Some issues in the classification of DAIS hyperspectral data

Abstract

Classification accuracy depends on a number of factors, of which the nature of the training samples, the number of bands used, the number of classes to be identified relative to the spatial resolution of the image and the properties of the classifier are the most important. This paper evaluates the effects of these factors on classification accuracy using a test area in La Mancha, Spain. High spectral and spatial resolution DAIS data were used to compare the performance of four classification procedures (maximum likelihood, neural network, support vector machines and decision tree). There was no evidence to support the view that classification accuracy inevitably declines as the data dimensionality increases. The support vector machine classifier performed well with all test data sets. The use of the orthogonal MNF transform resulted in a decline in classification accuracy. However, the decision‐tree approach to feature selection worked well. Small increases in classifier accuracy may be obtained using more sophisticated techniques, but it is suggested here that greater attention should be given to the collection of training and test data that represent the range of land surface variability at the spatial scale of the image.

Acknowledgements

Mahesh Pal thanks the Association of Commonwealth Universities (ACU), London, for providing a scholarship for this study under the Commonwealth Scholarship scheme. The DAIS data were collected and processed by DLR and were kindly made available by Professor J. Gumuzzio of the Autonomous University of Madrid. The support vector machine program was made available by AT&T, Royal Holloway College, University of London. Computing facilities were provided by the School of Geography, University of Nottingham. Special thanks are due to two anonymous referees for their comments in improving the quality of this paper.