Multimedical Image Denoising Framework by Optimal Shearlet Transform-Based Weighted Wiener Filtering
D. Pavunraj, Shanmugasundaram Singaravelan · Fluctuation and Noise Letters · 2025
Studies indicate that integrating filtering in both the spatial and transform domains yields superior outcomes for image denoising compared to using either method independently. We suggest combining weighted Wiener filtering (WWF) with the optimal shearlet transform (OST) for multimedical images (MMI). The shearlet transform breaks down the noisy image at all scales and in all directions. First, linear black window optimization (LBWO) is used to determine the high and low frequency coefficients of ST. Denoising enhances the image quality for more precise analysis and diagnosis. The technique of removing noise from a picture that has been organically distorted by noise is called image denoising. This research proposes an efficient WWF-based noise reduction technique to improve the image quality of several medical imaging modalities. It is discovered that the suggested technique improves the image and carefully reduces noise. In addition, it produces an output with improved mean square error (MSE) and peak signal-to-noise ratio (PSNR). Ultimately, the denoised image is obtained by using the inverse OST. The trials at the end of the paper demonstrate that the suggested approach may produce better results than others.