Title: Research on lossless restoration method of digital media image based on regularisation method
Authors: Mi Tian
Addresses: School of Design and Art, Changsha University of Science and Technology, Changsha, 410114, China
Abstract: In order to overcome the problems of high mean square error (MSE) and low peak signal-to-noise ratio (PSNR) in existing image lossless restoration methods, a new digital media image lossless restoration method based on regularisation method is proposed. The image is segmented by two-dimensional histogram and distance measure. According to the above processing results, a degradation model of digital media image is constructed. On the basis of the degradation model, a joint variation regularised image restoration model is constructed by regularisation method, and the pre-processed digital media image is restored losslessly. The experimental results show that compared with the traditional methods, the MSE of the proposed method is significantly reduced, the minimum MSE is only 0.17, the PSNR is high and the improved signal-to-noise ratio is significantly improved.
Keywords: regularisation method; digital media image; image degradation model; image lossless restoration; mean square error; MSE; peak signal-to-noise ratio; PSNR.
DOI: 10.1504/IJICT.2022.124809
International Journal of Information and Communication Technology, 2022 Vol.21 No.2, pp.93 - 110
Received: 11 Sep 2020
Accepted: 26 Oct 2020
Published online: 09 Aug 2022 *