A manifold learning based feature extraction method for hyperspectral classification

T Manifold learning methods have widely used in ordinary image processing domain. It has many advantages, depending on the different formulation of the manifold. Hyperspectral images are kind of images acquired by air-borne or space-born platforms. This paper introduces a novel manifold learning method into hyperspectral classification. The purpose is to fully utilize the spectral and spatial information from hyperspectral images to get confidential landcover and land use class results.