Adaptive fractional masks and super resolution based approach for image enhancement

Abstract

This paper presents an image enhancement technique based on super-resolution approach. The method uses fractional filters and reconstructs the output image by projection on convex sets (POCS) method. First, we generate a reference frame by using low-resolution frames and enhanced it by an adaptive fractional mask. Then the speed-up robust feature (SURF) is used to find the matching between low-resolution frames and the reference frame. Finally, the residuals between matching are reduced by the POCS reconstruction approach. To recover the high-frequency components, we have used a fractional integral mask in the POCS reconstruction process. We have compared the experimental results with some other existing methods from literature. Simulation results show that the proposed approach is efficient and returns a good quality image. © 2020, Springer Science+Business Media, LLC, part of Springer Nature.