Methods for Blind Estimation of the Variance of Mixed Noise and Their Performance Analysis

In image processing and computer vision, the denoising process is an important step before several processing tasks. This paper presents a new adaptive noisereducing anisotropic diffusion (ANRAD) method to improve the image quality, which can be considered as a modified version of a speckle-reducing anisotropic diffusion (SRAD) filter. The SRAD works very well for monochrome images with speckle noise. However, in the case of images corrupted with other types of noise, it cannot provide optimal image quality due to the inaccurate noise model. The ANRAD method introduces an automatic RGB noise model estimator in a partial differential equation system similar to the SRAD diffusion, which estimates at each iteration an upper bound of the real noise level function by fitting a lower envelope to the standard deviations of pre-segment image variances. Compared to the conventional SRAD filter, the proposed filter has the advantage of being adapted to the color noise produced by today's CCD digital camera. The simulation results show that the ANRAD filter can reduce the noise while preserving image edges and fine details very well. Also, it is favorably compared to the fast non-local means filter, showing an improvement in the quality of the restored image. A quantitative comparison measure is given by the parameters like the mean structural similarity index and the peak signal-to-noise ratio.

show abstract

“…8a. The points in the scatter plot have the coordinatesr 2 l for the vertical axis andû l for the horizontal axis [85].…”

Section: Noise Model Estimationmentioning

confidence: 99%

Adaptive noise-reducing anisotropic diffusion filter

Abdallah

Malek

Azar

et al. 2015

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…Then, if statistical characteristics of mixed noise are estimated not accurately, variance stabilization is not perfect and this leads to reduction of filtering efficiency. Note that blind estimation of mixed noise parameters is not able nowadays to provide quite accurate estimation of parameters for all images and all possible sets of mixed noise parameters (Zabrodina et al, 2011). Besides, non-local filtering methods are usually not fast enough since search for similar patches requires intensive computations.…”

Section: Component-wise and Vector Filteringmentioning

confidence: 99%

“…Finally, the third problem deals with accurate estimation of signal-dependent noise statistical characteristics (Zabrodina et al, 2011). Even assuming a proper variance stabilizing transform exists as, e.g., generalized Anscombe transform (Murtag et al, 1995) for mixed Poisson-like and additive noise, parameters of transform are to be adjusted to mixed noise statistics.…”

Section: Component-wise and Vector Filteringmentioning

confidence: 99%

Classification of Pre-Filtered Multichannel Remote Sensing Images

Ponomarenko¹,

Fevralev²,

Vozel³

et al. 2012

Remote Sensing - Advanced Techniques and Platforms

Self Cite

View full text Add to dashboard Cite

“…This value is pre-estimated by one of blind noise parameters estimation methods (e.g. [8]) if not known in advance.…”

Section: Description Of the Proposed Filtermentioning

confidence: 99%

Weighted MSE based spatially adaptive BM3D

Ponomarenko

Pismenskova

Егиазарян

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

Abstract-Weighted MSE (wMSE), recently introduced modification of MSE, is an image quality metric used to estimate visual quality of filtered images. It provides better than MSE correspondence to a human perception in consideration of distortions introduced by image filters. In this paper, wMSE is used both as a criterion to evaluate filtering efficiency of the modification of BM3D filter with spatially varying parameters, as well as to train a specially designed neural network to predict filters' parameters. Extensive analysis on three image datasets demonstrates that the proposed modification of BM3D provides lower values of wMSE than those of BM3D, both effectively suppressing noise in homogeneous regions as well as preserving fine details and texture.

show abstract

Methods for Blind Estimation of the Variance of Mixed Noise and Their Performance Analysis

Abstract: International audienc

Cited by 14 publications

References 34 publications

Adaptive noise-reducing anisotropic diffusion filter

Adaptive noise-reducing anisotropic diffusion filter

Classification of Pre-Filtered Multichannel Remote Sensing Images

Weighted MSE based spatially adaptive BM3D

Contact Info

Product

Resources

About