FSG-Net: Frequency-Spatial Synergistic Gated Network for High-Resolution Remote Sensing Change Detection

2025-09-08

AI Summary

Addressing two key challenges in high-resolution remote sensing change detection—severe radiometric false alarms (e.g., caused by illumination or seasonal variations) and difficulty in fusing shallow-detail and deep-semantic features—this paper proposes the Frequency-Spatial Gated Network (FSGNet). Methodologically: (i) a difference-aware wavelet interaction module is introduced in the frequency domain to suppress radiometric pseudo-changes; (ii) a time-spatial collaborative attention mechanism is designed in the spatial domain to enhance the saliency of genuine changes; and (iii) a lightweight gated fusion unit bridges semantic gaps across multi-scale features. Evaluated on CDD, GZ-CD, and LEVIR-CD benchmarks, FSGNet achieves F1 scores of 94.16%, 89.51%, and 91.27%, respectively—outperforming state-of-the-art methods. It delivers precise boundary localization and significantly reduces false positives.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Change detection from high-resolution remote sensing images lies as a cornerstone of Earth observation applications, yet its efficacy is often compromised by two critical challenges. First, false alarms are prevalent as models misinterpret radiometric variations from temporal shifts (e.g., illumination, season) as genuine changes. Second, a non-negligible semantic gap between deep abstract features and shallow detail-rich features tends to obstruct their effective fusion, culminating in poorly delineated boundaries. To step further in addressing these issues, we propose the Frequency-Spatial Synergistic Gated Network (FSG-Net), a novel paradigm that aims to systematically disentangle semantic changes from nuisance variations. Specifically, FSG-Net first operates in the frequency domain, where a Discrepancy-Aware Wavelet Interaction Module (DAWIM) adaptively mitigates pseudo-changes by discerningly processing different frequency components. Subsequently, the refined features are enhanced in the spatial domain by a Synergistic Temporal-Spatial Attention Module (STSAM), which amplifies the saliency of genuine change regions. To finally bridge the semantic gap, a Lightweight Gated Fusion Unit (LGFU) leverages high-level semantics to selectively gate and integrate crucial details from shallow layers. Comprehensive experiments on the CDD, GZ-CD, and LEVIR-CD benchmarks validate the superiority of FSG-Net, establishing a new state-of-the-art with F1-scores of 94.16%, 89.51%, and 91.27%, respectively. The code will be made available at https://github.com/zxXie-Air/FSG-Net after a possible publication.

Problem

• Reducing false alarms from radiometric variations in remote sensing change detection
• Bridging semantic gap between deep and shallow features for better fusion
• Improving boundary delineation in high-resolution Earth observation applications

Innovation

• Frequency domain processing to reduce false alarms
• Spatial attention to enhance genuine change regions
• Gated fusion unit to bridge semantic gaps