A Study on a Combination of Sparse Coding and Dictionary Learning Algorithms for Sparse Super-Resolution

This study proposed a method based on sparse coding for super-resolution, which acquires images from low resolution to high resolution. The method combines two parts: Dictionary learning and sparse coding. This combination typically affects the overall performance. The optimization of the algorithms for each part has been extensively studied, but these combinations were rarely studied. This study analyzes how these two parts of the algorithm interact with each other and finds the optimum combination to improve the accuracy of the dictionary learning algorithm.