Pavement Distress Evaluation using 3D Depth Information from Stereo Vision

Principal Investigators

  • Dr. Ezzatollah Salari
    The University of Toledo

Co-Principal Investigators

  • Dr James Lynch
    University of Detroit Mercy


  • Dr. Eddie Chou
    The University of Toledo
  • Dr. Utpal Dutta
    University of Detroit Mercy

Project Dates:

09/01/2010 to 07/15/2012

Project year: Year 1

MIOH-UTC Project Identifier: TS 43

Focus Area:

  • Research: Intelligent Transportation Systems


Countries such as the United States which have significant temperature changes are continually exposed to extreme heat and cold conditions throughout the year. This results in a high rate of expansion and contraction of pavement surfaces leading to extensive road surface anomalies such as cracking and potholes.  In areas where there is an extremely high level of transport load traveling across roadways, the rapid destruction of road surfaces is inevitable. Therefore, pavement inspection and maintenance becomes a very important part of U.S. Department of Transportation (DOT) spending and the spending of states.  Each state is spending millions of dollars annually to maintain and repair roadways.  Poor road conditions are often the major sources of automobile damage claims, e.g. there were more than 7,500 pothole damage claims in the state of Michigan alone in 2005[1].  Therefore, an automatic surface condition evaluation system is a necessity for our extensive national roadway system.  Given the complexity of pavement surface texture and outdoor lighting conditions, the development of an automated pavement inspection system possesses significant challenges. 

 The performance of most existing image processing based pavement inspection systems is heavily dependent on parameters that are affected by shadows and variations in outdoor lighting conditions among other factors.  Recent advances in stereoscopic imaging offers the potential for road surface quality assessment in 3-Dimentional space.  The proposed project extends the scope of our current MIOH-UTC project by using 3D depth information taken from road surfaces to complement the existing inspection algorithms.  The 3D surface profile generated from stereo images can provide a depth map of a road surface which is viable information needed for the detection and measurement of the potholes and as well as other surface anomalies. 

Final Report:


Total Budget: $79,734


US DOT, The University of Toledo, University of Detroit Mercy.