Semi-supervised hyperspectral band selection via spectral-spatial hypergraph model

Band selection is an essential step towards effective and efficient hyperspectral image classification. Traditional supervised band selection methods are often hindered by the problem of lacking enough training samples. To address this problem, we propose a semi-supervised band selection method that allows contribution from both labelled and unlabelled hyperspectral pixels. This method first builds a hypergraph model from all hyperspectral samples in order to measure the similarity among pixels. We show that hypergraph can captures relationship among pixels in both spectral and spatial domain. In the second step, a semisupervised learning method is introduced to propagate class labels to unlabelled samples. Then a linear regression model with group sparsity constraint is used for band selection. Finally, hyperspectral pixels with selected bands are used to train a support vector machine classifier. The proposed method is tested on three benchmark datasets. Experimental results demonstrate its advantages over several other band selection methods.