EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Frequency-Aware Refinement Network with Multi-Scale Fusion for Remote Sensing Change Detection

Xu Zhang iD, Yue Du, Zeyu Zhang, Kaihua Zhang iD

Sensors · 03.06.2026 · 26 (11) · 3538

DOI: 10.3390/s26113538 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 03.06.2026. Verfasst von Xu Zhang und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Digital Low-Altitude, Suzhou Polytechnic University, Suzhou 215104, China. Die Arbeit nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Seventh Batch of Science and Technology Development Plan (Agriculture) Project of Suzhou, Jiangsu Higher Vocational College Teachers Visiting Study Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Remote sensing change detection (RSCD) identifies land cover variations by comparing bi-temporal images. However, conventional methods relying solely on RGB domain information often fail to distinguish changed objects from visually similar backgrounds, especially in complex scenarios. To overcome this limitation, we propose a frequency-aware refinement network (FARNet) that follows a coarse-to-fine strategy. In the first stage, we design a frequency-aware module (FAM) that learns frequency domain information to identify the blurred boundaries of changed objects that resemble the background, enabling coarse localization of potential change regions. In the second stage, recognizing that high-resolution RGB domain details provide richer spatial information than frequency-domain features, we design a refinement fusion module (RFM) that leverages these RGB details to correct and refine segmentation boundaries, ensuring precise detection. Finally, edge loss is applied to preserve high-frequency details, enhancing the precision of change detection. Extensive experiments on benchmark datasets demonstrate that FARNet significantly outperforms existing methods, achieving superior accuracy and robustness in complex change detection scenarios.

Schlagwörter

Stichwörter der Autor:innen

change detectioncoarse-to-finefrequency-aware refinementmulti-scaleremote sensing

Autor:innen und Institutionen

  • School of Digital Low-Altitude, Suzhou Polytechnic University, Suzhou 215104, China.
  • Information Center, Suzhou Polytechnic University, Suzhou 215104, China.
  • Department of Applied Mathematics and Computer Science, Duke Kunshan University, Suzhou 215316, China.
  • School of Automation, Southeast University, Nanjing 210096, China.

ORCID-Profile: Xu Zhang, Kaihua Zhang

Förderung

  • the Seventh Batch of Science and Technology Development Plan (Agriculture) Project of Suzhou
  • Jiangsu Higher Vocational College Teachers Visiting Study Program

Zitieren

APA

Zhang, X., Du, Y., Zhang, Z., & Zhang, K. (2026). Frequency-Aware Refinement Network with Multi-Scale Fusion for Remote Sensing Change Detection. Sensors, 26(11), 3538. https://doi.org/10.3390/s26113538

Vancouver

Zhang X, Du Y, Zhang Z, Zhang K. Frequency-Aware Refinement Network with Multi-Scale Fusion for Remote Sensing Change Detection. Sensors. 2026;26(11):3538. doi:10.3390/s26113538

BibTeX
@article{Zhang2026,
  author = {Zhang, Xu and Du, Yue and Zhang, Zeyu and Zhang, Kaihua},
  title = {Frequency-Aware Refinement Network with Multi-Scale Fusion for Remote Sensing Change Detection},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {11},
  pages = {3538},
  doi = {10.3390/s26113538}
}
Kostenlos starten →