Fusion Based Watermarking Scheme with Multi Level DWT & SVD

Abstract

This paper proposes "Fusion based watermarking with multi level DWT & singular value decomposition" has been implemented. In watermarking scheme, maintaining security and robustness is major hurdle. To address this issue we proposed a novel non blind embedding scheme with Discrete Wavelet transform (DWT) and Singular Value Decomposition (SVD) techniques. This paper details the design of the proposed watermarking scheme and analyses its robustness in the presence of various possible security attacks that involves in degrading the quality of watermark. In the beginning, the cover image color component (especially Red Component) is decomposed into LL, LH, HL, HH with 2 level DWT: these LH & HL coefficients are further divided into 8*8 blocks and each block is compared to build a fused coefficient. Apply SVD on fused coefficient of cover image and watermark image to embed the singular values with sigmod scaling factor. Finally the watermarked image is generated, after applying inverse SVD & 2 level DWT. At receiver side the inverse process has been implemented to extract watermark image. The efficiency and performance of the proposed method is verified with Peak Signal to Noise Ratio (PSNR), Root mean square error (RMSE) and Mean Square error (MSE) and compared with recent works of santhi [12] and harsha [13].