In this exercise you will learn advanced classification using the image processing capabilities of Idrisi Kilimanjaro. Start Idrisi and click Help and then “IDRISI Tutorial”. Scroll down to Tutorial Part 5. The data for this lab can be found in I:\Students\Instructors\Geoffrey_Duh\IDRISI\Advanced IP. Please copy the files to the “c:\temp” folder on your local computer. Complete exercises 5-1 through 5-6, which include the following topics:
1) Bayes' Theorem and Maximum Likelihood Classification
2) Fuzzy Signature Development
3) Bayclass Soft Classifier (evaluation of probability that a pixel is a member of a class)
4) Hardeners (forcing decision of class membership)
5) Dempster-Shafer Soft Classifier and BELCLASS
6) Dempster-Shafer and Classification Uncertainty
You may need to read/skim the help section called Classification of Remotely Sensed Imagery (pages 201-227) in “IDRISI Manual” before you begin. To quickly access lab data, you can click the data paths in the “File” menu and set the main working folder to where the data are. Getting familiar with IDRISI’s GUI probably will be the most time-consuming part of this lab. Be patient!
Write a report that describes the different results of the different classification techniques. Be sure to include a map of each resulting classification as well as any tabular results that will illustrate their differences.