Comparative Analysis of Supervised and Unsupervised Classification on Multispectral Data
The aim of this study is to compare two methods of image classification, i.e. ML (Maximum Likelihood), a supervised method, and ISODATA (Iterative Self- Organizing Data Analysis Technique), an unsupervised method. The former is knowledge-driven, while the latter is data-driven. The former needs a p...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
HIKARI LTD
2013
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/9032/1/ahmadAMS73-76-2013_published.pdf http://eprints.utem.edu.my/id/eprint/9032/ http://www.m-hikari.com/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The aim of this study is to compare two methods of image classification, i.e. ML (Maximum Likelihood), a supervised method, and ISODATA (Iterative Self-
Organizing Data Analysis Technique), an unsupervised method. The former is knowledge-driven, while the latter is data-driven. The former needs a priori knowledge about the study area but the latter does not. In practice, the former can classify land covers with a higher accuracy and therefore is more widely used but there have been very few attempts to investigate this. Here we use both methods
in our study area, Selangor, Malaysia and compare the outcomes by means of qualitative and quantitative analyses to have a better understanding of the underlying reasons that drive the performance of both methods. |
|---|