Improvement of Image Super-resolution Algorithms using Iterative Back Projection
Ercole De Castro, Mariko Nakano-Miyatake, Gabriel Sanchez-Perez, Hector Manuel Perez-Meana · IEEE Latin America Transactions · 2017
This paper presents a scheme for improving the image quality produced by most super-resolution (SR) algorithms. In the proposed scheme the quality of a given image whose size has been increased by a SR system, is improved by using an iterative back projection and sharpening process. The improvement of the image quality obtained by using the proposed method used together with several previously proposed image interpolation algorithms is compared with those obtained by using classic interpolation methods and some other state-of-the-art algorithms. In all cases, the image is evaluated using several Image quality assessment models such as: Visual Information Fidelity (VIF), Structural Similarity Index Measure (SSIM), and Spatial-Spectral Entropy-based Quality index (SSEQ), as well as subjective way. Evaluation results show the desirable features of the proposed scheme.