An efficient method for image denoising based on a new nonlinear wavelet thresholding function

Image noise is random variation of brightness or color information in images, and is usually an aspect of electronic noise. It can be produced by the image sensor and circuitry of a scanner or digital camera. In fact, there are different kinds of noise functions such as Gaussian noise, salt and pepper noise and speckle noise. Hence, image denoising is the process of removing noise from an image using one of the denoising methods such as spatial or transform techniques. The main aim of an image denoising technique is to achieve both noise reduction and feature preservation. In this context, the wavelet denoising method works in the transform domain, where the noise is uniformly spread throughout coefficients while most of the image information is concentrated in a few large ones.Therefore, the first wavelet-based denoising methods were based on thresholding of detail subband coefficients. However, most of the wavelet thresholding methods suffer from the drawback that the chosen threshold may not match the specific distribution of the image and noise components. Another thing: Studies have proven that the thresholding function has an effective role in obtaining better quality of the denoised image. To address these disadvantages, in this paper, we propose an efficient method for image denoising in wavelets domain. This method is based on a new nonlinear wavelet thresholding function, that is characterized by main mathematical properties and a shape parameter. The theoretical study of this new method proves that we can overcome the drawbacks of classical thresholding methods and by freely adjusting the shape parameter, we achieve a compromise between Hard and Soft thresholding. The experimental results show that our proposed method provides better performance compared to classical thresholding methods in terms of the visual quality of the denoised image.