Single image super-resolution algorithm using PSNR in the wavelet domain
Source of Publication
Journal of Advanced Research in Dynamical and Control Systems
© 2020, Institute of Advanced Scientific Research, Inc.. All rights reserved. This article introduces a new algorithm for enhancing the resolution of interpolated images. The aim of this proposed algorithm is to obtain a high resolution image from a single low resolution image. The new algorithm uses the interpolation technique with a two dimensional filter that is specifically designed for the image by maximizing the peak signal to noise ratio in the wavelet domain. In addition to the 2D filter, a prediction method to estimate the difference between interpolated images will be used. The processed images are assessed by comparing them with ground truth images which are used for assessment purpose only. The proposed algorithm performance is evaluated using the peak signal to noise ratio in the spatial domain (PSNR), the peak signal to noise ratio in the wavelet domain (WPSNR) and the structural similarity index measurement (SSIM). The proposed new algorithm gives a better result than the previous algorithms.
Mahmou, Abdulqader; Taher, Fatma; Kunhu, Alavi; and Al-Ahmad, Hussain, "Single image super-resolution algorithm using PSNR in the wavelet domain" (2020). Scopus Indexed Articles. 406.