Denoising of Images using Wavelet Transform

Denoising images corrupted by noise is a critical task in image processing, where wavelet shrinkage methods have proven to be highly effective. Traditional approaches, such as the SCAD and Soft thresholding functions, are commonly used for noise suppression. This paper proposes a novel shrinkage function designed to improve the accuracy of image denoising. The proposed function is evaluated under various methods, including the Top method, Universal method, and Translation Invariant method, to handle images contaminated by additive white Gaussian noise. Performance is benchmarked against SCAD, Soft thresholding, and Wiener filter methods using Root Mean Square Error (RMSE) and Peak Signal-to-Noise Ratio (PSNR) metrics. Experimental results demonstrate that the novel shrinkage function consistently achieves superior noise reduction and image quality restoration compared to existing methods, making it a robust solution for denoising applications.