The National Academies

NCHRP 20-102(06) [Completed]

Road Markings for Machine Vision
[ NCHRP 20-102 (Impacts of Connected Vehicles and Automated Vehicles on State and Local Transportation Agencies--Task-Order Support) ]

  Project Data
Funds: $200,000
Research Agency: Texas A&M Transportation Institute
Principal Investigator: Adam Pike
Effective Date: 7/13/2016
Completion Date: 8/12/2018

The objective of the research was to develop information on the performance characteristics of pavement markings that affect the ability of machine vision systems to recognize them. This information is expected to be useful to the AASHTO/SAE Working Group as they develop guidelines and criteria. Markings to be studied are center lines [Manual on Uniform Traffic Control Devices (MUTCD) Section 3B.01], no-passing zone markings [MUTCD 3B.02], lane lines [MUTCD 3B.04, including dotted extensions for ramps], and edge lines [MUTCD 3B.06]. Factors to be considered include pavement marking presence, type of marking (flush, raised [MUTCD 3B.11, 3B.12, 3B.13, 3B.14], recessed, or temporary [MUTCD 6F.77, 6F.78, 6F.79]), contrast between the pavement and the marking during daytime conditions (including contrast markings, different angles of the sun, and the effects of shadows), retroreflectivity of the marking during nighttime conditions (including the effect of illumination) and different weather conditions (rain, fog, etc.), pavement uniformity (including sealed cracks and patching), vehicle speed, and the impact of other substances on the road such as snow, sand, salt, and water. It is intended that the work include a range of forward-facing machine vision systems so that the current technologies and those on the horizon can be accommodated.


The Contractor's Final Report is available and is being used by the joint AASHTO SAE Working Group.
Task 1. Conduct a kick-off meeting with the panel on June 1, 2016 at SAE headquarters in Troy, Michigan.
Task 2. Review policies and specifications that relate to the performance of pavement markings relative to machine vision systems.
Task 3. Identify the specific testing conditions that will be used in Task 4.
Task 4. Conduct testing to generate the data needed to meet the project objective.
Task 5. Process the data from the Task 4 field study and conduct the analyses to establish the results needed to identify performance characteristics of pavement markings that affect the ability of machine vision systems to recognize them.
Task 6. Prepare the final documents, which will include:
  • Performance data on the machine vision recognition of different marking approaches for lane departure warning (LDW) and line keeping applications (LKA) that clearly lays out the assumptions and limitations of the research approach and the level of confidence in the results;
  • Prioritized list of marking characteristics that are important to the LDW and LKA performance of machine vision systems;
  • Insofar as practical, recommendations for good pavement marking practices that do not degrade the performance of the markings for human vision;
  • Estimation of the costs and safety impacts that could be expected through implementation of the various pavement marking recommendations;
  • If appropriate, recommended changes to the MUTCD; and
  • Gaps in research that still need to be addressed.
Task 7. Finalize the deliverables of the project based on the panel review.

To create a link to this page, use this URL: http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=4004