Classification Oriented Semi-supervised Band Selection for Hyperspectral Images

This paper proposes a new framework of band selection for object classification in hyperspectral images. Different from traditional approaches where the selected bands are shared from all classes, in this work, different subsets of bands are selected for different class pairs. Without prior knowledge of spectral database, we estimate the spectral characteristic of objects with the labeled and unlabeled samples, benefiting from the concept of semi-supervised learning. Under the assumption of Gaussian mixture model (GMM), the vectors of mean values and covariance matrices for each class are estimated. The separabilities for all pairs of classes are thus calculated on each band. The bands with the highest separabilities are then selected. To validate our band selection result, support vector machine (SVM) is employed using a strategy of one against one (OAO). Experiments are conducted on a real data set of hyperspectral image, and the results can validate our algorithm.