Low-Light Image Enhancement Network Based on Improved RRD-Net
To address the issue of insufficient exposure in low-light environments,this paper proposes an image enhancement method based on the Retinex model,which decomposes the input image using RRD-Net as priori knowledge.Adaptive equalization is performed on the reflectance image,adjusting the original image to generate a noise image,and combining it with the illumination component to create a grayscale image.Finally,the images are integrated to form a newly exposed-repaired image.This method aims to improve the quality of images under low-light conditions,enhance visibility,highlight object details,and reduce noise,demonstrating significant practical value.The enhanced images are then objectively assessed using three methods:namely Natural Image Quality Evaluation(NIQE),Peak Signal-to-Noise Ratio(PSNR)and Structural Similarity Index(SSIM).The NIQE value is assessed to be 5.63,the SSIM value is 0.78,and the PSNR reaches 16.73.The results indicate that the proposed algorithm significantly outperforms other algorithms in image enhancement.
万方数据
维普
