The National Academies

NCHRP IDEA 20-30/IDEA 106 [Completed (IDEA)]

Pavement Performance Evaluation and Prediction using Automated Real-Time Crack Detection Classification System
[ NCHRP 20-30 (NCHRP-IDEA) ]

  Project Data
Staff Responsibility: Dr. Inam Jawed

Pavement crack detection and analysis is an important component of pavement management systems. Conventional visual and manual pavement crack analysis techniques are very costly, time-consuming, dangerous, labor-intensive, tedious, and subjective; have a high degree of variability; are unable to provide meaningful quantitative information; and almost always lead to inconsistencies in distress detail over space and across evaluations. An ideal automated crack detection system should detect all types of cracking and other surface distress, of any size and at any collection speed. It should be affordable, easy to operate, and capable of daylight operation. Pavement management programs would be very adequately served by devices that give meaningful, repeatable distress ratings to sections of pavement, providing the critical information for maintenance decisions. Problems with the existing automated crack detection systems are as follows: 
  1. They require special devices (special lights, lasers, etc.) that would increase cost and limit the application of the system or method;
  2. Some systems have very low processing speed and low accuracy; 
  3. They can deal only with certain kinds of distresses; if there is a need to deal with more kinds of distresses, the complexity of the system would greatly increase or it
  4. could not handle the additional work; They cannot achieve real-time processing; some systems can only record in realtime (on film or videotape) and then perform offline processing and analysis. 
  5. The distress severity can only be measured qualitatively.  The final report is available from the National Technical Information Service (NTIS # PB2007-107318).

The final report for this IDEA project can be found at:

To create a link to this page, use this URL: http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=2076