EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Symmetrical Learning and Transferring: Efficient Knowledge Distillation for Remote Sensing Image Classification

Huaxiang Song, Junping Xie, Liang Liang, Yan Su, Yao Xiao, Xinyuan Zhang, Yuqi Ouyang, Xinling Li, Siyi Chen, Yucheng Li

Symmetry · 25.06.2025 · 17 (7) · 1002

DOI: 10.3390/sym17071002 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Symmetry, online erschienen am 25.06.2025. Verfasst von Huaxiang Song und 9 weiteren Autor:innen. Die Arbeit wurde bisher 56-mal zitiert (Crossref) und nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hunan Provincial Department of Education’s Scientific Research Project, Research Foundation of Hunan University of Arts and Science, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Knowledge distillation (KD) is crucial for remote sensing image (RSI) classification, particularly as the operating environment in remote sensing is often constrained by hardware limitations. However, prior research has not fully addressed the challenge of leveraging KD to develop lightweight, high-accuracy models for RSI classification. A key issue is the sparse distribution of training data, which often results in asymmetry within the data. This asymmetry impedes the transfer of prior knowledge during the distillation process, diminishing the overall efficacy of KD techniques. To overcome this challenge, we propose a novel, symmetry-enhanced approach that augments the logit-based KD process, improving its effectiveness and efficiency for RSI classification. Our method is distinguished by three core innovations: a symmetrically generative algorithm to enhance the symmetry of the training data, an efficient algorithm for constructing a robust teacher ensemble model, and a quantitative technique for feature alignment. Rigorous evaluations on three benchmark datasets demonstrate that our method outperforms 14 existing KD-based approaches and 30 other state-of-the-art methods. Specifically, the student model trained with our approach achieves accuracy improvements of up to 22.5% while reducing the model size and inference time by as much as 96% and 88%, respectively. In conclusion, this research makes a significant contribution to RSI classification by introducing an efficient and effective data symmetry-driven method to enhance the knowledge transferring efficiency of the logit-based KD process.

Förderung

  • Hunan Provincial Department of Education’s Scientific Research Project
  • Research Foundation of Hunan University of Arts and Science
  • National Natural Science Foundation of China
  • Xuzhou Basic Research Program Dual Carbon Special Project
  • “343” Industrial Development Project of Xuzhou

Zitieren

APA

Song, H., Xie, J., Liang, L., Su, Y., Xiao, Y., Zhang, X., Ouyang, Y., Li, X., Chen, S., & Li, Y. (2025). Symmetrical Learning and Transferring: Efficient Knowledge Distillation for Remote Sensing Image Classification. Symmetry, 17(7), 1002. https://doi.org/10.3390/sym17071002

Vancouver

Song H, Xie J, Liang L, Su Y, Xiao Y, Zhang X, et al. Symmetrical Learning and Transferring: Efficient Knowledge Distillation for Remote Sensing Image Classification. Symmetry. 2025;17(7):1002. doi:10.3390/sym17071002

BibTeX
@article{Song2025,
  author = {Song, Huaxiang and Xie, Junping and Liang, Liang and Su, Yan and Xiao, Yao and Zhang, Xinyuan and Ouyang, Yuqi and Li, Xinling and Chen, Siyi and Li, Yucheng},
  title = {Symmetrical Learning and Transferring: Efficient Knowledge Distillation for Remote Sensing Image Classification},
  journal = {Symmetry},
  year = {2025},
  volume = {17},
  number = {7},
  pages = {1002},
  doi = {10.3390/sym17071002}
}
Kostenlos starten →