EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

D2Former: Dual-Domain Transformer for Change Detection in VHR Remote Sensing Images

Huanhuan Zheng, Hui Liu, Lei Lu, Shiyin Li, Jiyan Lin

Electronics · 2024-06-05 · 13 (11) · 2204

DOI: 10.3390/electronics13112204 ↗

At a glance

Journal article in Electronics, published online on 2024-06-05. Written by Huanhuan Zheng and 4 co-authors. Cited 4 times so far (Crossref); 79 references. The full text is freely available (open access, CC BY 4.0). Funded by: Yulin Science and Technology Bureau Industry University Research Project, National Natural Science Foundation Regional Fund Project.

AbstractBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Computational intelligence technologies have been extensively applied for the interpretation of remote sensing imagery. Recently, the computational-intelligence-based Transformer change detection (CD) approach has attracted increasing attention. However, the current Transformer-based CD method can better capture global features, but there is no good solution for the loss of local detail information. For this reason, introducing semantic and frequency information from the perspective of a dual-domain can be beneficial for improving the representation of detailed features to improve CD performance. To overcome this limitation, a dual-domain Transformer (D2Former) is proposed for CD. Firstly, we adopt a semantic tokenizer to capture the semantic information, which promotes the enrichment and refinement of semantic change information in the Transformer. Secondly, a frequency tokenizer is introduced to acquire the frequency information of the features, which offers the proposed D2Former another aspect and dimension to enhance the ability to detect change information. Therefore, the proposed D2Former employs dual-domain tokenizers to acquire and fuse the feature representation with rich semantic and frequency information, which can refine the features to acquire more fine-grained CD ability. Extensive experiments on three CD benchmark datasets demonstrate that the proposed D2Former obviously outperforms some other existing approaches. The results present the competitive performance of our method on the WHU-CD, LEVIR-CD, and GZ-CD datasets, for which it achieved F1-Score metrics of 92.85%, 90.60%, and 87.02%, respectively.

Funding

  • Yulin Science and Technology Bureau Industry University Research Project
  • National Natural Science Foundation Regional Fund Project

Cite

APA

Zheng, H., Liu, H., Lu, L., Li, S., & Lin, J. (2024). D2Former: Dual-Domain Transformer for Change Detection in VHR Remote Sensing Images. Electronics, 13(11), 2204. https://doi.org/10.3390/electronics13112204

Vancouver

Zheng H, Liu H, Lu L, Li S, Lin J. D2Former: Dual-Domain Transformer for Change Detection in VHR Remote Sensing Images. Electronics. 2024;13(11):2204. doi:10.3390/electronics13112204

BibTeX
@article{Zheng2024,
  author = {Zheng, Huanhuan and Liu, Hui and Lu, Lei and Li, Shiyin and Lin, Jiyan},
  title = {D2Former: Dual-Domain Transformer for Change Detection in VHR Remote Sensing Images},
  journal = {Electronics},
  year = {2024},
  volume = {13},
  number = {11},
  pages = {2204},
  doi = {10.3390/electronics13112204}
}
Start for free →