Automated and Contactless Identification of Asphalt Pavement Surface Friction Based on Computer Vision Image-based 3D Reconstruction and Machine Learning Techniques

The objective of this research is to develop an automated and contactless 3D scanning and analysis method to identify hot-mix asphalt (HMA) pavement surface friction using close-range photogrammetry. To achieve this objective, a pipeline of (a) computer vision methods for image-based 3D reconstruction and (b) machine learning methods will be used to generate 3D point cloud and mesh models, characterize the geometry and appearance of the pavement surfaces, and infer the observed surface friction. Pavement surface texture will be obtained based on any feed of overlapping images and videos. To streamline the reality capture process, an automatic image acquisition system will be employed. Building on prior work by Ibrahim et al. (2022), a pipeline of image-based 3D reconstruction, multi-view dense reconstruction and mesh modelling will be used to reconstruct the geometry and appearance of the pavement surface texture. Through a user interface, the obtained texture information will be labelled with ground-truth data, and the mapping will be stored in a data set along with the corresponding friction measurements.

Language

• English

Project

• Status: Active
• Funding: $750000
• Contract Numbers:

  R27-247

• Sponsor Organizations:

  Illinois Department of Transportation

  Bureau of Materials and Physical Research
  126 East Ash Street
  Springfield, IL  United States  62704-4766
• Managing Organizations:

  University of Illinois, Urbana-Champaign

  Illinois Center for Transportation
  1611 Titan Drive
  Rantoul, IL  United States  61866
• Principal Investigators:

  Al-Qadi, Imad

• Start Date: 20220701
• Expected Completion Date: 20260630
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Computer vision; Data collection; Detection and identification; Friction; Hot mix asphalt; Machine learning; Photogrammetry; Skid resistance
• Subject Areas: Data and Information Technology; Highways; Materials; Pavements;

Filing Info

• Accession Number: 01850660
• Record Type: Research project
• Source Agency: Illinois Department of Transportation
• Contract Numbers: R27-247
• Files: RIP, STATEDOT
• Created Date: Jun 29 2022 4:31PM