Abstract
Multiplicative speckle noise removal is a challenging task in image processing. Motivated by the performance of anisotropic diffusion in additive noise removal and the structure of the standard deviation of a compressed speckle noisy image, we address this problem with anisotropic diffusion theories. Firstly, an anisotropic diffusion model based on image statistics, including information on the gradient of the image, gray levels, and noise standard deviation of the image, is proposed. Although the proposed model can effectively remove multiplicative speckle noise, it does not consider the noise at the edge during the denoising process. Hence, we decompose the divergence term in order to make the diffusion at the edge occur along the boundaries rather than perpendicular to the boundaries, and improve the model to meet our requirements. Secondly, the iteration stopping criteria based on kurtosis and correlation in view of the lack of ground truth in real image experiments, is proposed. The optimal values of the parameters in the model are obtained by learning. To improve the denoising effect, post-processing is performed. Finally, the simulation results show that the proposed model can effectively remove the speckle noise and retain minute details of the images for the real ultrasound and RGB color images.
Highlights
Image denoising is a very important problem in image processing
It can be revealed that our method provides the optimal peak signal-to-noise ratio (PSNR), mean absolute deviation error (MAE) and structural similarity (SSIM) values in all examples, which values in allinexamples, which are shown in bold face infurther
Resultsobtained obtained with different denoising methods applied to the Shepp-Logan head image corrupted with noise (σ phantom image corrupted with noise (σ=3): (a) noisy; (b) ours; (c) speckle reducing anisotropic diffusion (SRAD); (d) OBNLM; (e) ADMSS; (f)
Summary
A variety of image denoising methods have been developed to deal with additive noise in recent decades, including variational-based methods [1,2], partial differential equation (PDE)-based methods [3,4,5,6], filter-based methods [7,8], sparse representation (SR)-based methods [9,10,11], wavelet-based methods [12,13], deep neural network (DNN)-based methods [14,15,16], etc. Multiplicative noise is still a difficult problem to deal with it by various methods. Motivated by Zhou’s work, we focus on anisotropic diffusion-based multiplicative speckle noise removal methods in this work
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have