Hyperspectral Image Classification Based on Stacked Principal Components Analysis

In this paper, a simple hierarchical model called Stacked Principal Components Analysis (SPCA) is proposed for hyperspectral image (HSI) classification. It can learn discriminative spectral-spatial features. Specifically, SPCA learns useful high-level features by alternating between spectral and spatial feature learning stages. Finally, Kernel-based Extreme Learning Machine (KELM) is used to obtain strong generalization. Extensive experiments are performed on AVIRIS Indian Pines and University of Pavia to validate the effectiveness of our method. The proposed method is simple and effective.