Image Denoising Algorithm Using Regional Threshold by Wavelet-Based Contourlet Transform

Abstract

Due to the good characteristics of multi-resolution and multi-direction for wavelet-based contourlet transform, a new threshold method is proposed, which considers the influence of decomposition level, energy distribution of different directional sub-band in the same level and property of regional coefficients in the same directional sub-band. The designed threshold function improves the problems of soft and hard threshold functions, and uses the feature of Gaussian noise. The simulation experiment results indicate that, compare to the related methods, the method proposed in this paper has better de-noising results for different degree and different type noise images.