GISC 7365: Remote Sensing Digital Image Processing
Instructor: Dr. Fang Qiu
Lab 10: Unsupervised Image Classification
Objective:

To generate clusters using an unsupervised classification approach (ISODATA)
and evaluate the clusters in feature image space.
Image:
Quickview
File - charleston_11-9-82tm.img
Landsat TM Data
Band 1 = Blue (.45-.52)
Band 2 = Green (.52-.60)
Band 3 = Red (.63-.69)
Band 4 = NIR (.76-.90)
Band 5 = MIR (1.55-1.75)
Band 6 = MIR (2.08-2.35)
Band 7 = TIR (10.4-12.6)
In unsupervised classification the computer develops the signatures that results in a
number of spectral classes, which the analyst must then assign (a posteriori) to
information classes of interest. This requires the knowledge of the terrain present in the
scene as well as its spectral characteristics. The Iterative Self-Organizing Data Analysis
Technique (ISODATA) is a widely used clustering algorithm and is different from the
formerly used chain method because it makes a large number of passes through the
remote sensing dataset, not just two passes. It uses the minimum spectral distance
formula to form clusters. It begins with either arbitrary cluster means or means of an
existing signature set, and each time the clustering repeats, the means of these clusters are
shifted. The new cluster means are used for the next iteration.
The ISODATA utility repeats the clustering of the image until either a maximum number
of iterations have been performed, or a maximum percentage of unchanged pixels have
been reached between two iterations. Performing an unsupervised classification is
simpler than a supervised classification, because the signatures are automatically
generated by the ISODATA algorithm. However, as stated before, the analyst must have
ground truth information and knowledge of the terrain, or ancillary high resolution data if
this approach is to be successful.







Open a color infra-red composite of charleston_11-9-82tm.img in a viewer (RGB=
bands 4, 3, 2) and fit to frame.
To begin the unsupervised classification, click on the Classifier icon and then select
Unsupervised Classification. You will notice that the Unsupervised Classification
dialog box states that it is an ISODATA unsupervised classification (Title bar).
Fill in the input file as Charleston_11-9-82tm.img and output information in the
Unsupervised Classification dialog box.
Give both the Output Cluster Layer and Output Signature Set a similar name.
Make sure that under Clustering Options, the Initialize from Statistics box is on and
set Number of Classes to 15.
Under Processing Options, set Maximum Iterations to 20 and leave the Convergence
Threshold set to 0.950. Maximum Iterations is the number of times that the
ISODATA utility will recluster the data. It prevents the utility from running too long,
or from getting stuck in a cycle without reaching the convergence threshold. The
convergence threshold is the maximum percentage of pixels whose cluster
assignments can go unchanged between iterations. This prevents the ISODATA
utility from running indefinitely.
Leave everything else in its default state. When you have entered the entire relevant
information click OK to begin the process.
Cluster Identification
To aid in evaluation we will need to view the results of the clustering so that we may see
how the clusters are arranged in feature space and thereby make informed decisions about
the nature of the cluster. The first step that will allow us to do so is the creation of feature
space images.







Go to the Classification menu and click the Feature Space Image button. A dialog box
will appear saying Create Feature Space Images at the top.
Select the original image (charleston_11-9-82tm.img) as the Input Raster Layer and
make sure the Output Root Name is (charleston_11-9-82tm) and the directory path is
correct.
The number of combinations of two bands out of 7 TM bands is 21. By default,
Imagine will create 21 feature space images! Under feature space layer, select the
layers which are the band combinations of 1 and 3, 2 and 4, 3 and 4, 4 and 5. Four
feature space images will be created.
Leave the rest of the selections at their default settings and click OK.
When the processing is complete open a new viewer and view the output images (i.e.
charleston_11-82tm_1_3.fsp.img, charleston_11-82tm_2_4.fsp.img, charleston_1182tm_ 3_4.fsp.img, charleston_11-82tm_4_5.fsp.img). Examine those four feature
space images.
Open the Signature Editor (under the Classification menu) with the *.sig file you
created in the unsupervised classification.
Select all the clusters (they should all be highlighted in yellow).




In the Signature Editor main menu select Feature and then in that pull-down menu
select Objects.
This will display a Signature Objects dialog box that allows you to tell Imagine which
viewer you want to receive the signature editor information about the clusters. In this
case we want the viewer in which you have displayed your chosen feature space
image. Select that viewer # in the Signature Objects space provided that represents
this viewer.
Select Plot Ellipses, Plot Means and Label (or you can try the others if you like).
Leave everything else in its default state and click OK. Only selected clusters in the
Signature Editor window will be drawn.
More than likely your ellipses and means are multi-tonal in nature. If you would like
them all to be white, red, green, etc..., select all the classes in the Signature Editor
dialog box using the mouse and change the color to the one you desire.
To analyze the content of the clusters, you should use a combination of techniques.







You will more than likely have to zoom in to get a better look at some of the clusters
given the close proximity of clusters to each other.
You should also have a viewer open with the original scene displayed. This will
further help you identify the land cover class.
Overlay your classified image (The cluster image file generated when you did the
ISODATA classification) on the original image.
Set all the clustered image's colors to transparent using the Raster Attribute Editor
found under the Viewer menu and changing all the opacity values to 0.
Once you have set all classes to transparent then you can individually color them by
making them opaque (opacity value – 1) particular classes and see where they are on
the image.
Another method may be to use the Utility - Swipe or the Utility - Flicker tools in the
Viewer by opening the classified image on top of the raw data (do not Clear Display
after opening the first raw image).
When you have decided upon the class breakdowns, use the Raster Attribute editor to
assign class names (“Class names” column) and colors to the classification image.
Create the same four classes you used in the supervised classification (i.e., urban,
forest, wetland, and water) and place each of the clusters into one of the classes by
giving it the same color and class name as every other cluster in that class.
Homework
Q3. Create another new map composition containing the completed unsupervised
classification. Make sure the colors are somewhat appropriate to the class type. Include
all appropriate cartographic elements. Capture the print screen of the map and paste in the
word file.
Q4. Compare the advantages and disadvantages between using a supervised and
unsupervised classification approach. When would one approach be more appropriate
than the other?