Medical Image Watermarking based on SVD-DWT technique
Khashayar Yaghmaie, Reza Sabbaghi‐Nadooshan · 2012
This paper presents a new semi-blind digital watermarking technique using demographic text data and iris code for identification as multiple watermarks. The original image is decomposed into 2-level wavelet transform. In the second level of wavelet transform, the two sub-bands with higher energy are selected as a place to embed the watermark bits. These sub-bands are divided into 8*8 pixels blocks and then singular value decomposition (SVD) is computed to each block.The singular values of host image are modified by the singular values of the watermarks. Experiments conducted on four types of medical images prove that these modifications are visually imperceptible while it has a good robustness against some common attacks such as compression, filtering, and noise.