Artificial Immune System Optimization Technique for Robust and Secure Image Watermarking

For a secure data image transmission purpose, digital watermarking techniques are used by embedding and extracting the same watermark into digital content. One of the widely used transform domains for watermarking of most digital images is the Discrete Wavelet Transform domain (DWT)and the hybridization in the digital domain of the image watermarking scheme using (DWT) and Singular Value Decomposition (SVD),which has shown its efficiency in many watermarking schemes. A scaling factor is used to modify the coefficients before inserting them into singular values of the suitable sub-band for host image. In this paper, we propose an optimal watermarking scheme based on DWT and SVD techniques using an Artificial Immune System (AIS) algorithm. This is done with the aim of optimizing imperceptibility and robustness by finding the best scaling factor and searching for the appropriate location for the watermark insertion. Besides, Arnold Transform encryption is used for adding security to the watermark image embedded in the host image. After embedding the watermark, the watermarked image is submitted to several attacks like median filtering, JPEG compression, cropping, and rotation with different distortion strengths. The imperceptibility and robustness of the watermarked image are checked out by measuring the Peak Signal to Noise Ratio (PSNR) and Normalized Correlation Coefficient (NCC). The proposed scheme provides some competitive simulations results proving the efficiency of our proposed method. Our arguments are motivated by the attractive obtained values of PSNR and NCC parameters, compared to those discussed in the recent literature.