Dr. Yuxiang Zhang’s recent publication on IEEE Transactions on Circuits and Systems for Video Technology
- leshili
- 7 days ago
- 1 min read
Cross-domain Hyperspectral Image Classification based on Bi-directional Domain Adaptation
Congratulations to Dr. Yuxiang Zhang on publishing a research article on IEEE Transactions on Circuits and Systems for Video Technology (Early Access). The title of the article is “Cross-domain Hyperspectral Image Classification based on Bi-directional Domain Adaptation” and it is now available at https://doi.org/10.1109/TCSVT.2025.3586282, and the code will be available from the website: https://github.com/YuxiangZhang-BIT/IEEE_TCSVT_BiDA
This study introduces a novel Bi-directional Domain Adaptation (BiDA) framework that significantly improves the accuracy of land cover classification using hyperspectral remote sensing data. The proposed approach addresses a key challenge in cross-domain hyperspectral image (HSI) classification by simultaneously extracting domain-invariant features and domain-specific information in the independent adaptive space, thereby enhancing both adaptability and separability to target scenes. Experimental results on cross-temporal and cross-scene datasets demonstrate that BiDA outperforms state-of-the-art methods, with improvements of 3%–5% in a cross-temporal tree species classification task, establishing a new benchmark for domain adaptation in hyperspectral image.
Reference: Y. Zhang, W. Li, W. Jia, M. Zhang, R. Tao and S. Liang, "Cross-domain Hyperspectral Image Classification based on Bi-directional Domain Adaptation," in IEEE Transactions on Circuits and Systems for Video Technology (Early Access), doi: 10.1109/TCSVT.2025.3586282.
Keywords: Hyperspectral Image Classification, Cross-domain, Domain adaptation, Transformer, Feature extraction, Bidirectional control