Comparison of multisource image fusion methods and land cover classification

Abstract

The aim of this study is to explore the performances of different data fusion techniques for the enhancement of urban features and evaluate the features obtained by the fusion techniques in terms of separation of urban land cover classes when multisource images are under consideration. For the data fusion, multiplicative method, Brovey transform, principal component analysis (PCA), Gram–Schmidt fusion, wavelet-based fusion and Elhers fusion are used and the results are compared. Of these methods, the best result is obtained by the use of the optical/synthetic aperture radar (SAR) wavelet-based fusion. The classification methods of multisource images, statistical maximum likelihood classification (MLC) and the knowledge-based method are used and the results are compared. The knowledge-based method is based on a hierarchical rule-based approach and it uses a hierarchy of rules describing different conditions under which the actual classification has to be performed. Overall, the research indicates that multisource information can significantly improve the interpretation and classification of land-cover types and the knowledge-based method is a powerful tool in the production of a reliable land-cover map.

Acknowledgement

A part of this research was conducted under the sponsorship for Humboldt Fellows.