D3FNet: A Differential Attention Fusion Network for Fine-Grained Road Structure Extraction in Remote Perception Systems

2025-08-21

AI Summary

Extracting narrow roads—characterized by low width, topological discontinuities, and severe occlusion—from high-resolution remote sensing imagery remains challenging. To address this, we propose an enhanced D-LinkNet incorporating three key innovations: (1) a differential attention dilation module that amplifies weak road responses while suppressing background interference; (2) a dual-stream decoding mechanism that jointly optimizes local details and global contextual information; and (3) a multi-scale atrous convolution strategy to better model irregular, narrow road structures. Evaluated on DeepGlobe and CHN6-CUG benchmarks, our method achieves superior IoU and recall over state-of-the-art approaches. Notably, it demonstrates significantly improved continuity and robustness in heavily occluded and topologically fragmented regions. These results validate the effectiveness of our design in enabling fine-grained, accurate road reconstruction under complex real-world conditions.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Distributed Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Extracting narrow roads from high-resolution remote sensing imagery remains a significant challenge due to their limited width, fragmented topology, and frequent occlusions. To address these issues, we propose D3FNet, a Dilated Dual-Stream Differential Attention Fusion Network designed for fine-grained road structure segmentation in remote perception systems. Built upon the encoder-decoder backbone of D-LinkNet, D3FNet introduces three key innovations:(1) a Differential Attention Dilation Extraction (DADE) module that enhances subtle road features while suppressing background noise at the bottleneck; (2) a Dual-stream Decoding Fusion Mechanism (DDFM) that integrates original and attention-modulated features to balance spatial precision with semantic context; and (3) a multi-scale dilation strategy (rates 1, 3, 5, 9) that mitigates gridding artifacts and improves continuity in narrow road prediction. Unlike conventional models that overfit to generic road widths, D3FNet specifically targets fine-grained, occluded, and low-contrast road segments. Extensive experiments on the DeepGlobe and CHN6-CUG benchmarks show that D3FNet achieves superior IoU and recall on challenging road regions, outperforming state-of-the-art baselines. Ablation studies further verify the complementary synergy of attention-guided encoding and dual-path decoding. These results confirm D3FNet as a robust solution for fine-grained narrow road extraction in complex remote and cooperative perception scenarios.

Problem

• Extracting narrow roads from high-resolution remote sensing imagery
• Addressing fragmented topology and frequent occlusions in roads
• Improving continuity in fine-grained road structure segmentation

Innovation

• Differential Attention Dilation Extraction module
• Dual-stream Decoding Fusion Mechanism
• Multi-scale dilation strategy mitigating artifacts