Hyperspectral image classification method and system based on spectrum features and spatial feature fusion

The invention provides a hyperspectral image classification method and system based on spectrum features and spatial feature fusion, and belongs to the field of remote sensing image classification. The method comprises the steps: carrying out the extraction of the spatial features of a hyperspectral image according to a region connectivity of the hyperspectral image; carrying out the denoising of the spectrum features and spatial features; carrying out the fusion of the spectrum features and the spatial features; predicting the class of pixels in an unknown class on the hyperspectral image through a classification model. The system comprises a spatial feature extraction module, a feature denoising module, a feature fusion module and a classification module. The beneficial effects of the invention are that the method improves the classification precision of the hyperspectral image through the low-rank expression and feature fusion of the spectrum features and spatial features of the hyperspectral image, and is