REVERSIBLE WATERMARKING BASED SECURITY

Abstract

Reversible watermarking technique is used to transfer some small information through the medium of images, and both the image and the data could be recovered at the receiving end. It’s widely used in medical imaging to transfer the patient information corresponding to his/her medical image for the purpose of diagnostic. Here we propose a technique based on difference expansion, in which the transferred image consist of the original image embedded with the payload consisting of patient info, hash of original image and original bits required for the recovery. In order to provide confidentiality, the payload is encrypted using symmetric key based encryption algorithm, this provides that the patient information is confidential and the image too could not be recovered for diagnoses under attack. Hash of the recovered image is calculated and compared with the recovered embedded hash to make sure that the image has not been tampered. This provides for the confidentiality, authentication and the integrity of the transferred image. We have also extended the concept of reversible watermarking earlier confined only to images and audio data, to other forms of data formats as well. It could be widely used in copyright protection and could be effectively used to distribute .docx, .xls, .ppt, .pdf, .exe and many other file formats securely over the network. The approach is based on sliding window approach in which the embedding space is not the whole available space of the document or the file but a small window is selected out of the available space.