Integration of spatial-spectral information for hyperspectral image classification

Classification of hyperspectral image data has drawn much attention in recent years. Consequently, it contains not only spectral information of objects, but also spatial arrangement of objects. The most established Hyperspectral classifiers are based on the observed spectral signal, and ignore the spatial relations among observations. Information captured in neighboring locations may provide useful supplementary knowledge for analysis. To combine the spectral and spatial information in the classification process, in this paper, an integration of spatial-spectral information for hyperspectral classification method is proposed. Based on this measure, a collaborative classification method is proposed, which integrates the spectral and spatial autocorrelation during the decisionmaking process. The trials of our experiment are conducted on Washington DC Mall hyperspectral imagery. Quantitative measures of local consistency (smoothness) and global labeling, along with class maps, demonstrate the benefits of applying this method for unsupervised classification.