Skip to main content
ICTD 2025 Main banner
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

Pavements 2025

Pavement Design, Evaluation, Preservation, Maintenance, and Rehabilitation

POSTER RECEPTION (View list of posters here)

Scalable Method for Pavement Texture Evaluation Using Computer Vision Techniques

Tuesday, June 10, 2025
4:45 PM – 6:00 PM MT
Abstract: Pavement texture plays a critical role in highway safety due to its impact on tire-pavement friction. Texture data are usually collected using expensive high-speed laser scanners, which may create a barrier for network-level data collection. Structure from Motion (SfM), which is a photogrammetry technique, has been used in recent research to measure pavement texture due to its relatively lower cost, but it is limited to small patches. This is due to limitations in the image acquisition techniques and the methods to resolve the SfM scale ambiguity. This research presents a framework for capturing images over long distances rather than small patches while preventing distortions that are usually associated with planar surfaces, like bowling. RTK GNSS technology was utilized to resolve the scale ambiguity using AprilTag ground control points as well as a mobile module attached to the camera. While using the mobile module was a more scalable approach, ground control using AprilTag yielded more accurate results due to better equipment quality. Mean profile depth measured with the SfM-based approach was compared to that from AMES 9400HD laser scanner. The results were within 15% of those from the laser scanner. With more tuning, validation, and advanced equipment, the methodology can be extended to network application.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • Upon completion, the participant will be able to understand the limitations of the existing research on using computer vision techniques to measure pavement texture.
  • Upon completion, the participant will be able to describe the techniques that can be used to produce measurable point clouds and list the advantages and disadvantages of each.
  • Upon completion, the participant will be able to understand the limitation of this research and the steps that are being taken to reach a more accurate and implementable solution.