Section Article

Enhancing Images Using the GSR Algorithm: A Study on Blur and Noise Reduction
Author(s): Akanksha Goel

Abstract
The increasing availability of digital imaging tools has led to a corresponding rise in the demand for high-quality visual data across fields such as medical imaging satellite analysis surveillance biometric identification and consumer photography. Image degradation through blurring and noise significantly affects the integrity and usefulness of such images leading to inaccuracies in interpretation and analysis. The Group Sparse Representation (GSR) algorithm has emerged as a promising model for enhancing image quality by reducing blur and noise while preserving underlying structural details. This paper investigates the role of GSR in image enhancement highlighting its theoretical foundations working principles and comparative performance against traditional denoising and deblurring algorithms. The study also reviews pre-2018 literature to demonstrate how GSR evolved from classical sparse representation frameworks patch-based processing and self-similarity models. The findings show that GSR efficiently enhances degraded images by utilizing patch grouping sparse coding and adaptive dictionary learning. The paper concludes that the GSR algorithm provides an effective balance between preservation of edges suppression of artifacts and computational efficiency making it relevant for advanced image restoration tasks.