A NEW IMAGE DENOISING TECHNIQUE USING BOX BILATERAL FILTER AND CURVELET TRANSFORM WITH EDGE PRESERVING STRATEGY

Abstract

Images are often get corrupted during its acquisition and transmission process. Image denoising is a process in which image data is manipulated to produce a visually high-quality image. This paper introduces a new framework for the removal of noise embedded in the image using curvelet transform known as unequally spaced fast Fourier transform (USFFT) with box bilateral filtering. The noisy image is subjected to box bilateral filter and the residual image is further processed by applying curvelet transform. Here, the residual noise in modified curvelet coefficients is further suppressed by Bayesian shrink estimator during curvelet projection. Finally, the recovered edge information is fused with filtered image to form the denoised image. The performance of the proposed method is evaluated based on the parameters peak signal to noise ratio (PSNR) and structural similarity index (SSIM) demonstrate that the proposed method is very efficient in removing noise as compared to some of the popular denoising algorithms.