2026, Vol. 13, No. 1. - go to content...

Permanent address of this page - https://resources.today/en/22nzor126.html

Метаданные этой статьи так же доступны на русском языке

DOI: 10.15862/22NZOR126 (https://doi.org/10.15862/22NZOR126)

Full article in PDF format (file size: 1.2 MB)

For citation:

Zuev V.A. Automated system for assessing the visual informativeness of three-dimensional maps of urban areas based on artificial intelligence methods. Russian journal of resources, conservation and recycling. 2026; 13(1). Available at: https://resources.today/PDF/22NZOR126.pdf (in Russian). DOI: 10.15862/22NZOR126

Automated system for assessing the visual informativeness of three-dimensional maps of urban areas based on artificial intelligence methods

Zuev Vladislav Aleksandrovich
Siberian State University of Geosystems and Technologies, Novosibirsk, Russia
E-mail: veuzdalv638@gmail.com
RSCI: https://elibrary.ru/author_profile.asp?id=1163924

Abstract. The article presents the development and experimental testing of an automated system for assessing the visual informativeness of three-dimensional maps of urban areas based on artificial intelligence methods. The relevance of the study is обусловлена the need to improve the efficiency and objectivity of analyzing the quality of three-dimensional cartographic visualizations, since traditional evaluation methods require significant time expenditures and depend on the subjective judgments of experts. The proposed system is based on a previously developed qualimetric methodology and provides automatic calculation of a set of visual informativeness indicators.The architecture of the system includes modules for loading and preprocessing 3D models, generating perspective images, analyzing visual characteristics using computer vision algorithms and deep learning methods, calculating integral indicators, and visualizing the results. Algorithms have been implemented for the automatic determination of key parameters, including the number of visible objects, color contrast coefficient, level of detail, characteristics of map symbols, texture quality, readability, as well as visualization parameters (viewing angle and field of view).The system was tested on a dataset of three-dimensional models of an urban territory with different levels of detail and visualization parameters. The results demonstrate a high correlation between automated and expert assessments while significantly reducing the analysis time — from 30–60 minutes to approximately 2 minutes per visualization variant. The developed system can be used for automated quality control of three-dimensional maps, optimization of visualization parameters, and integration into cartographic and geoinformation platforms.

Keywords: three-dimensional maps; automated evaluation; artificial intelligence; computer vision; visual informativeness; qualimetry; deep learning

Download article in PDF format