Interpretability of Transformer Structures Based on Attribution Inference for Polarimetric SAR Image Classification
Remote Sensing · 08.11.2026 · 18 (21) · 3840
DOI: 10.3390/rs18213840 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 08.11.2026. Verfasst von Lijia Dong und 4 weiteren Autor:innen. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Basic Research Program of Shaanxi, Innovation Foundation for Doctor Dissertation of Northwestern Polytechnical University.
Abstract (Originalsprache)
Transformer-based deep networks have gained prominence in Polarimetric Synthetic Aperture Radar (PolSAR) image classification due to their strong capability for modeling long-range dependencies. However, existing Transformer-based methods fail to reveal the polarimetric scattering characteristics and their spatial interactions in classification decisions, which leads to limited interpretability and reduced reliability of the models. To address this challenge, this paper proposes an interpretability analysis framework based on attribution inference for Transformer-based PolSAR image classification. Specifically, an attribution inference module is first constructed by integrating structural dependency relationships with gradient-based saliency information, enabling explicit simulation of polarimetric scattering features and their interactions within the model. To enhance the interpretability of PolSAR classification models, we further conduct a full-chain analysis of the Transformer’s internal structure based on the attribution inference matrix, rather than relying on qualitative visualization from a single attention map. This integrated approach enables precise localization of critical attention heads and information flows, while simultaneously identifying discriminative regions that exert a decisive influence on the prediction results. To validate the proposed method, quantitative and qualitative experiments are performed on three PolSAR image classification datasets, which clearly demonstrate the effectiveness of the proposed interpretability framework.
Förderung
- Natural Science Basic Research Program of Shaanxi
- Innovation Foundation for Doctor Dissertation of Northwestern Polytechnical University
Zitieren
APA
Dong, L., Geng, J., Zhang, Y., Zhou, Y., & Jiang, W. (2026). Interpretability of Transformer Structures Based on Attribution Inference for Polarimetric SAR Image Classification. Remote Sensing, 18(21), 3840. https://doi.org/10.3390/rs18213840
Vancouver
Dong L, Geng J, Zhang Y, Zhou Y, Jiang W. Interpretability of Transformer Structures Based on Attribution Inference for Polarimetric SAR Image Classification. Remote Sensing. 2026;18(21):3840. doi:10.3390/rs18213840
BibTeX
@article{Dong2026,
author = {Dong, Lijia and Geng, Jie and Zhang, Yuhang and Zhou, Yuhan and Jiang, Wen},
title = {Interpretability of Transformer Structures Based on Attribution Inference for Polarimetric SAR Image Classification},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {21},
pages = {3840},
doi = {10.3390/rs18213840}
}
EUVIMED