Evaluating Crack Identification Performance of 3D Pavement Imaging Systems Using Portable High-Resolution 3D Scanning

Abstract

With the increasing adoption of three-dimensional (3D) pavement imaging systems by highway agencies for automated pavement condition assessment, coupled with the advent of diverse systems from different manufacturers, there is a need for standard procedures for the verification and certification of systems’ performance in regard to distress identification, especially for cracking that is a key contributor for triggering maintenance and rehabilitation activities. Although some procedures were adopted by agencies for a rough verification of cracking identification accuracy using ground reference established subjectively by trained raters, a more rigorous and objective method is needed to match the continuous advancement in the systems’ capabilities and data quality requirements. As portable high-resolution 3D scanning technologies have become commercially available, there is an opportunity to leverage them for establishing a more trustable ground reference for the data quality evaluation. This paper proposes a methodology that uses high-resolution 3D scanners to establish the ground reference for field pavement cracking distress to evaluate the crack identification capability of 3D pavement imaging systems in regard to crack quantity, position, and width. A case study was performed by scanning sample pavement cracking spots using “FARO Arm Quantum S” scanner to collect ground reference images and a 3D pavement imaging system installed on the “Georgia Tech Sensing Van” to collect test images to validate the feasibility of the proposed methodology.