Mixed Noise Removal Algorithm Combining Adaptive Directional Weighted Mean Filter and Improved Adaptive Anisotropic Diffusion Model

Abstract

A mixed noise removal algorithm combining adaptive directional weighted mean filter and improved adaptive anisotropic diffusion model is proposed. Firstly, a noise classification method is introduced to divide all pixels into two types as the pixels corrupted by impulse noise and the pixels corrupted by Gaussian noise. Then an adaptive directional weighted mean filter is developed to remove impulse noise, which can adaptively select the optimal direction template from twelve direction templates and replace the gray level of each impulse noise corrupted pixel by the weighted mean gray level of pixels on the optimal direction template. Finally, an improved adaptive anisotropic diffusion model is developed to remove Gaussian noise in the initial denoised image, which can finely classify image features as smooth regions, edges, corners, and isolated noises by characteristic parameters and variance parameter and conduct adaptive diffusion for different image features by designing reasonable eigenvalues of diffusion tensor. A large number of experimental results show that the proposed algorithm outperforms many existing main mixed noise removal methods in terms of image denoising and detail preservation.

COVID-19 ground-glass opacity segmentation based on fuzzy c-means clustering and improved random walk algorithm

“…In the lung CT image, due to the high gray difference between the chest wall area and both sides, in the process of filtering and denoising the lung CT image using the P-M model, the smooth denoising can be carried out in the lung area, while smoothing can be suppressed where there are edges [21] , [22] . Compared with Gaussian filtering, mean filtering and other methods, on the premise of ensuring the important feature information of the image, it can remove the noise and preserve the image edge more effectively [23] , [24] . The nonlinear anisotropic diffusion equation is: where, represents the image to be processed; represents the divergence; represents the gradient operator; represents the diffusion coefficient, which controls the diffusion rate.…”

CVRRSS‐CHD: Computer vision‐related roadside surveillance system using compound hierarchical‐deep models

“…These visual images are need not used in the parking gap, space available process. Initially, noise in the visual surveillance image is removed with the help of a Gaussian filter [26], which eliminates unwanted information using the Gaussian impulse function response. The Gaussian removes the blur information from the image by reducing the contrast of the image.…”

An improved nonlocal means‐based correction strategy for mixed noise removal

“…However, it is more difficult to remove mixed noise due to the complex distribution of the noise. Researchers have developed many effective algorithms [23–32]. Xiao et al.…”

“…Ma et al. [26] combined an adaptive directional weighted mean filter with an anisotropic diffusion model for mixed noise removal. A smart switching bilateral filter (SBF) was proposed [27] to compensate for the shortcomings of the traditional SBF filter.…”

“…However, it is more difficult to remove mixed noise due to the complex distribution of the noise. Researchers have developed many effective algorithms [23][24][25][26][27][28][29][30][31][32]. Xiao et al [23] applied one kind of filter to identify IN, and a dictionary was learned from the free-noise pixels.…”