Enhancing semantic information of vector road networks using tile maps
International Journal of Digital Earth · 02.01.2026 · 19 (1)
DOI: 10.1080/17538947.2026.2658294 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Digital Earth, online erschienen am 21.04.2026. Verfasst von Mengwei Zhang und 6 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Semantic information is a critical component of spatial data; however, it is often incomplete, limiting the usability of vector road networks. Tile maps are freely accessible, information-rich raster data containing complete road annotations. However, optical character recognition (OCR) applied to tile maps often produces fragmented text boxes and detects non-road text owing to resolution limitations and varying text orientations. To address these challenges, a method for enhancing the semantic information of vector road networks using tile maps is proposed in this study. The vector road network and tile maps are co-registered and segmented. An improved text extraction and merging strategy based on PaddleOCR is developed using geometric constraints to reconstruct fragmented road-name text. Furthermore, a geometry-constrained text classification model, termed the RoadSense Classifier, is introduced to filter non-road text and retain accurate road names. Matching the extracted road text to its corresponding road segments enriches the semantic information. Experiments using tile map data from the Third Ring Road area of Zhengzhou City and the main urban area of Xinxiang show that the proposed method achieves a precision of 98.86% and 97.64%, recall of 85.07% and 83.98%, and F1-score of 91.45% and 89.32%, respectively, improving vector road attribute completeness. Article highlights • Integrated Geometric-Semantic Enhancement Method for Vector Road Networks: A novel method integrating geometric constraints and semantic analysis is proposed to enhance the semantic information of vector road networks using tile maps, addressing the problem of incomplete road name attributes.• Improved Method for PaddleOCR-Based Text Extraction and Merging with Geometric Constraints: An improved PaddleOCR-based text extraction and merging strategy is developed, which merges fragmented text boxes via parallel edge distance and optimizes results based on skeleton morphology, improving the completeness of road name extraction.• Novel RoadSense Classifier Model for Road Name Text Classification Incorporating Geometric Constraints: A geometry-constrained RoadSense Classifier is introduced, integrating geometric constraint-based preliminary filtering with BERT-based encoding (augmented by data augmentation), FGM adversarial training, and multi-head self-attention. This synergy effectively filters non-road text and ensures accurate road name recognition.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhang, M., Qian, H., Liu, X., Liu, C., Gong, X., Wang, X., & Huang, Z. (2026). Enhancing semantic information of vector road networks using tile maps. International Journal of Digital Earth, 19(1). https://doi.org/10.1080/17538947.2026.2658294
Vancouver
Zhang M, Qian H, Liu X, Liu C, Gong X, Wang X, et al. Enhancing semantic information of vector road networks using tile maps. International Journal of Digital Earth. 2026;19(1). doi:10.1080/17538947.2026.2658294
BibTeX
@article{Zhang2026,
author = {Zhang, Mengwei and Qian, Haizhong and Liu, Xingui and Liu, Chengyi and Gong, Xianyong and Wang, Xiao and Huang, Zhekun},
title = {Enhancing semantic information of vector road networks using tile maps},
journal = {International Journal of Digital Earth},
year = {2026},
volume = {19},
number = {1},
doi = {10.1080/17538947.2026.2658294}
}
EUVIMED