Record Details

Unsupervised classification of remote sensing data using graph cut-based initialization

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title Unsupervised classification of remote sensing data using graph cut-based initialization
 
Creator TYAGI, M
MEHRA, AK
CHAUDHURI, S
BRUZZONE, L
 
Description In this paper we propose a multistage unsupervised classifier which uses graph-cut to produce initial segments which are made up of pixels with similar spectral properties, subsequently labelled by a fuzzy c-means clustering algorithm into a known number of classes. These initial segmentation results are used as a seed to the expectation maximization (EM) algorithm. Final classification map is produced by using the maximum likelihood (ML) classifier, performance of which is quite good as compared to other unsupervised classification techniques.
 
Publisher SPRINGER-VERLAG BERLIN
 
Date 2011-10-23T18:52:26Z
2011-12-15T09:11:17Z
2011-10-23T18:52:26Z
2011-12-15T09:11:17Z
2005
 
Type Article; Proceedings Paper
 
Identifier PATTERN RECOGNITION AND MACHINE INTELLIGENCE, PROCEEDINGS,3776,206-211
3-540-30506-8
0302-9743
http://dspace.library.iitb.ac.in/xmlui/handle/10054/15205
http://hdl.handle.net/100/1975
 
Source 1st International Conference on Pattern Recognition and Machine Intelligence,Kolkata, INDIA,DEC 20-22, 2005
 
Language English