Scribd
Upload
95 views

Lab 5 Unsupervised Classification

This document provides instructions for performing unsupervised classification of a digital image into 10 classes. The process involves using an unsupervised classification tool in GIS software to automatically classify pixels in the image. The results are then analyzed and edited by assigning color codes and renaming classes to match land cover types in the area, with reference to other materials. The final output is a new classified image map displaying 10 distinct land cover classes represented by different colors.

Uploaded by

Muhammad Aiman
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
95 views

Lab 5 Unsupervised Classification

This document provides instructions for performing unsupervised classification of a digital image into 10 classes. The process involves using an unsupervised classification tool in GIS software to automatically classify pixels in the image. The results are then analyzed and edited by assigning color codes and renaming classes to match land cover types in the area, with reference to other materials. The final output is a new classified image map displaying 10 distinct land cover classes represented by different colors.

Uploaded by

Muhammad Aiman
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

SUG330/SUG351

LAB 5 : UNSUPERVISED CLASSIFICATION

1.0 Introduction

Unsupervised classification is a process to classify surface features by assigning the digital


values of surfaces according to the given class. In this task, you are asked to perform 10
classes of surface features. While performing this task, other references might be required to
assist your analysis.

2.0 Objective

The main objective of unsupervised classification is to classify digital image of your study area
into 10 classes. The classes should be representing the classes on the ground.

3.0 Methodology
1. Prepare your image from your previous laboratory session on your current viewer.
2. At Raster menu at main menu, click on Unsupervised Classification button.
(Unsupervised)

Unsupervised Classification window will appear at your screen.

Siti Nor Maizah Saad


JSUG/FSPU/UiTM Perlis
SUG330/SUG351

Output File

Set number of classes

3. After that, fill in your output file in Unsupervised Classification window and set
Number of Class to 10 classes.
4. Click OK to proceed with the classification process.
5. After that, open your Unsupervised Classification output in the viewer. A grey scale
image will appear on your viewer.
6. Check the attributes of the classes by clicking at Table in main menu and select
Show Attribute at its dropped down function.

Siti Nor Maizah Saad


JSUG/FSPU/UiTM Perlis
SUG330/SUG351

Grey Scale Image

Show Attributes
Table
7. A table of attributes will appear as below. Rearrange the column position by
adjusting the Colum Properties

Siti Nor Maizah Saad


JSUG/FSPU/UiTM Perlis
SUG330/SUG351

8. Change the classes colour in the table to make the classes more informative and
attractive. While rename the class name and changing the colour, you should use a
relevant reference to assist your classification.

Change to

9. After finish editing the color of classes, save the table and re open the saved image
at. Last result should be appearing on your screen and the image also already edited
with suitable color.

Siti Nor Maizah Saad


JSUG/FSPU/UiTM Perlis
SUG330/SUG351

Last Output

Siti Nor Maizah Saad


JSUG/FSPU/UiTM Perlis

You might also like