Adaptive Image Super-Resolution Algorithm Based on Fractional Fourier Transform

Abstract

Super-resolution imaging is a critical image processing stage that improves visual image quality. Super-resolution imaging has a wide array of use in different fields, such as medical, satellite, and astronomical images. The correct execution of this stage could increase the accuracy and quality of any available processes in different executive fields. Learning methods, especially deep learning, have become much more popular in recent years for performing the super-resolution imaging process. Methods with this approach have high-quality levels but lack appropriate performance times. This study intends to perform super-resolution imaging using an algorithmic approach based on the particle swarm optimization algorithm and the fractional Fourier transform. The test results on a dataset show the 92.16% accuracy of this proposed method.