A universal noise removal algorithm with an impulse detector

Abstract

Abstract-We introduce a local image statistic for identifying noise pixels in images corrupted with impulse noise of random values. The statistical values quantify how different in intensity the particular pixels are from their most similar neighbors. We continue to demonstrate how this statistic may be incorporated into a filter designed to remove additive Gaussian noise. The result is a new filter capable of reducing both Gaussian and impulse noises from noisy images effectively, which performs remarkably well, both in terms of quantitative measures of signal restoration and qualitative judgements of image quality. Our approach is extended to automatically remove any mix of Gaussian and impulse noise.

Robust local similarity filter for the reduction of mixed Gaussian and impulsive noise in color digital images

“…To alleviate this problem, various designs introduce additional weighting factors which diminish the influence of outliers on the final filtering outcome. In [14], an extension of the bilateral filter based on the ROAD measure was proposed. The so-called trilateral filter assigns to each pixel its measure of impulsiveness, which is considered when building the average over the pixels belonging to the local processing region.…”

“…The authors of [14] described the Rank of Ordered Absolute Differences (ROAD) statistic for impulse detection and combined it with the bilateral denoising scheme [52], designing a trilateral filter able to efficiently reduce the mixed noise. This filter, although intended for gray-scale images, can be also applied to color images, replacing the difference in intensities by the distance in a given color space.…”

Single-Patch Low-Rank Prior for Non-pointwise Impulse Noise Removal

“…The median filter and its extensions are often effective to remove salt-and-pepper noise but can corrupt some textures. To reduce undesired corruption, various two-stage methods have been developed, first detecting the locations of noisy pixels, then recovering intensities only at noisy locations using certain filtering or variational methods, e.g., adaptive center weighted median filter (ACWMF) [4], rankordered absolute difference (ROAD) noise detector followed by a trilateral filtering [9], and a logarithmic version of the ROAD followed by edge-preserving regularization (EPR) for pixel restoration (ROLD-EPR) [7]. Both ROADtrilateral and ROLD-EPR methods can preserve edges better than median-type methods like ACWMF as they consider local structures during pixel restoration.…”

“…In practice, the binary matrix W 0 can be generated by any existing impulse noise detectors (e.g., ROAD [9] or ROLD [7]), or even by our low-rank matrix recovery framework by setting W = 1. This is because our framework (when W = 1) also recovers the sparse matrix S representing the initial spatial distribution of impulse noise.…”

A multiresolution framework for local similarity based image denoising

“…In order to overcome the limitations of bilateral filtering, Garnett et al [31] proposed a trilateral filter employing a local image statistic for identifying the noisy pixels. The trilateral filter proposed in [31] was mainly aimed at denoising images corrupted with impulse noise, although it was shown to be effective for removing Gaussian and mixed noise too.…”