PAFNet: A Parallel Attention Fusion Network for Water Body Extraction of Remote Sensing Images

Abstract

Highlights: What are the main findings? A Parallel Attention Fusion Network (PAFNet) is proposed to refine features, apply parallel attention, and fuse multi-level semantics for accurate water body extraction. Experiments on four representative remote sensing datasets demonstrate that PAFNet delivers more precise boundary delineation and stronger robustness compared with existing methods. What are the implications of the main findings? The proposed method reduces misclassification of spectrally similar features (e.g., shadows, built-up areas), enabling reliable water body mapping in diverse scenarios. PAFNet shows strong potential for real-world applications such as flood monitoring, watershed management, and ecological conservation. Water body extraction plays a crucial role in remote sensing, supporting applications such as environmental monitoring and disaster prevention. Although Deep Convolutional Neural Networks (DCNNs) have achieved remarkable progress, their hierarchical architectures often introduce channel redundancy and hinder the joint representation of fine spatial structures and high-level semantics, leading to ineffective feature fusion and poor discrimination of water features. To address these limitations, a Parallel Attention Fusion Network (PAFNet) is proposed to achieve more effective multi-scale feature aggregation through parallel attention and adaptive fusion. First, the Feature Refinement Module (FRM) employs multi-branch asymmetric convolutions to extract multi-scale features, which are subsequently fused to suppress channel redundancy and preserve fine spatial details. Then, the Parallel Attention Module (PAM) applies spatial and channel attention in parallel, improving the discriminative representation of water features while mitigating interference from spectrally similar land covers. Finally, a Semantic Feature Fusion Module (SFM) integrates adjacent multi-level features through adaptive channel weighting, thereby achieving precise boundary recovery and robust noise suppression. Extensive experiments conducted on four representative datasets (GID, LandCover.ai, QTPL, and LoveDA) demonstrate the superiority of PAFNet over existing state-of-the-art methods. Specifically, the proposed model achieves 94.29% OA and 95.95% F1-Score on GID, 86.17% OA and 88.70% F1-Score on LandCover.ai, 98.99% OA and 98.96% F1-Score on QTPL, and 89.01% OA and 85.59% F1-Score on LoveDA.