In recent years, Vision Transformers (ViTs) have gained significant traction in the field of hyperspectral image classification due to their advantages in modeling long-range dependency relationships between spectral bands and spatial pixels. However, after stacking multiple Transformer encoders, challenges pertaining to information degradation may emerge during the forward propagation. That is to say, existing Transformer-based methods exhibit certain limitations in retaining and effectively utilizing information throughout their forward transmission. To tackle these challenges, this paper proposes a novel dual-branch spatial–spectral Transformer model that incorporates similarity propagation (DBSSFormer-SP). Specifically, this model first employs a Hybrid Pooling Spatial Channel Attention (HPSCA) module to integrate global information by pooling across different dimensional directions, thereby enhancing its ability to extract salient features. Secondly, we introduce a mechanism for transferring similarity attention that aims to retain and strengthen key semantic features, thus mitigating issues associated with information degradation. Additionally, the Spectral Transformer (SpecFormer) module is employed to capture long-range dependencies among spectral bands. Finally, the extracted spatial and spectral features are fed into a multilayer perceptron (MLP) module for classification. The proposed method is evaluated against several mainstream approaches on four public datasets. Experimental results demonstrate that DBSSFormer-SP exhibits excellent classification performance.

Hyperspectral image classification is an important research topic in the field of remote sensing. Its core objective is to accurately identify different land-cover categories by jointly exploiting the spatial information and spectral characteristics of pixels. However, the complex and diverse spatial structures of ground objects in hyperspectral images make it challenging for existing methods to fully capture intricate spatial patterns. To address this issue, this paper proposes a dual-branch spatial–spectral transformer with spatial feature enhancement under a spatial–spectral joint representation learning framework. Specifically, in the spatial branch, a deformable spatial feature fusion module is introduced, which combines grouped deformable convolution with Transformer to achieve fine-grained modeling of irregular spatial structures. Subsequently, a global spatial semantic aggregation module is designed to further integrate spatial features and enhance their discriminative capability. In the spectral branch, Transformer is employed to model long-range spectral dependencies, while a convolutional feed-forward network is introduced to strengthen the representation ability of critical spectral information. Experimental results demonstrate that the proposed method achieves competitive classification performance on four benchmark datasets.