DefinePK

Abstract

Wavelet transform has played an important role in Image processing task such as compression and restoration. Unlike most of existing denoising algorithms, using the curvelet makes it needless to hypothesize a statistical model for the noiseless image. This wavelet transform fails to represent the images, which has edges and treated them as smooth functions with discontinuity along curves. The curvelet transforms, where frame elements are indexed by scale, location and orientation parameters. This curvelet transform is designed to represent edges and other singularities along the curves which are more efficient than the traditional wavelet transform. Moreover, the curvelet transform and Gaussian filter are used for an effective image denoising system. This process will be based on the block-based noise estimation technique, in which an input image will be contaminated by the additive white Gaussian noise and filtering process to be performed by an adaptive Gaussian filter and curvelet transform. Coefficients of the Gaussian filter will be selected, as the functions of the standard deviation of the Gaussian noise will be estimated from an input noisy image. Denoising an image is carried out by processing an noisy image through Gaussian filter and using curvelet transform which gives better PSNR. The obtained PSNR values can be compared with that of many wavelet and curvelet in RGB regions. The renowned index Peak Signal to Noise Ratio (PSNR) and Root Mean Square Error (RMSE) demonstrate marked improvement of image denoising over other methods.