Assessing OpenStreetMap Road Currency with Sentinel-2 Time Series: Seven High-Growth Sites in Türkiye
Dr and
Abdullah Karaağaç
International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 2026, vol. 12, issue 5, 185-211
Abstract:
OpenStreetMap (OSM) is the most widely used open road dataset, yet how late it records new roads is rarely measured. This study assesses the currency of the OSM major road network at seven high-growth sites in Türkiye by comparing monthly reconstructions of the map from its full edit history with Sentinel-2 summer composites of 2020, 2022, 2024 and 2025. The 16 km × 16 km sites, one per geographical region, were selected purposively among the cells with the strongest growth of major roads in a national screening. A gradient boosting classifier using spectral, index and multiscale Hessian line features was trained on stable OSM roads and evaluated with leave-one-site-out validation. Road segments of 50 m were scored and compared with the date on which each segment entered the map, with inference based on block bootstrap resampling of road stretches. Separability on held-out sites was high (AUC 0.993), but detection declined for roads in open contexts with little built-up cover. Between 30 and 46 percent of the major road length added to OSM after 2021 was already visible in the preceding composite, and 62 percent of the new length showed persistent image evidence before mapping, with a median lag of at least 15.5 months between first image evidence and OSM entry (a lower bound, since completion dates are not observed). Motorways were usually mapped during construction, whereas the delay concentrated on secondary and primary roads. These results were stable under an alternative entry-date definition and when any single site was removed. Candidate centrelines extracted against earlier map versions coincided with later additions four to seven times more often than the spatial baseline. Free 10 m imagery therefore offers an external, repeatable signal of where the OSM major road network is likely to be out of date in rapidly changing areas.
Keywords: OpenStreetMap; Sentinel-2; Road Network; Data Currency; Volunteered Geographic Information; Machine Learning; Türkiye (search for similar items in EconPapers)
Date: 2026
Note: Article URL: https://ijsrcseit.com/home/article/view/CSEIT2612518
References: Add references at CitEc
Citations:
Downloads: (external link)
https://ijsrcseit.com/home/article/view/CSEIT2612518 Article URL (text/html)
https://ijsrcseit.com/home/article/download/CSEIT2612518/CSEIT2612518 Full text (application/pdf)
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:jbh:ijsrcs:v12:y2026:i5:id:2168
DOI: 10.32628/CSEIT2612518
Access Statistics for this article
More articles in International Journal of Scientific Research in Computer Science, Engineering and Information Technology from International Journal of Scientific Research in Computer Science, Engineering and Information Technology
Bibliographic data for series maintained by Pankaj Sharma ().