ML Models for Winter Road Surface Condition Recognition in the Case of Insufficient Coverage

2022 63rd International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS 2022): Proceedings 2022
Viesturs Pavlovs, Martins Zviedris, Dainis Dosbergs

To maintain roads during the winter season, it is critical to determine their surface conditions, for efficient allocation of cleaning efforts such as plowing or salting. Current techniques for determining road surface conditions include the use of expensive roadside weather stations or sending trained personnel for on-site inspection. However, these stations are only placed on some roads, and they have a limited coverage. This paper proposes a method to determine road surface conditions in places with insufficient coverage.

Atslēgas vārdi
instance segmentation | interpolation | machine learning | mask r-cnn | xgboost
DOI
10.1109/ITMS56974.2022.9937101
Hipersaite
https://ieeexplore.ieee.org/document/9937101

Pavlovs, V., Zviedris, M., Dosbergs, D. ML Models for Winter Road Surface Condition Recognition in the Case of Insufficient Coverage. No: 2022 63rd International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS 2022): Proceedings, Latvija, Riga, 6.-7. oktobris, 2022. Piscataway: IEEE, 2022, 1.-4.lpp. ISBN 979-8-3503-9986-8. e-ISBN 979-8-3503-9985-1. ISSN 2771-6953. e-ISSN 2771-6937. Pieejams: doi:10.1109/ITMS56974.2022.9937101

Publikācijas valoda
English (en)