Abstract:
Full-depth patching is one of the commonly used asphalt pavement maintenance and rehabilitation methods in which deteriorated base and surface layers are repaired to restore strength and improve ride quality. During resurfacing projects, areas requiring full-depth patching are identified and quantified as construction priorities because of the high costs associated with the labor and materials for the procedure. Currently, the manual surveys conducted to identify these locations are time-consuming and labor-intensive. Thus, large projects often cannot easily quantify the full-depth patching need because of the significant labor that would be required. This paper proposes a method that uses emerging 3D laser technology to identify the full-depth patching need by processing and analyzing the pavement distresses automatically extracted from 3D laser images. The proposed method consists of five steps: (1) 3D data acquisition, calibration, and validation, (2) crack detection, (3) crack classification, (4) rutting detection and measurement, and (5) determination of image-based patching need using the established decision tree. A case study of one mile of 3D pavement images, collected from US 80/S.R. 26, was conducted to demonstrate the use and feasibility of the proposed method. Results show the proposed method is capable of correctly classifying 95.4% of the images that show pavements requiring patching and 84.2% of the images showing pavements not requiring patching for a combined accuracy of 94.1%. The method shows promise for identifying patch locations in a cost-effective manner and will save money and time for transportation agencies.
