Graph-Structured Convolution-Guided Continuous Context Threshold-Aware Networks for Hyperspectral Image Classification

Although convolutional neural networks (CNNs) have shown superior performance to traditional machine learning algorithms for hyperspectral image (HSI) classification tasks, the ability of traditional CNNs to model remote dependencies in the spatial orientation of HSIs is still limited, and they always extract similar low-level features, leading to feature redundancy. To cope with this limitation, this article proposes a novel multiorder statistical representation-guided graph convolution and continuous context threshold-aware network for the classification of HSIs with limited training samples. Initially, the spectral–spatial information is separately modeled using first-order features and second-order pooling operators. Secondly, we graph-structured the patch features, and by employing a random walk transition probability matrix, the graph-structured convolution can mine more discriminative directional features. In addition, a continuous context threshold-aware network is designed to model multidimensional spatial relationships, which enhances the feature representation of graph features. Specifically, the cross-attention mechanism is used to calculate the attention weights in the vertical and horizontal directions, and the features are divided into two levels—important and secondary—by solving the cosine distance between feature vectors; the former is retained and the latter is punished. Extensive experiments on multiple HSI datasets demonstrated that the proposed method delivers competitive performance. The code will be available at
https://github.com/vivitsai/GSC-CCTA
.