Scribd
Upload
113 views

Lab 6 Supervised Classification

This document provides instructions for performing a supervised land use classification on an image using Erdas Imagine software. It describes drawing training samples of different land use classes in the image, recording the sample signatures, running the supervised classification using the signatures, and checking the resulting classified image. The objective is to classify land use features in a study area using supervised classification.

Uploaded by

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

Lab 6 Supervised Classification

This document provides instructions for performing a supervised land use classification on an image using Erdas Imagine software. It describes drawing training samples of different land use classes in the image, recording the sample signatures, running the supervised classification using the signatures, and checking the resulting classified image. The objective is to classify land use features in a study area using supervised classification.

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

SUG330/351 AP120

LAB 6 : SUPERVISED CLASSIFICATION

1. Introduction
A supervised classification is used to identify information classes of interest in the image
which is called training samples or training sites. Image processing software is then used to
develop a statistical characterization of the reflectance for each information class. This
technique is more suitable if the analyst have prior knowledge about study area.

2. Objective

The objective of this task is to classify land use features at study area using supervised
classification technique.

3. Methodology
1. Prepare your data in Erdas Imagine viewer.
2. At Raster menu, select Classification – Supervised Classification – Signature
Editor

3. Signature Editor will appear on your screen. This box allows you to record the name of
your training samples.

Siti Nor Maizah Saad


Jabatan Sains Ukur dan Geomatik
FSPU, UiTM Perlis
SUG330/351 AP120

4. At main menu, select Drawing menu. You will see drawing option on its menu bar.
Drawing icon allows you to draw training samples of your image.

5. By using polygon or any drawing icon, draw samples of features at your study area.
Click Add to insert the sample detail in Signature Editor Box.

6. Write the name of all classes in the Signature Editor table.

Siti Nor Maizah Saad


Jabatan Sains Ukur dan Geomatik
FSPU, UiTM Perlis
SUG330/351 AP120

Class
name

7. Save your signature data and and AOI layer

8. After that, at Raster menu, select Classification - Supervised Classification

Siti Nor Maizah Saad


Jabatan Sains Ukur dan Geomatik
FSPU, UiTM Perlis
SUG330/351 AP120

9. At Supervised Classification window, fill in data input(.img), data output(.img) and


signature data (.sig).

10. Click at Attribute Option and select Standard Deviation. After that click Close.

11. Click OK to proceed with classification process. Once the process successful, open
Supervised image and check its attributes. You can edit the colour of classes and class
name.

Siti Nor Maizah Saad


Jabatan Sains Ukur dan Geomatik
FSPU, UiTM Perlis
SUG330/351 AP120

4. Result

Siti Nor Maizah Saad


Jabatan Sains Ukur dan Geomatik
FSPU, UiTM Perlis

You might also like