EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Land Cover Classification of UAV Remote Sensing Based on Transformer–CNN Hybrid Architecture

Tingyu Lu, Luhe Wan, Shaoqun Qi, Meixiang Gao

Sensors · 2023-06-02 · 23 (11) · 5288

DOI: 10.3390/s23115288 ↗

At a glance

Journal article in Sensors, published online on 2023-06-02. Written by Tingyu Lu and 3 co-authors. Institution: College of Geographical Sciences, Harbin Normal University, Harbin 150025, China. Major MEDLINE topics: Remote Sensing Technology, Electric Power Supplies. Cited 48 times so far (Crossref); 40 references. The full text is freely available (open access, CC BY 4.0). Funded by: National Natural Science Foundation of China.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

High-precision land cover maps of remote sensing images based on an intelligent extraction method are an important research field for many scholars. In recent years, deep learning represented by convolutional neural networks has been introduced into the field of land cover remote sensing mapping. In view of the problem that a convolution operation is good at extracting local features but has limitations in modeling long-distance dependence relationships, a semantic segmentation network, DE-UNet, with a dual encoder is proposed in this paper. The Swin Transformer and convolutional neural network are used to design the hybrid architecture. The Swin Transformer pays attention to multi-scale global features and learns local features through the convolutional neural network. Integrated features take into account both global and local context information. In the experiment, remote sensing images from UAVs were used to test three deep learning models including DE-UNet. DE-UNet achieved the highest classification accuracy, and the average overall accuracy was 0.28% and 4.81% higher than UNet and UNet++, respectively. It shows that the introduction of a Transformer enhances the model fitting ability.

Subjects

MeSH terms (MEDLINE)

Remote Sensing TechnologyElectric Power SuppliesIntelligenceNeural Networks, ComputerImage Processing, Computer-Assisted

Author keywords

CNNUAV remote sensingdeep learningsemantic segmentationtransformer

Authors and institutions

  • College of Geographical Sciences, Harbin Normal University, Harbin 150025, China.
  • Heilongjiang Province Key Laboratory of Geographical Environment Monitoring and Spatial Information Service in Cold Regions, Harbin Normal University, Harbin 150025, China.
  • Department of Geography and Spatial Information Techniques, Ningbo University, Ningbo 315211, China.

Funding

  • National Natural Science Foundation of China

Cite

APA

Lu, T., Wan, L., Qi, S., & Gao, M. (2023). Land Cover Classification of UAV Remote Sensing Based on Transformer–CNN Hybrid Architecture. Sensors, 23(11), 5288. https://doi.org/10.3390/s23115288

Vancouver

Lu T, Wan L, Qi S, Gao M. Land Cover Classification of UAV Remote Sensing Based on Transformer–CNN Hybrid Architecture. Sensors. 2023;23(11):5288. doi:10.3390/s23115288

BibTeX
@article{Lu2023,
  author = {Lu, Tingyu and Wan, Luhe and Qi, Shaoqun and Gao, Meixiang},
  title = {Land Cover Classification of UAV Remote Sensing Based on Transformer–CNN Hybrid Architecture},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {11},
  pages = {5288},
  doi = {10.3390/s23115288}
}
Start for free →