View License

Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.

» Watch video

Highlights from
Fast normalised cross correlation for gray-level and color images

4.5 | 2 ratings Rate this file 9 Downloads (last 30 days) File Size: 3.7 KB File ID: #25365 Version: 1.0

Fast normalised cross correlation for gray-level and color images


Rami (view profile)


An extention of the normxcorr2 function.

| Watch this File

File Information

A fast vectorized implementation of the ncc that handles color (3 channel) images as well as gray level.
The algorithm for colored images is based on the ncc formulation for vector valued functions.
Note: A bug that I found in the normxcorr2 was also fixed here.

Required Products Image Processing Toolbox
MATLAB release MATLAB 7.4 (R2007a)
Tags for This File   Please login to tag files.
Please login to add a comment or rating.
Comments and Ratings (7)
28 Jun 2014 Balaji

Balaji (view profile)

Hi...when i run my main code
"cc=Fast_NCC_Corr(Image2,Image1); "
in my main file..

i get the error like
"Error using >
Matrix dimensions must agree.

Error in Fast_NCC_Corr (line 87)
i_nonzero = find(cc_d > tol);"

what is this error in line 87.

Please help.. Thanks

Comment only
09 Sep 2013 JAVIER moreno

couldnt have equal size?

Comment only
09 Sep 2013 JAVIER moreno

Hi. In the normxcorr2 documentation C = normxcorr2(template, A) I read the A must be larger than "template" for the normalization to be meaningful. Why this happens. Thanks

Comment only
22 May 2013 Qiao Liyong

13 Dec 2011 Renuka

Renuka (view profile)


For Executing this file it need the Template As i am new to Matlab I dont know how to create a template and exevute this file Please help me...

Comment only
18 Aug 2010 Rob Campbell

Rob Campbell (view profile)

NOT a fast version of the Mathworks function: Doesn't return a matrix, as does Matlab's normxcorr2, but the x and y location of the peak. Poorly documented.

Comment only
05 Feb 2010 Steve

Steve (view profile)

Runs fast on a first test, but how can I get the reliability out from this (height of the peak)?

Contact us