File Exchange

image thumbnail

Noise estimators / estimations for various noises

version 1.0.3 (165 KB) by Olivier LALIGANT
Code of a new noise estimator and code of some other noise estimators, comparisons.

13 Downloads

Updated 27 Apr 2020

View License

NOISE ESTIMATION distribution

This a companion distribution of the Ref. paper:
O. Laligant, F. Truchetet, E. Fauvet, 'Noise estimation from digital step-model signal', IEEE Trans. Image Processing, 2013 Dec., 22(12):5158:67

Contact : olivier.laligant@u-bourgogne.fr

This distribution permits to:
- introduce a new noise estimator (NOLSE) with interesting performances on various types of noise
- test various noise estimators on real images corrupted by various synthetic noises
- estimate noise level in image with various noise estimators
The results can be used for various applications. The title image shows an example of image restoration where the parameter of the restoration method is obtained through the noise estimators.

Estimators:
- nolse.m, fnolse.m like-script and function versions of the new estimator NOLSE
- averageN.m noise estimation by S.I. Olsen (see help of averageN)
- FNVE.m noise estimation by J. Immerkær (see help of FNVE)
- mad.m noise estimation by D. L. Donoho (see help of mad)
- TaiYang.m noise estimation by S.C. Tai and S.M. Yang (see help of TaiYang) ;
Rk : please see the above Ref. for the correct equation of the Tai-Yang estimator.

Main:
- testAllGWN.m : test of the estimators on an image corrupted by synthetic GWN
- testAllSpeckle.m : test of the estimators on an image corrupted by speckle noise
- testAllPoisson.m : test of the estimators on an image corrupted by Poisson noise
- testAllImpulse.m : test of the estimators on an image corrupted by impulse noise
- estimNoise.m : estimation of the noise level in an image with various estimators

Tools:
- binarise.m provides a binary image
- fit1p2d.m polynomial fitting
- histo.m histogram (variant)
- jordanOL.m jordan resolution (variant)
- mse.m mean square deviation calculus
- thresh.m low thresholding

Test images are included in the distribution

Cite As

Olivier LALIGANT (2020). Noise estimators / estimations for various noises (https://www.mathworks.com/matlabcentral/fileexchange/63172-noise-estimators-estimations-for-various-noises), MATLAB Central File Exchange. Retrieved .

O. Laligant, F. Truchetet, E. Fauvet, 'Noise estimation from digital step-model signal', IEEE Trans. Image Processing, 2013 Dec., 22(12):5158:67 Olivier LALIGANT (2020). Noise estimators / estimations for various noises - MATLAB Central File Exchange. Retrieved April 27, 2020.

Comments and Ratings (0)

Updates

1.0.3

Reference correction

1.0.2

Some image displays and comments added.

1.0.0.0

Files updated

1.0.0.0

Some comments updated

1.0.0.0

tags updated

1.0.0.0

tags added

1.0.0.0

minor corrections of typo error

MATLAB Release Compatibility
Created with R2012a
Compatible with any release
Platform Compatibility
Windows macOS Linux