A Hybrid Local and Global Enhancement Approach for Low-Light Image Quality Improvement

Abstract

Low-light images are less effective for autonomous driving, medical imaging, and surveillance due to their relatively low contrast, noise, and visibility. In order to enhance low-light photographs, this investigation will implement both local and global enhancement strategies. The primary objective is to enhance the luminosity, contrast, and clarity of the image while retaining its natural characteristics. In order to enhance low-light photographs, the proposed approach applies interference reduction, adaptive histogram equalization, gamma correction, and contrast elongation. PSNR and SSIM experiments are implemented in this assessment of structural protection and enhancement. Conventional methods are outperformed by the proposed method in terms of contrast, visibility, and detail preservation. This research established a critical foundation for future developments that will enhance real-time image processing applications and low-light computer vision systems.