Tamper detection of medical images using statistical moments against various attacks

Abstract

Medical Imaging has evolved to digital means to make it suitable for transmission and storage purposes. This aids in transfer of medical images for medical transcriptions, tele-medicine and Content Based Image Retrieval (CBIR) applications. Medical images need to be transported with enough security to preserve the contents from intruders, to verify the contents of Hospital Information system for legal purposes and to protect the images from disclosure to unauthorized persons. Data authentication using watermarking preserves the content against modification and also can be used for ready reference anytime. In this work, zernike and Hu moments were used for generation of Hash for the original and distorted images. Hash generated is compared with the Hausdorff distance. It was found that the variance of Hu moments is higher than that of zernike moments and hence the use of Hu moments enhances the security of medical images used for transmission and storage purposes.