Enhanced Image Denoising Using Color Wiener Filtering with Optimized Low-Rank Approximation

Abstract

Sparse representation and low-rank approximation are popular for image denoising, but struggle with complex structures in heavily degraded images due to inadequate local descriptors and coefficient shrinkage rules. Hence, this research introduces a novel approach for image denoising using Color Wiener Filtering with Optimized Low-Rank Approximation. The proposed model integrates gamma correction for contrast enhancement with Color Wiener Filtering and Optimized Low-Rank Approximation to effectively remove noise from various images. Additionally, L0 smoothing is employed to address blurring effects while preserving image edges. The study evaluates the proposed model's performance using key parameters, including Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM). Results indicate that the proposed model achieves an average MSE of 0.075, PSNR of 44.47 decibels (dB) and SSIM of 0.961. The suggested model is compared to other traditional state-of-the-art approaches to validate its performance, and it is found that in terms of PSNR and SSIM, the proposed model surpassed all other state-of-the-art methods. Overall, this research highlights the challenges addressed by the proposed model, its attained results, and its superiority compared to existing approaches, positioning it as a promising solution for image-denoising applications.