Please use this identifier to cite or link to this item: https://doi.org/10.1109/LGRS.2009.2025059
Title: Unsupervised change detection in satellite images using principal component analysis and κ-means clustering
Authors: Celik, T. 
Keywords: κ-means clustering
Change detection
Multitemporal satellite images
Optical images
Principal component analysis (PCA)
Remote sensing
Issue Date: Oct-2009
Citation: Celik, T. (2009-10). Unsupervised change detection in satellite images using principal component analysis and κ-means clustering. IEEE Geoscience and Remote Sensing Letters 6 (4) : 772-776. ScholarBank@NUS Repository. https://doi.org/10.1109/LGRS.2009.2025059
Abstract: In this letter, we propose a novel technique for unsupervised change detection in multitemporal satellite images using principal component analysis (PCA) and κ-means clustering. The difference image is partitioned into h ×h nonoverlapping blocks. S,S≤h2, orthonormal eigenvectors are extracted through PCA of h×h nonoverlapping block set to create an eigenvector space. Each pixel in the difference image is represented with an S-dimensional feature vector which is the projection of h×h difference image data onto the generated eigenvector space. The change detection is achieved by partitioning the feature vector space into two clusters using κ-means clustering with κ = 2 and then assigning each pixel to the one of the two clusters by using the minimum Euclidean distance between the pixel's feature vector and mean feature vector of clusters. Experimental results confirm the effectiveness of the proposed approach. © 2009 IEEE.
Source Title: IEEE Geoscience and Remote Sensing Letters
URI: http://scholarbank.nus.edu.sg/handle/10635/95380
ISSN: 1545598X
DOI: 10.1109/LGRS.2009.2025059
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

237
checked on Jul 16, 2018

WEB OF SCIENCETM
Citations

165
checked on Jun 20, 2018

Page view(s)

218
checked on Jun 29, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.