AN EFFICIENT MULTIRESOLUTION IMAGE DENOISING USING SMOOTHING SPLINE ESTIMATION
G. M. Rajathi, Raghuraman Rangarajan, Sri Ramakrishna · 2014
The image data is normally corrupted by additive noise during acquisition. This reduces the accuracy and reliability of any automatic analysis. For this reason, denoising methods are often applied to restore the original image. The effectiveness of image denoising depends on the estimation of noise variance of noisy image. In this paper, the Discrete Wavelet Transform(DWT) is used, which offers simultaneous localization in time and frequency domain and the noise variance is estimated using Smoothing Spline Estimation method. Multiresolution bilateral filter is the extension of bilateral filter. Bilateral filter is a nonlinear filter that does spatial averaging without smoothing out the edges. Multiresolution bilateral filtering is applied to the approximation sub bands of the transformed noisy image and thresholding is applied to the detailed sub bands which uses the estimated noise variance. With this process, a better PSNR is achieved.