A New Feature Fusion Method for Hyperspectral Image Classification

In this paper, we propose a filter bank for feature extraction of hyperspectral image when the number of training samples is small. The designed filter bank tries to extract discriminant features and is not sensitive to rotation. The number of features is small. Thus, the number of bands that extract features will be increased. The extracted spatial features and the spectral ones are stacked to each other. Finally, the stacked features are classified by support vector machine (SVM). Experimental results on two popular data sets, namely, Pavia University and Salinas, show that the proposed method is superior to some of the state-of-the-art spatial-spectral hyperspectral image classification methods.