Machine learning techniques for pavement condition evaluation
Pavement management systems play a significant role in country's economy since road authorities are concerned about preserving their priceless road assets for a longer time to save maintenance costs. An essential part of such systems is how to collect and analyze pavement condition data. This paper reviews the state-of-the-art techniques in pavement condition data evaluation using machine learning techniques, more specifically, the application of machine learning methods: image classification, object detection, and segmentation in pavement distress assessment is investigated. Furthermore, the pavement automated data collection tools and pavement condition indices have been reviewed from the lens of machine learning applications. The review concludes that the overall trends in pavement condition evaluation is to apply machine learning techniques although there are some limitations not only in detection of few pavement distresses with complicated patterns but also in indication of the severity and density of distresses leading to avenues for future research.
- Record URL:
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/09265805
-
Supplemental Notes:
- © 2022 Elsevier B.V. All rights reserved. Abstract reprinted with permission of Elsevier.
-
Authors:
- Sholevar, Nima
- Golroo, Amir
- Esfahani, Sahand Roghani
- Publication Date: 2022-4
Language
- English
Media Info
- Media Type: Web
- Features: Figures; References; Tables;
- Pagination: 104190
-
Serial:
- Automation in Construction
- Volume: 136
- Issue Number: 0
- Publisher: Elsevier
- ISSN: 0926-5805
- Serial URL: http://www.sciencedirect.com/science/journal/09265805
Subject/Index Terms
- TRT Terms: Image processing; Machine learning; Monitoring; Pavement maintenance; Pavement management systems
- Subject Areas: Highways; Maintenance and Preservation; Pavements;
Filing Info
- Accession Number: 01840276
- Record Type: Publication
- Files: TRIS
- Created Date: Mar 25 2022 12:36PM