Lesson: Hyperspectral Images

About the Data:

The data for the ENVI tutorials are found in the class/gly560/class/envidata directory.

About the Lesson:

This lesson will consist of completing two ENVI tutorials, with some minor additions (to be turned in):

1. ENVI Tutorial Number 12: Introduction to Hyperspectral Analysis
   • This tutorial will gives a nice introduction to spectroscopy, particularly with respect to mineral identification. You'll work with "famous" AVIRIS data from Cuprite, Nevada.
     
2. ENVI Tutorial Number 13: Basic Hyperspectral Analysis.
   • This tutorial is gives and introduction to the ENVI tools for handing hyperspectral data, and provides an opportunity to work with library spectra.
3. Create a classification map based upon the results of ENVI tutorial 13.
   • At the end of Tutorial 13, you work with a region of interest file: cup95_av.roi. Use this ROI file, along with a parallelpiped classification tool, to classify the into the class distinguished by the ROI.
   • FIRST read the ENVI user help files on supervised classification and the parellelpiped method.
   • Use parallelpipeds with a 2 standard deviation distribution. You do not need to generate rule images, unless you want to inspect them.
   • Use Overlay ->Classify-> to create a Map Key for the image. Expand the boarder of the image (by about 250 pixels) to allow the map key to be placed next to the image. Add a title, your name, and save the image as a jpg file.
   • Prepare an identically dimensioned image using AVIRIS bands 183, 193, 207 for RGB. Save this image also as a jpeg file.
   • Import both jpg files into an html file, placing one image below the other. Publish the page to your home directory. Browse the file and compare the two images. Does the classification make sense? Are there obvious mistakes in the classification?
   • Print the page, and make sure that the url appears on the page. Note, in pen or pencil, and problems that you identified in the classification. Turn this page in.