Scribd
Upload
9 views2 pages

RS Assignment 5 - Unsupervised Classification

The document discusses unsupervised classification algorithms, particularly the ISODATA clustering method, which categorizes data into distinct groups based on similarity. It outlines the steps for implementing this algorithm, including setting parameters for cluster specification and processing images. Additionally, it provides references for further reading on the topic.

Uploaded by

jiteshmaheshwari07
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views2 pages

RS Assignment 5 - Unsupervised Classification

The document discusses unsupervised classification algorithms, particularly the ISODATA clustering method, which categorizes data into distinct groups based on similarity. It outlines the steps for implementing this algorithm, including setting parameters for cluster specification and processing images. Additionally, it provides references for further reading on the topic.

Uploaded by

jiteshmaheshwari07
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Unsupervised Classification

Shivam Kubavat Jitesh Maheshwari Parakh


201030011023 201030011008 Maheshwari
IITRAM, Ahmedabad IITRAM, Ahmedabad 201030011020
IITRAM, Ahmedabad

● Abstract—The primary objective of Report/Map To” Section.


unsupervised classification algorithms is to ● Now select “ISODATA..” in Algorithm
categorize data records into distinct groups, section.
each containing similar members while
being noticeably different from other
groups. This task is a fundamental aspect of
exploratory data mining and is frequently
employed in statistical data analysis.

I. INTRODUCTION
● The "ISO" prefix in the ISO data clustering
algorithm stands for "Iterative Self-Organizing,"
which describes the method of performing
clustering. In this clustering approach, an iterative
process is employed, where, in each iteration, all Fig. 2 – Set ISODATA Cluster specification dialogbox
data samples are assigned to current cluster centres,
and new means are computed for each cluster. ● We can new dialog box in other option write 7
in “number of clusters” and write 100 in
“convergence percentage”.
II. METHOD
● Then go for “Ok”.
● Begin by opening the image in which you intend to ● The output will be as show in fig. 3.
perform unsupervised classification. ● To toggle cluster visibility, use the "Red O"
● Next, go to the "Processor" menu and choose the button on the toolbar.
"Cluster" option.

Fig. 1 – Set Cluster specification dialog box

● Select “Do Not Save” in pop menu.


● Select “Cluster mask file” and select “Image
window overlay” under “Write Cluster Fig.3
IV. REFERENCES
● To get more control over it open Saved image by
“File>Open Image”.
● Then you have to select your image and then click ● https://engineering.purdue.edu/~biehl/MultiSpec/t
open. utorials/MultiSpec_Tutorial_3.pdf

● https://www.usgs.gov/faqs/what-are-band-desig
nations-landsat-satellites

Fig. 4
● New dialog box will open “Set Thematic Display
specification” all the default value are fine and
click ok
● You can see File output in Fig. 5.

III. RESULTS

Fig. 5 Thematic Display

You might also like