Published Paper Details:

Digital image watermarking refers to embedding a digital data into a cover image to later prove the ownership or authentication of the document. Image watermarking find its applications in a number of areas including copyright protection, authentication, scientific analysis, medical analysis etc. to name a few. There are two important issues that watermarking algorithms need to address. Firstly, watermarking schemes are required to provide trustworthy evidence for protecting rightful ownership; Secondly, good watermarking schemes should satisfy the requirement of robustness and resist distortions due to common image manipulations (such as filtering, compression, etc.). In addition to these two requirements, the watermarking must be embedded in such a way so that the perceptible quality of the image will not go below a certain threshold. In past years, a number of image watermarking algorithms have been proposed, based on LSB, DFT, DCT, DWT etc. In this dissertation, a watermarking algorithm is presented to embed watermark in transform domain using SVD transform. SVD transform is a type of compression in which singular values of a matrix are approximated, thereby implementing a lossy compression at the cost of image quality. In this dissertation, U matrix is used for watermark embedding, by redundantly embedding the watermark bits in the 4X4 non overlapping blocks of the image. Thus, the effect of 4 bits of payload data is scattered to 10 pixels of the cover watermark image thereby giving a smooth embedding with high fidelity