A New Hybridization of Bilateral and Wavelet Filters for Noisy De-Noisy Images

In this work we propose, a hybrid noise reduction algorithm that is a combination of a spatial field binary filter and a hybrid wave field threshold function. These two methods are used to stop Gaussian noise. The hybrid filter is a nonlinear filter that deals with spatial averaging of non-uniform edges. We found it to be an effective technique for image reduction. Determining filter parameters for the mixed filter is important to avoid large differences in results, besides the issue of acceleration velocity. This hybrid model, binary filtering, and Wavelet Thresholding have tried standard images, such as normal eyes, MRI, Roya Face, Ultrasound, X-Ray, and Rawa. Different Gaussian noise was added with different standard deviations σ = 10, 20, 35, 40, and 50. The peak-to-noise ratio (PSNR) signal, MSE, VIF, IQI, and the proposed model MSE between pixels were used as quantitative measures of performance of the relative noise reduction algorithms and then were compared to the models.