EDGE oriented image denoising through an adaptive thresholding in the complex wavelet domain

Abstract

Noise reduction is a fundamental process in the enhancement of image quality. In recent years, large bodies of approaches have been developed to minimize the effect of noise in the image based applications. In this study, the authors proposed a novel image denoising framework based on applying adaptive thresholding on complex wavelet transform methods. In the proposed approach, the adaptive thresholding has high capacity to tune its parameters according to the noise type and noise intensity. Further, focusing over the preservation of edges with minimum complexity, this paper proposed a new patch grouping mechanism based on the Gabor wavelet coefficients. Simulation experiments are employed over the image samples to evaluate the performance of proposed mechanism by quantifying the signal strength, structural preservation and edge preservation with respect to the PSNR, SSIM and FOM. In the experiments, the proposed approach had shown an optimal performance in both the edge preservation and quality enhancement with less computational burden.