Automatic Road Extraction and Vectorization From Scanned Topographic Maps

Keywords:
Object Detection, Automatic Vectorization, Scanned Maps Processing

Abstract:
The previous practice has shown that older editions of topographic maps of the Military Geographical Institute can be used as a reliable source of data in the process of creating digital topographic maps at a scale of 1:25 000. Since vectorization of the topographic data for the complete scale series of topographic maps has not been done, these maps are used exclusively in the form of raster services as a supplement to primary cartographic sources. Therefore, the large amount of data on these topographic maps can provide insight into how some phenomena moved and developed over time, but remained in raster form unsuitable for processing, analysis, and comparison with vector data obtained from primary cartographic sources and depicting the real character of some occurrences. The paper shows the process by which, using the Python programming language in combination with the ArcScan vectorization tool, line symbols of paved roads are extracted from the 1:25 000 topographic maps and then translated into a vector form suitable for further use. The data obtained in this way become available for multiple applications with a great saving of time, considering that the process is completely automated. A proposal for using the results through a comparison of the higher-order road network between the situation in 1969 and the situation in 2022 is given.

CITATION:

IEEE format

Copy to clipboard

M. Mrlješ, M. Basarić, S. Bakrač, S. Radojčić, “Automatic Road Extraction and Vectorization From Scanned Topographic Maps,” in Sinteza 2023 - International Scientific Conference on Information Technology, Computer Science, and Data Science, Belgrade, Singidunum University, Serbia, 2023, pp. 227-234. doi:10.15308/Sinteza-2023-227-234

APA format

Copy to clipboard

Mrlješ, M., Basarić, M., Bakrač, S., Radojčić, S. (2023). Automatic Road Extraction and Vectorization From Scanned Topographic Maps. Paper presented at Sinteza 2023 - International Scientific Conference on Information Technology, Computer Science, and Data Science. doi:10.15308/Sinteza-2023-227-234

BibTeX format
Download

RefWorks Tagged format
Download