File Exchange

image thumbnail

Fast Noise Estimation in Images

version (1.6 KB) by Tolga Birdal
Estimate the standard deviation of the noise in a gray-scale image.


Updated 31 May 2012

View License

Editor's Note: This file was selected as MATLAB Central Pick of the Week

This is an extremely simple m-file which implements the method described in :
J. Immerkær, “Fast Noise Variance Estimation”, Computer Vision and Image Understanding, Vol. 64, No. 2, pp. 300-302, Sep. 1996

The function inputs a grayscale image I and returns Sigma, the noise estimate. Here is a sample use:

I = rgb2gray(imread('sample.jpg'));

The advantage of this method is that it includes a Laplacian operation which is almost insensitive to image structure but only depends on the noise in the image.

Cite As

Tolga Birdal (2021). Fast Noise Estimation in Images (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (9)

Berk Cem Arslan

Namwon Kim

sang min sung

Maaria Rantala

David Niles

Antal Horváth

@Rukundo, following the above-mentioned paper, M=[1 -2 1; -2 4 -2; 1 -2 1] is the difference between two laplacians:
M=2(L_2-L_1), where L_2 is the laplacian on the diagonals L_2=[1 0 1;0 -4 0;1 0 1] (1/2) and L_1 the standard laplacian L_1=[0 1 0;1 -4 1;0 1 0].
The author proposes the difference between the two masks, to make the noise estimater "more insensitive" to the laplacian of the "real, desirable image without noise".


In this source code, is M=[1 -2 1; -2 4 -2; 1 -2 1] correct ? I am asking this because the Laplacian kernel is M=[0 -1 0; -1 4 -1; 0 -1 0]. Please tell me which one is correct.

Sanchari Sengupta

This might sound very lame but can you please tell guide me from point to point as to how to run this file? Its very urgent

Youssef Khmou

MATLAB Release Compatibility
Created with R2009b
Compatible with any release
Platform Compatibility
Windows macOS Linux

Inspired by: Noise Level Estimation from a Single Image

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!