S2A-Swin: Spectral Smoothing–Guided Spectral–Spatial Windows with Generative Augmentation for Hyperspectral Image Classification Under Class Imbalance and Limited Labels
Remote Sensing · 19.03.2026 · 18 (6) · 935
DOI: 10.3390/rs18060935 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 19.03.2026. Verfasst von Baisen Liu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Science and Technology Project of the Department of Transportation of Heilongjiang Province, the Longjiang Project Young Goose Innovation Team Support Program, the Natural Science Foundation of Heilongjiang Province for Key projects.
Abstract (Originalsprache)
Hyperspectral image (HSI) classification faces the challenges of scarce labeled data and severe class imbalance, which limits the effective training and generalization capabilities of models. To address these issues, we propose S2A-Swin, a joint spatial–spectral hybrid Swin Transformer framework. First, we develop a spectral–spatial conditional generative adversarial network (SSC-cGAN), which combines spectral and spatial smoothing regularizers to synthesize class-specific image patches, thus alleviating the problems of data scarcity and class imbalance while maintaining spectral continuity and local spatial structure consistent with real data. Second, we introduce a dimension-aware hybrid Transformer module, which adds local windows along the spectral dimension to the standard spatial window, thereby facilitating cross-dimensional feature interactions and ensuring that each spectral band is modeled using the local spatial context for more efficient joint spatial–spectral modeling. In this module, attention mechanisms for spectral and spatial windows are applied alternately (“cross-sequence” attention mechanisms), the execution order of which is guided by hyperspectral prior knowledge to enhance cross-dimensional representation learning. This module is embedded in the lightweight Swin backbone and extends the traditional spatial window mechanism through spectral window attention, capturing spectral continuity while maintaining spatial structure consistency. Extensive experiments on multiple datasets demonstrate that, compared to mainstream CNN and Transformer baselines on four benchmark datasets, the proposed method achieves overall accuracy (OA) improvements of 2.45%, 7.05%, 5.17%, and 0.85%.
Förderung
- the Science and Technology Project of the Department of Transportation of Heilongjiang Province
- the Longjiang Project Young Goose Innovation Team Support Program
- the Natural Science Foundation of Heilongjiang Province for Key projects
Zitieren
APA
Liu, B., Chen, J., Zhang, W., Dang, Z., Li, X., & Kong, W. (2026). S2A-Swin: Spectral Smoothing–Guided Spectral–Spatial Windows with Generative Augmentation for Hyperspectral Image Classification Under Class Imbalance and Limited Labels. Remote Sensing, 18(6), 935. https://doi.org/10.3390/rs18060935
Vancouver
Liu B, Chen J, Zhang W, Dang Z, Li X, Kong W. S2A-Swin: Spectral Smoothing–Guided Spectral–Spatial Windows with Generative Augmentation for Hyperspectral Image Classification Under Class Imbalance and Limited Labels. Remote Sensing. 2026;18(6):935. doi:10.3390/rs18060935
BibTeX
@article{Liu2026,
author = {Liu, Baisen and Chen, Jianxin and Zhang, Wulin and Dang, Zhiming and Li, Xinyao and Kong, Weili},
title = {S2A-Swin: Spectral Smoothing–Guided Spectral–Spatial Windows with Generative Augmentation for Hyperspectral Image Classification Under Class Imbalance and Limited Labels},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {6},
pages = {935},
doi = {10.3390/rs18060935}
}
EUVIMED