Center-Pixel Guided Dual-Path Multihead Attention Fusion Transformer for Hyperspectral Image Classification
IEEE Transactions on Geoscience and Remote Sensing · 2026 · 64 · 1-18
DOI: 10.1109/tgrs.2026.3662353 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2026. Verfasst von Haobo Jiang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 67 Literaturangaben. Gefördert von: Fundamental Research Funds for the Central Universities of South–Central Minzu University, Hubei Province Technology Innovation Special Fund Project.
Abstract (Originalsprache)
The performance of hyperspectral image classification (HSIC) relies on the deep integration of spatial and spectral information. Although mainstream Transformer architectures can capture global context, they often overlook the key inductive bias that “the central pixel determines the category,” and also incur high computational costs. Moreover, traditional linear fusion strategies (such as concatenation or element-wise addition) struggle to capture the complex nonlinear interactions between spatial and spectral dimensions. To address these issues, this article proposes a central-pixel guided dual-path multihead attention transformer (CPFormer). The model first introduces a spectral–spatial feature embedding (SSFE) module, which generates robust initial representations while preserving inherent spectral characteristics. Guided by the central pixel-guided dual-path multihead attention (CP-DMA) module is then proposed: the spatial path uses the central pixel as the query and incorporates 2-D Gaussian positional encoding to enhance local perception; the spectral path employs depthwise separable convolution and bidirectional spectral decay prior to model spectral (channel) continuity. Furthermore, CP-DMA adopts element-wise multiplication for spatial–spectral fusion, enabling adaptive and nonlinear feature interaction. Information plane analysis further confirms that, compared with linear fusion, the multiplication strategy can more effectively compress redundant information and drive feature representations closer to the “minimum sufficient statistic” in information theory. Extensive experiments on four benchmark datasets under limited training samples demonstrate that CPFormer outperforms state-of-the-art methods in classification performance. The source code will be available at: https://github.com/Jianghaob/CPFormer
Förderung
- Fundamental Research Funds for the Central Universities of South–Central Minzu University
- Hubei Province Technology Innovation Special Fund Project
Zitieren
APA
Jiang, H., Gao, Q., Chen, H., & Zhou, B. (2026). Center-Pixel Guided Dual-Path Multihead Attention Fusion Transformer for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 64, 1-18. https://doi.org/10.1109/tgrs.2026.3662353
Vancouver
Jiang H, Gao Q, Chen H, Zhou B. Center-Pixel Guided Dual-Path Multihead Attention Fusion Transformer for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing. 2026;64:1-18. doi:10.1109/tgrs.2026.3662353
BibTeX
@article{Jiang2026,
author = {Jiang, Haobo and Gao, Qiankun and Chen, Haobing and Zhou, Bin},
title = {Center-Pixel Guided Dual-Path Multihead Attention Fusion Transformer for Hyperspectral Image Classification},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
year = {2026},
volume = {64},
pages = {1--18},
doi = {10.1109/tgrs.2026.3662353}
}
EUVIMED