GAUSSIAN MINIMUM ERROR CLASSIFIER EXAMPLE

OVERVIEW

This example assumes the training and target data sets (images) has pixel data that follow
the Gaussian (normal) distribution. The traing data set is composed of two images. One
is for pixel data. The other is for class label. The class label data should be 
greater than or eaqual to 0 and there should be no missing value between 0 and the largest
value. For example, there are four classes. the class lables should 0, 1, 2, and 3.

The target data set (image) is an image that the classifier will assign class labels 
for each pixel in the image.

For more information, please refer to the Statistical Pattern Classification chapter of the
Users Guide. You can get the document by checking out InsightDocumentation repository from
itk.org using cvs. You can find download instruction for the InsightDocumentation at
http://www.itk.org/HTML/Download.htm.

NOTE: The example supports only metaimage file format (.mha or .mhd).

HOW TO COMPILE

Before compile the utilities in this example, you have to turn on the
BUILD_AUXILIARY option and BUILD_EXAMPLE option which you can find
in the CMakeList.txt file in the root of the Insight source
directory. After all necessary libraries are compiled, cd to this directory
and run make or open the dsp file for Visual C++.


HOW TO USE THE APPLICATION

Just type the name of the application you want to use in command line
without any option. The application will show its usage.

