EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Frequency-Spatial Segmentation Network for High-Resolution Remote Sensing Images

Qiyuan Zhang iD, Jianshun Liu

Sensors · 24.09.2026 · 26 (19) · 6059

DOI: 10.3390/s26196059 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 24.09.2026. Verfasst von Qiyuan Zhang und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Engineering, Sichuan University of Science and Engineering, Yibin 644000, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Sichuan University of Science and Engineering.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Semantic segmentation of high-resolution remote sensing images faces three major challenges in frequency-spatial feature fusion: background clutter mixed into high-frequency components, semantic discontinuities within large homogeneous regions, and loss of fine rigid boundaries caused by convolutional downsampling. Traditional dual-stream networks rely on complex attention mechanisms and directly concatenate heterogeneous features, which often leads to representation conflicts, information loss, and excessive edge smoothing. To address these issues, we propose FSOD-Net, a Frequency-Spatial Feature Decoupling Network built upon the SFFNet architecture. The feature mapping stage is reconstructed to incorporate physical edge priors and frequency-domain attention-based denoising. Specifically, a Spatial Attention Wavelet Transform Feature Decomposer (SA-WTFD) is introduced in the frequency branch to adaptively suppress high-frequency background clutter and enhance discriminative features. In the spatial global branch, a lightweight pyramid pooling module (L-PPM) captures macro-contextual information to fill semantic gaps within areal objects. In the local branch, a fixed Laplacian operator is embedded as an explicit physical edge prior to preserving micro-scale rigid boundaries during downsampling. These three modules operate along independent paths, enabling decoupled yet collaborative multi-domain feature learning. Extensive experiments on the ISPRS Vaihingen and Potsdam datasets demonstrate that FSOD-Net achieves mean Intersection over Union (mIoU) scores of 84.24±0.11% and 86.79±0.09%, respectively, indicating its potential in segmentation accuracy and generalization.

Schlagwörter

Stichwörter der Autor:innen

edge guidanceremote sensingsemantic segmentationspatial-frequency fusionwavelet transform

Autor:innen und Institutionen

  • School of Computer Science and Engineering, Sichuan University of Science and Engineering, Yibin 644000, China.

ORCID-Profile: Qiyuan Zhang

Förderung

  • National Natural Science Foundation of China
  • Sichuan University of Science and Engineering

Zitieren

APA

Zhang, Q., & Liu, J. (2026). A Frequency-Spatial Segmentation Network for High-Resolution Remote Sensing Images. Sensors, 26(19), 6059. https://doi.org/10.3390/s26196059

Vancouver

Zhang Q, Liu J. A Frequency-Spatial Segmentation Network for High-Resolution Remote Sensing Images. Sensors. 2026;26(19):6059. doi:10.3390/s26196059

BibTeX
@article{Zhang2026,
  author = {Zhang, Qiyuan and Liu, Jianshun},
  title = {A Frequency-Spatial Segmentation Network for High-Resolution Remote Sensing Images},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {19},
  pages = {6059},
  doi = {10.3390/s26196059}
}
Kostenlos starten →