A Large-Scale Remote Sensing Dataset and VLM-based Algorithm for Fine-Grained Road Hierarchy Classification

TL;DR

This paper introduces SYSU-HiRoads, a large-scale remote sensing dataset with hierarchical road annotations, and RoadReasoner, a vision-language-geometry framework that improves multi-grade road mapping accuracy and semantic consistency.

Contribution

The work provides a new large-scale hierarchical road dataset and a novel VLM-based algorithm for detailed road hierarchy classification from remote sensing imagery.

Findings

RoadReasoner achieves 72.6% OA in hierarchy classification.

The dataset enables joint training of segmentation and hierarchy tasks.

RoadReasoner outperforms existing road extraction methods.

Abstract

In this work, we present SYSU-HiRoads, a large-scale hierarchical road dataset, and RoadReasoner, a vision-language-geometry framework for automatic multi-grade road mapping from remote sensing imagery. SYSU-HiRoads is built from GF-2 imagery covering 3631 km2 in Henan Province, China, and contains 1079 image tiles at 0.8 m spatial resolution. Each tile is annotated with dense road masks, vectorized centerlines, and three-level hierarchy labels, enabling the joint training and evaluation of segmentation, topology reconstruction, and hierarchy classification. Building on this dataset, RoadReasoner is designed to generate robust road surface masks, topology-preserving road networks, and semantically coherent hierarchy assignments. We strengthen road feature representation and network connectivity by explicitly enhancing frequency-sensitive cues and multi-scale context. Moreover, we perform hierarchy inference at the skeleton-segment level with geometric descriptors and geometry-aware textual prompts, queried by vision-language models to obtain linguistically interpretable grade decisions. Experiments on SYSU-HiRoads and the CHN6-CUG dataset show that RoadReasoner surpasses state-of-the-art road extraction baselines and produces accurate and semantically consistent road hierarchy maps with 72.6% OA, 64.2% F1 score, and 60.6% SegAcc. The dataset and code will be publicly released to support automated transport infrastructure mapping, road inventory updating, and broader infrastructure management applications.