Ground feature classification method and system

The invention provides a ground feature classification method and system, and relates to the technical field of image processing, and the method comprises the steps: constructing a sample data set based on an obtained multispectral remote sensing image of a designated region and DEM data; extracting shallow multi-modal pixel features in the sample data set, and stacking the pixel features to obtain a stacking result; a graph convolutional neural network and a residual network are used to analyze the stacking result, spectral space depth features and terrain depth features are obtained and fused, and fused depth features are obtained; and inputting the fused depth feature into a classifier to obtain a ground feature label of the specified area output by the classifier. By obtaining the fusion depth features, the deep-level diversified ground feature features in the data can be utilized, the ground features can be represented better, the classifier can identify the ground features more accurately and perform fi