A Kinect-Based Approach for 3D Pavement Surface Reconstruction and Cracking Recognition

Pavement surface distress conditions are critical inputs for quantifying roadway infrastructure serviceability. Numerous computer-aided automatic examination techniques have been deployed for pavement distress condition assessments, such as digital image processing methods. However, their effectiveness and applicability are impeded due to information losses in 2-D image combination processes or extremely high costs in 3-D geo-referenced data set. In this paper, a cost-effective Kinect-based approach is proposed for 3-D pavement surface reconstruction and cracking recognition. We propose a comprehensive computational solution for the detection and recognition of pavement distress feature identification. Various cracking measurements such as alligator cracking, traverse cracking, longitudinal cracking, and so on. are identified and recognized for their severity examinations based on associated geometrical features. The experimental results indicate that this method is effective in reducing data collection costs and extracting analytical information on pavement cracking measurements. The research findings confirm that the proposed approach provides a viable, applicable solution to an automatic pavement surface condition detection and evaluation. The proposed methodology is transferable for pavement surface reconstruction and distress condition detection based on the other 3-D cloud point data. It provides an alternative inexpensive complement to existing pavement examination methodologies.