Record Details

A context-sensitive clustering technique based on graph-cut initialization and expectation-maximization algorithm

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title A context-sensitive clustering technique based on graph-cut initialization and expectation-maximization algorithm
 
Creator TYAGI, MAYANK
BOVOLO, F
MEHRA, AK
CHAUDHURI, SUBHASIS
BRUZZONE, L
 
Subject graph theory
statistical methods
remote sensing
fuzzy rules
 
Description This letter presents a multistage clustering technique for unsupervised classification that is based on the following: 1) a graph-cut procedure to produce initial segments that are made up of pixels with similar spatial and spectral properties; 2) a fuzzy c-means algorithm to group these segments into a fixed number of classes; 3) a proper implementation of the expectation-maximization (EM) algorithm to estimate the statistical parameters of classes on the basis of the initial seeds that are achieved at convergence by the fuzzy c-means algorithm; and 4) the Bayes rule for minimum error to perform the final classification on the basis of the distributions that are estimated with the EM algorithm. Experimental results confirm the effectiveness of the proposed technique.
 
Publisher IEEE
 
Date 2009-01-19T16:23:06Z
2011-11-25T16:30:20Z
2011-12-26T13:05:31Z
2011-12-27T05:52:39Z
2009-01-19T16:23:06Z
2011-11-25T16:30:20Z
2011-12-26T13:05:31Z
2011-12-27T05:52:39Z
2008
 
Type Article
 
Identifier IEEE Geoscience and Remote Sensing Letters 5 (1), 21-25
1545-598X
http://dx.doi.org/10.1109/LGRS.2007.905119
http://hdl.handle.net/10054/561
http://dspace.library.iitb.ac.in/xmlui/handle/10054/561
 
Language en