Region of Interest based robust watermarking scheme exploiting the homogeneity analysis

Abstract

Region of Interest (ROI) based watermarking is employed where some regions of the image content are more important as compared to the other portions. Hence, the aim is to protect these ROI regions to retain the commercial value of the image. Towards this end, a robust watermarking scheme has been proposed in this paper that extracts feature from the image itself and embeds it imperceptibly into the region of interest portions of the image. To enhance the security of the scheme, the feature is encoded using the convolutional based encoding prior to embedding into the appropriate sites of the ROI. The homogeneity analysis of the ROI is carried out using the quadtree based image segmentation to select homogeneous blocks of a fixed size depending upon the requirement of the application. The robustness efficacy of the scheme has been tested against various attacks and evaluated using peak signal to noise ratio (PSNR) and normalized cross correlation (NCC) metric values. © 2016 IEEE.