EUVIMEDEUVIMED entdecken
Crossref

MSG: Robust Multimodal Remote Sensing Image Matching Using Side Window Gaussian Space

Chongyue Zheng, Shanshan Li, Chengyou Wang, Bing Zhang

IEEE Transactions on Geoscience and Remote Sensing · 2025 · 63 · 1-23

DOI: 10.1109/tgrs.2025.3591504 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2025. Verfasst von Chongyue Zheng und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 60 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Robust and accurate image matching and registration are foundational tasks for numerous applications. However, current methods often fail when dealing with multimodal remote sensing images (MRSIs) that exhibit significant spatial geometric differences (SGDs) and nonlinear radiometric differences (NRDs). To address these challenges, this article proposes a novel MRSI matching (MRSIM) method: matching using side window Gaussian space (MSG). MSG leverages an intuitive concept that human visual perception relies heavily on salient features at image edges for precise matching. Specifically, the proposed method: 1) constructs a multiscale side window Gaussian filter scale space (MSGSS) that preserves edge information at different scales while blurring the image; 2) enhances the repeatability of keypoints by performing corner detection on edge maps; 3) increases descriptor robustness by using second-order gradients combined with steerable filtering; and 4) further utilizes a two-stage matching strategy within a constrained search space and designs a new distance, making full use of densely distributed edge keypoints. Quantitative and qualitative experiments conducted on five datasets spanning 957 image pairs across nine multimodal types demonstrate that MSG outperforms nine advanced algorithms (six feature-based methods: SIFT, OS-SIFT, RIFT, CoFSM, HOWP, and POS-GIFT; and three deep learning-based methods: SuperPoint + SuperGlue, LoFTR, and ReDFeat). The results indicate that MSG achieved a number of correct matches (NCMs) much higher than the compared algorithms, with the highest success rate (SR), lowest RMSE, and good time efficiency while achieving both scale and rotation invariance. Codes are available at https://github.com/ZCYla/MSG

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China

Zitieren

APA

Zheng, C., Li, S., Wang, C., & Zhang, B. (2025). MSG: Robust Multimodal Remote Sensing Image Matching Using Side Window Gaussian Space. IEEE Transactions on Geoscience and Remote Sensing, 63, 1-23. https://doi.org/10.1109/tgrs.2025.3591504

Vancouver

Zheng C, Li S, Wang C, Zhang B. MSG: Robust Multimodal Remote Sensing Image Matching Using Side Window Gaussian Space. IEEE Transactions on Geoscience and Remote Sensing. 2025;63:1-23. doi:10.1109/tgrs.2025.3591504

BibTeX
@article{Zheng2025,
  author = {Zheng, Chongyue and Li, Shanshan and Wang, Chengyou and Zhang, Bing},
  title = {MSG: Robust Multimodal Remote Sensing Image Matching Using Side Window Gaussian Space},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2025},
  volume = {63},
  pages = {1--23},
  doi = {10.1109/tgrs.2025.3591504}
}
Kostenlos starten →