Illumination Map-Based Enhancement for lowlight Images

Low-light images often suffer from poor visibility, which impacts their visual quality and hinders various computer vision applications.This paper presents an improved low-light image enhancement model aimed at increasing the Peak Signal-to-Noise Ratio (PSNR) by integrating advanced image enhancement techniques, specifically Image Adjustment and Histogram Equalization.The proposed approach first utilizes the LIME (Low-light Image Enhancement) method, which estimates illumination by extracting maximum intensity from RGB channels and refines the illumination map using a structureprior-based regularization technique.Guided filtering is applied to achieve smooth illumination adjustments, followed by gamma correction and local adaptation using kernel filtering.Additionally, BM3D denoising is employed to suppress noise, enhancing image clarity.To further improve the enhancement quality, Image Adjustment and Histogram Equalization are introduced, effectively boosting contrast and brightness.Experimental evaluation using PSNR, Mean Squared Error (MSE), and Correlation Coefficient demonstrates that the enhanced model achieves superior visual quality and higher PSNR compared to the base model.This integrated approach proves effective in real-world low-light image enhancement applications.