust against common image processing operations and geometrical distortions.