Content Based Image-in-Image Watermarking Using DCT and SVD

In this paper we consider content based features like texture, luminance, corners and the edges based attacks including JPEG compression, noise, filtering and rotation etc. Experimental results are evaluated with correlation and PSNR. Experimental results show that the proposed watermarking algorithm DCT-SVD-JND is more robust than the combined DCT - SVD information to generate mask JND (just Noticeable Distortion) that helps hidden water mark less perceptible to the human eye. Proposed watermarking algorithm for digital image based on combined DCT and SVD along with JND values. This algorithm is more robust and the watermark can be retrieved with reasonable amount of distortion even in the case of attacks such as frequency method.