A neighbourhood-constrainedk-means approach to classify very high spatial resolution hyperspectral imagery
- 1 August 2012
- journal article
- research article
- Published by Taylor & Francis in Remote Sensing Letters
- Vol. 4 (2) , 161-170
Abstract
In classifying very high spatial resolution (VHR) hyperspectral imagery, intra-class variation often adversely affects classification accuracy, mainly due to a low signal-to-noise ratio (SNR) and high spatial heterogeneity. To address this problem, this article develops a neighbourhood-constrained k-means (NC-k-means) algorithm by incorporating the pure neighbourhood index into the traditional k-means algorithm. The performance of the NC-k-means algorithm was assessed through a series of simulated images and a real hyperspectral image. The results indicate that the classification accuracy of NC-k-means algorithm is consistently better than that of the traditional k-means algorithm, in particular for the images with significant spatial autocorrelations among neighbouring pixels.Keywords
This publication has 10 references indexed in Scilit:
- A de-texturing and spatially constrained K-means approach for image segmentationPattern Recognition Letters, 2010
- Recent advances in techniques for hyperspectral image processingRemote Sensing of Environment, 2009
- Initializing Partition-Optimization AlgorithmsIEEE/ACM Transactions on Computational Biology and Bioinformatics, 2009
- Extracting impervious surfaces from medium spatial resolution multispectral and hyperspectral imagery: a comparisonInternational Journal of Remote Sensing, 2008
- Remote Sensing Digital Image Analysis2006
- Integration of spatial and spectral information by means of unsupervised extraction and classification for homogenous objects applied to multispectral and hyperspectral dataIEEE Transactions on Geoscience and Remote Sensing, 2005
- Sensitivity of hyperclustering and labelling land cover classes to Landsat image acquisition dateInternational Journal of Remote Sensing, 2004
- An evolutionary technique based on K-Means algorithm for optimal clustering in RNInformation Sciences, 2002
- Spectral unmixingIEEE Signal Processing Magazine, 2002
- Observation of the Earth and Its Environment1996