Learning-Based Super-Resolution for Image Upscaling Using Sparse Representation
Kavya T. M, Yogish Naik G. R · 2024
The use of super-resolution techniques is essential for improving the visual appeal of low- resolution pictures. In this work, we offer a unique learning-based method that takes use of sparse representation to achieve better image upscaling. The effectiveness of our suggested strategy is demonstrated by experimental findings on common benchmark datasets. Comparing quantitative evaluations to state-of-the-art techniques, considerable gains are shown in peak signal-to-noise ratio (PSNR) and structural similarity index (SSI). Additionally, visual comparisons show that the up scaled photographs have better features and textures, which makes them more aesthetically pleasing and more similar to high-resolution images taken in the real world.