Hyperspectral image classification method based on pseudo label assistance

The invention discloses a hyperspectral image classification method based on pseudo label assistance, which mainly solves the problem of low cross-domain hyperspectral image classification accuracy in the prior art, and the implementation scheme of the method comprises the following steps: obtaining and constructing a hyperspectral source domain and target domain data set; respectively extracting feature maps of a source domain and a target domain through a deep neural network, and inputting the feature maps into two dual classifiers with the same structure to obtain respective two prediction probabilities; updating the deep neural network and the dual classifier for the first time by using the prediction probabilities of the source domain and the target domain; constructing a class perception maximum mean difference loss function by using the source domain and target domain feature maps, and performing secondary updating on the primarily updated deep neural network and the dual classifier by minimizing the l