Sparse Multiple Kernel Learning for Hyperspectral Image Classification Using Spatial-Spectral Features

The increase in spatial and spectral resolution of the satellite sensors has provided high-quality data for remote sensing image classification. However, the high-dimensional feature space induced by using many spatial information precludes the use of simple classifiers. This paper proposes to classify the hyperspectral images and simultaneously to learn significant features in such high-dimensional scenarios. Group lasso regularized multiple kernel learning (GLMKL) is used to incorporate extended multi-attribute profile (EMAP) for hyperspectral image classification. We formulate a closed-form solution for optimizing the kernel weights based on the equivalence between group-lasso and MKL, and the derived variant equivalence leads to an efficient algorithm for MKL. Experiments are conducted on three high spatial resolution hyperspectral data sets. The results show that the proposed method achieves better performance for hyperspectral image classification compared to several state-of-the-art algorithms.