Fuzzy-BPN Based Improved Technique for Color Image Watermarking

Digital image watermarking is a technique to hide confidential information into data like image, audio and video which is extracted later for several purposes like ensuring identification and authentication. In the recent years the advancement in the medical science motivated researcher to work in color image watermarking domain. In this paper we proposed watermarking technique for color cover image to hide the patient's/doctor's confidential information. We hide different amount of information into RGB component by considering the contribution of color to the luminance of image to get better imperceptibility. To achieve better Robustness and efficiency we use Fuzzy-BPN architecture to train the system based on Human Visual System (HVS) parameters. In the proposed approach we find improved hiding capacity and robustness and at the same time better imperceptibility. To analyze the proposed watermarking technique Normalized-CrossCorrelation (NC), Structural Similarity Index (SSIM) and Peak-signal-to-noise-ratio (PSNR) are used as the performance parameters.