Medical Image Authentication using Watermarking and Blockchain

Abstract

In recent years the use of images in communication has significantly increased. It is important to secure images from unauthorized access and manipulation. Copyright protection approaches protect data from attacks that result in modification, illegal copying or removal. Our proposed approach uses watermarking and blockchain to confirm authentication. Watermarking is a data-hiding technique whereby data is hidden inside an image to make it difficult to modify. Blockchain at its core, is a distributed-ledger that stores secured and encrypted information while ensuring that it is not changed by someone. We present an imperceptible and secure watermarking approach for digital images using the discrete wavelet transform (DWT). The encrypted watermark is saved in the blockchain. When an image is received and it is to be authenticated, the watermark is retrieved from the blockchain and compared with the watermark that is extracted from the given image. The visual quality of the proposed watermarking approach is tested by comparing the original and watermarked images using peak signal-to-noise ratio (PSNR). The basic challenge that we face is finding suitable locations for inserting the watermark while satisfying the imperceptibility aspects along with the security of the algorithm to reach its highest effectiveness.