Artificial Intelligence for Pavement Condition Assessment from 2D/3D Surface Images

While manual quality assurance is inefficient and expensive, the proprietary data storing and processing methods have prevented Texas Department of Transportation (TxDOT) from developing automated methods for data validation. Recently, with the national initialization of standard format for two-dimensional/three-dimensional (2D/3D) pavement surface images and the development of Artificial Intelligence (AI)/Machine Learning (ML) in Computer Vision, TxDOT sees the opportunity of developing new methods for automated pavement condition assessment, with more independence from vendors and their equipment. The main objective of this research is to develop ML-based application software to assess pavement conditions using the standard format 2D/3D pavement surface images. The three main components of this research include the development of a standard format 2D/3D pavement surface image library, a set of ML models for pavement distress measurement, and application software for pavement condition evaluation. The proposed project will assist TxDOT to enhance the quality of the automated pavement condition data, which would eventually help the State of Texas improve its pavement performance.

• Record URL:
  https://library.ctr.utexas.edu/Presto/project=7150

Language

• English

Project

• Status: Active
• Funding: $451875
• Contract Numbers:

  0-7150

• Sponsor Organizations:

  Texas Department of Transportation

  125 E. 11th Street
  Austin, TX  United States  78701-2483
• Managing Organizations:

  Texas Department of Transportation

  125 E. 11th Street
  Austin, TX  United States  78701-2483
• Project Managers:

  Adediwura, Jade

• Performing Organizations:

  Texas State University, San Marcos

  JCK Building, Suite 489
  San Marcos, TX  United States 
• Principal Investigators:

  Wang, Feng

• Start Date: 20220901
• Expected Completion Date: 20250831
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Artificial intelligence; Automatic data collection systems; Computer vision; Evaluation and assessment; Image analysis; Machine learning; Surface course (Pavements)
• Subject Areas: Data and Information Technology; Highways; Maintenance and Preservation; Pavements;

Filing Info

• Accession Number: 01863345
• Record Type: Research project
• Source Agency: Texas Department of Transportation
• Contract Numbers: 0-7150
• Files: RIP, STATEDOT
• Created Date: Nov 3 2022 4:22PM