Applications Of Compressed Sensing In Super-Resolution Problem

This article discusses the application of compressed sensing in thesuper-resolution problem. According to the number of low-resolution images,we can divide the super-resolution problem into two kinds of problem: singleimage super-resolution problem and multiple images super-resolution problem.In this paper, we succeed in figuring out the single image super-resolutionalgorithm, and then we do a lot of research about the dictionary in the singleimage super-resolution algorithm. We have made three different kinds ofdictionaries, and analyzed and compared the experimental results of a lot ofexperiments using the dictionaries in the algorithm.After figuring out the single image super-resolution algorithm, weindependently design a relative norm which evaluates the effect the singleimage super-resolution algorithm and a norm which evaluates the effect ofdictionaries in the single image super-resolution algorithm. Then we use thesenorms to compare and analyze our three different dictionaries.In the last part, we improved the single image super-resolution algorithm,and succeed in figuring out the multiple images super-resolution algorithm.Then we provide an example to show our algorithm’s effect.