EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

An Adaptive Remote Sensing Image-Matching Network Based on Cross Attention and Deformable Convolution

Peiyan Chen, Ying Fu, Jinrong Hu, Bing He, Xi Wu, Jiliu Zhou

Electronics · 2023-06-30 · 12 (13) · 2889

DOI: 10.3390/electronics12132889 ↗

At a glance

Journal article in Electronics, published online on 2023-06-30. Written by Peiyan Chen and 5 co-authors. Cited 4 times so far (Crossref); 44 references. The full text is freely available (open access, CC BY 4.0). Funded by: Sichuan Natural Science Foundation, Sichuan Science and Technology Program, Science and Technology Major Project of Sichuan.

AbstractBibliografiske oplysningerCiteRelated studies

Abstract (original language)

There are significant background changes and complex spatial correspondences between multi-modal remote sensing images, and it is difficult for existing methods to extract common features between images effectively, leading to poor matching results. In order to improve the matching effect, features with high robustness are extracted; this paper proposes a multi-temporal remote sensing matching algorithm CMRM (CNN multi-modal remote sensing matching) based on deformable convolution and cross-attention. First, based on the VGG16 backbone network, Deformable VGG16 (DeVgg) is constructed by introducing deformable convolutions to adapt to significant geometric distortions in remote sensing images of different shapes and scales; second, the features extracted from DeVgg are input to the cross-attention module to better capture the spatial correspondence of images with background changes; and finally, the key points and corresponding descriptors are extracted from the output feature map. In the feature matching stage, in order to solve the problem of poor matching quality of feature points, BFMatcher is used for rough registration, and then the RANSAC algorithm with adaptive threshold is used for constraint. The proposed algorithm in this paper performs well on the public dataset HPatches, with MMA values of 0.672, 0.710, and 0.785 when the threshold is selected as 3–5. The results show that compared to existing methods, our method improves the matching accuracy of multi-modal remote sensing images.

Funding

  • Sichuan Natural Science Foundation
  • Sichuan Science and Technology Program
  • Science and Technology Major Project of Sichuan
  • Project of Innovation Ability Enhancement of Chengdu University of Information Technology

Cite

APA

Chen, P., Fu, Y., Hu, J., He, B., Wu, X., & Zhou, J. (2023). An Adaptive Remote Sensing Image-Matching Network Based on Cross Attention and Deformable Convolution. Electronics, 12(13), 2889. https://doi.org/10.3390/electronics12132889

Vancouver

Chen P, Fu Y, Hu J, He B, Wu X, Zhou J. An Adaptive Remote Sensing Image-Matching Network Based on Cross Attention and Deformable Convolution. Electronics. 2023;12(13):2889. doi:10.3390/electronics12132889

BibTeX
@article{Chen2023,
  author = {Chen, Peiyan and Fu, Ying and Hu, Jinrong and He, Bing and Wu, Xi and Zhou, Jiliu},
  title = {An Adaptive Remote Sensing Image-Matching Network Based on Cross Attention and Deformable Convolution},
  journal = {Electronics},
  year = {2023},
  volume = {12},
  number = {13},
  pages = {2889},
  doi = {10.3390/electronics12132889}
}
Start for free →