What's new in Computer Vision System Toolbox in R2011a

Matlab have updated is Computer Vision System Toolbox.
Here's the documentation.

Here’s a list of what’s new:

• extractFeatures function for creating an array of feature vectors (descriptors) based on interest points within an image

• matchFeatures function for finding the best matches between two arrays of feature vectors (descriptors)

• Visualization of epipolar geometry for stereo images using epipolarLine, isEpipoleInImage, and lineToBorderPoints functions

• estimateUncalibratedRectification function for calculating projective transformations to rectify stereo images

• Video segmentation based on Gaussian Mixture Models using ForegroundDetector System object YCbCr video format support for ToVideoDisplay block and DeployableVideoPlayer System object

Calling this product a toolbox also allows us to clarify and highlight the MATLAB capabilities in the product that we’ve had since R2010a. Some of these algorithms overlap with Image Processing Toolbox, but provide support for C code generation and fixed-point modeling. Others are unique to Computer Vision System Toolbox, including these:

• 
  
• vision.BlockMatcher
• vision.Deinterlacer
• vision.GeometricTransformEstimator
• vision.OpticalFlow
• vision.TemplateMatcher

Source : Steve Blog

Matlab Visual Sudoku Solver

After the Google Goggles Sudoku Demo Matlab have released near the same demo.
Not purely real time. But a working one with explanation. (Basis bricks are a sudoku solver, some image processing (morphological operations) and number recognition).

http://www.mathworks.com/videos/matlab/sudoku.html

The source code is accessible on the same webpage.

Configure Matlab to use Visual 2010 C/C++ compiler

By default Matlab 2009 have only a C compiler (mex file compiler).
It could detects the Visual compiler for the 2005/2008 version but not for the 2010 version.

Hopefully MathWorks have released a patch ;)

Link : VS2010MEXSupport.zip

Subject:

How can I use Microsoft Visual C++ 2010 to create MEX files with MATLAB 7.10 (R2010a)?

Problem Description:

I would like to use Visual C++ 2010 to create MEX files for use in MATLAB.

Solution:

The Microsoft Visual C++ 2010 compiler is not supported on the released version of MATLAB 7.10 (R2010a), but support can be added by installing a patch. To install the patch:

1. Download the patch attached to this Solution. The patch includes files needed to support these combinations of MATLAB and Visual C++ 2010:
• Visual C++ 2010 Professional and 64-bit MATLAB 7.10 (R2010a)
• Visual C++ 2010 Professional and 32-bit MATLAB 7.10 (R2010a)
• Visual C++ 2010 Express (Windows SDK 7.1 also required) and 64-bit MATLAB 7.10 (R2010a)
• Visual C++ 2010 Express and 32-bit MATLAB 7.10 (R2010a)

2. Confirm that you are running MATLAB 7.10 (R2010a). You can check this using the About box available at Help->About MATLAB.

3. Unzip the contents of the patch into your MATLAB installation. This can be done from within MATLAB itself with the command:
unzip( path_to_zip_file, matlabroot);

4. Set up MEX to use Visual C++ 2010 with the command:

mex –setup Visual C++ 2010 should now be among the listed options.

Note that this patch will only work with MATLAB 7.10 (R2010a). Using this patch with any version of MATLAB is not supported.

Other MathWorks products such as Real-Time Workshop do not support Visual C++ 2010.

Source : MathWorks

Hany Farid : Fundamentals of Image Processing

This guide from Hany Farid gives you a mathematical viewpoint of common point use in Image Processing.
It's a good introduction.

Learn about the fundamentals of signal and image processing built upon a unifying linear algebraic framework.
http://www.cs.dartmouth.edu/farid/tutorials/fip.pdf

Other short guide are accessible here : http://www.cs.dartmouth.edu/farid/tutorials/

Fast euclidean distance in matlab/octave

On this topic we could see an interesting discussion about how make the fastest euclidean distance.

But I have see that on Octave, it is not true at all...

 a= rand(1000, 800);
b= rand(1000, 800);

%method A:
tic; d = sqrt( sum((a-b) .* (a-b),2)); toc;

%method B
tic; d = sqrt( sum(a.*a + b.*b - 2 *( a .*b),2)); toc;

%method C
tic;
aa=sum(a.*a,1); bb=sum(b.*b,1);
 d = abs(aa( ones(size(bb,2),1), :)' + bb( ones(size(aa,2),1), :) - 2*a'*b); toc

A => Elapsed time is 0.02084 seconds.
B => Elapsed time is 0.0335 seconds.
C => Elapsed time is 1.49 seconds.

So on my machine the naive method is the fastest, certainly due the fact that octave do not use SSE library for my part.

Octave/Matlab graphics Master

You will find here a powerful guide to make graphics/plot in octave matlab.

http://iste.epfl.ch/cours_matlab/graphiques.html

I know it's written in french, but a picture with universal source code is better than thousand words !

All the following list is explained :

Pie/Bar/Histogram/Stem//Plot/Polar/Meshgrid/Quiver (gradient field) / Slice (3d data)....