Essential Hierarchical Information That Warrants Attention: A Semi-Supervised Hyperspectral Hierarchical Classification Network

The hierarchical classification method can utilize the hierarchical structure to achieve fine classification from coarse-grained to fine-grained. Hierarchical classification has demonstrated outstanding performance across various fields, benefiting from the continuous expansion of dataset sizes. Nevertheless, the majority of current hyperspectral image (HSI) deep learning classification methods employ a flat classification strategy, disregarding the hierarchical structure present in HSI data. Can hierarchical structure information assist deep learning in completing HSI classification tasks? In this article, we validated the feasibility of using hierarchical structure information to assist deep learning in completing HSI classification tasks through intuitive experiments and then proposed a semi-supervised hierarchical classification network (Semi-HCN) to achieve hierarchical classification of HSIs. Semi-HCN first generates a large number of pseudo-labels through a fast self-training (FST) module for subsequent network training and then extracts hierarchical features through a multibranch network and progressively consolidates them across layers. Finally, hierarchical constraints are applied to the network output through hierarchical cross-entropy loss (CE Loss). Experiments on three real datasets have demonstrated that Semi-HCN can achieve competitive classification performance with limited training samples. The code will be released at https://github.com/jinyaoWHU/Semi-HCN