Abstract
Digital image watermarking is considered as a patent method to enhance medical data security, confidentiality and integrity. Medical image watermarking requires extreme care when embedding additional data, given their importance to clinical diagnosis, treatment, and research. In this paper, a novel image watermarking for medical images is presented which is robust to geometric attacks like rotation, scaling, translation or combination of these. The watermark is embedded in the feature space of the image, as embedding the watermark in feature space makes it invariant against geometric attacks (translation, scaling, rotation and combination of these) and then Arnold transform which is considered as a chaotic scrambling is applied on the watermarked image to make the watermark more secure. Moreover, the proposed watermarking scheme is blind as the watermark is extracted without knowing the original image at receiver end. Experiment results demonstrate that watermark is robust against geometric attacks.