Frequency-Guided Denoising Network for Semantic Segmentation of Remote Sensing Images

Li X, Xu F, Zhang J, Zhang H, Lyu X, Liu F, Gao H, Kaup A (2026)

Publication Type: Journal article

Publication year: 2026

Journal

IEEE Transactions on Geoscience and Remote Sensing Institute of Electrical and Electronics Engineers (IEEE)

Book Volume: 64

Article Number: 5400217

DOI: 10.1109/TGRS.2025.3648408

Abstract

Semantic segmentation of high-resolution remote sensing images (RSIs) remains challenging due to the degradation of high-frequency (HF) semantic cues during convolutional encoding and the lack of frequency consistency in multistage feature fusion. To address these issues, we propose a frequency-guided denoising network (FreDNet) that explicitly enhances frequency-sensitive representations throughout the segmentation process. Specifically, we introduce the denoising-bypassed residual block (DRB), which incorporates a frequency-guided cross-stage fusion module (FDM) and a frequency-aware fusion module (FFM) to suppress frequency-domain noise while preserving edge structures. Furthermore, we design a frequency-guided cross-stage fusion module (FCFM) that leverages frequency intensity response maps to adaptively fuse encoder and decoder features. These components work collaboratively to enhance the frequency robustness and spatial consistency of the segmentation predictions. Extensive experiments on three challenging benchmarks, ISPRS Vaihingen, ISPRS Potsdam, and LoveDA, demonstrate that FreDNet achieves superior performance, surpassing the latest state-of-the-art (SOTA) approaches by up to 0.8% in mean intersection over union (mIoU) and 0.9% in overall accuracy (OA), while maintaining a lightweight inference cost. In addition, ablation study confirms the contribution of each component of FreDNet.

Authors with CRIS profile

Andre Kaup Lehrstuhl für Multimediakommunikation und Signalverarbeitung (LMS)

Involved external institutions

Hohai University CN China (CN) The University of Hong Kong (HKU) / 香港大學 HK Hong Kong (HK) Griffith University AU Australia (AU)

How to cite

APA:

Li, X., Xu, F., Zhang, J., Zhang, H., Lyu, X., Liu, F.,... Kaup, A. (2026). Frequency-Guided Denoising Network for Semantic Segmentation of Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing, 64. https://doi.org/10.1109/TGRS.2025.3648408

MLA:

Li, Xin, et al. "Frequency-Guided Denoising Network for Semantic Segmentation of Remote Sensing Images." IEEE Transactions on Geoscience and Remote Sensing 64 (2026).

BibTeX: Download