Abstract:Convolutional neural networks (CNNs) and graph convolutional networks (GCNs) have led to promising advancements in hyperspectral image (HSI) classification; however, traditional CNNs with fixed square convolution kernels are insufficiently flexible to handle irregular structures. Similarly, GCNs that employ superpixel nodes instead of pixel nodes may overlook pixel-level features; both networks tend to extract features locally and cause loss of multilayer contextual semantic information during feature extraction due to the fixed kernel. To leverage the strengths of CNNs and GCNs, we propose a multiscale pixel-level and superpixel-level (MPAS)-based HSI classification method. The network consists of two sub-networks for extracting multi-level information of HSIs: a multi-scale hybrid spectral–spatial attention convolution branch (HSSAC) and a parallel multi-hop graph convolution branch (MGCN). HSSAC comprehensively captures pixel-level features with different kernel sizes through parallel multi-scale convolution and cross-path fusion to reduce the semantic information loss caused by fixed convolution kernels during feature extraction and learns adjustable weights from the adaptive spectral–spatial attention module (SSAM) to capture pixel-level feature correlations with less computation. MGCN can systematically aggregate multi-hop contextual information to better model HSIs' spatial background structure using the relationship between parallel multi-hop graph transformation nodes. The proposed MPAS effectively captures multi-layer contextual semantic features by leveraging pixel-level and superpixel-level spectral–spatial information, which improves the performance of the HSI classification task while ensuring computational efficiency. Extensive evaluation experiments on three real-world HSI datasets demonstrate that MPAS outperforms other state-of-the-art networks, demonstrating its superior feature learning capabilities.

Multiscale Principle of Relevant Information for Hyperspectral Image Classification

Multi-Scale Superpixels Dimension Reduction Hyperspectral Image Classification Algorithm Based on Low Rank Sparse Representation Joint Hierarchical Recursive Filtering

Hyperspectral image classification based on multiscale piecewise spectral-spatial attention network

Hyperspectral Image Classification Using Multi-Branch-multi-scale Residual Fusion Network

Multiscale Residual Weakly Dense Network with Attention Mechanism for Hyperspectral Image Classification

Multiscale Union Regions Adaptive Sparse Representation For Hyperspectral Image Classification

Multi-Scale Depthwise Separable Convolutional Neural Network for Hyperspectral Image Classification

Integrating Spectral And Spatial Features For Hyperspectral Image Classification Using Low-Rank Representation

Superpixel-based multi-scale multi-instance learning for hyperspectral image classification

A Modified Locality-Preserving Projection Approach for Hyperspectral Image Classification

The Classification of Hyperspectral Images: A Double-Branch Multi-Scale Residual Network

Spectral-Spatial Hyperspectral Image Classification Via Robust Low-Rank Feature Extraction and Markov Random Field.

Multiscale Pixel-Level and Superpixel-Level Method for Hyperspectral Image Classification: Adaptive Attention and Parallel Multi-Hop Graph Convolution

Toward Effective Hyperspectral Image Classification Using Dual-Level Deep Spatial Manifold Representation.

Spectral-Spatial Discriminant Feature Learning For Hyperspectral Image Classification

Integration of 3-dimensional discrete wavelet transform and Markov random field for hyperspectral image classification.

Multiarea Target Attention for Hyperspectral Image Classification

Multiscale Feature-Learning with a Unified Model for Hyperspectral Image Classification

Multi-Scale Spatial–Spectral Residual Attention Network for Hyperspectral Image Classification

Hyperspectral Classification Based on Coupling Multiscale Super-Pixels and Spatial Spectral Features

Hyperspectral Image Classification via Multiscale Multiangle Attention Network