Digital watermarking robust to geometric distortions in biomedical images

We propose a new normalization procedure, which is invariant to affine transform attacks .The resulting watermarking scheme is suitable for public watermarking applications, where the original image is not available for watermark extraction. , we employ watermarking scheme as a direct-sequence code division multiple access approach to embed multibit text information in DCT and DWT transform domain. In Biomedical images the watermark is embedded in the regions other than the area of interest so that the diagnosis is not affected. The area of interest of the biomedical image is found out using the K-means segmentation method. Performance of proposed watermarking scheme is robust against various types of attacks such as Affine transform, Gaussian noise, shearing, rotation, median filtering and JPEG compression. This proposed algorithm can be used in Hospital environment.