United States Department of Transportation
ROSA P serves as an archival repository of USDOT-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by USDOT or funded partners.
As a repository, ROSA P retains documents in their original published format to ensure public access to scientific information.
i
Quality Management of Cracking Distress Survey in Flexible Pavements Using LTRC Digital Highway Data Vehicle
-
2022-03-01
[PDF-3.98 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report, April 2016 – September 2021
-
Corporate Publisher:
-
Abstract:The Louisiana Department of Transportation and Development (DOTD) began to collect its Pavement Management System (LA-PMS)’s pavement condition data using a vendor’s 3D automatic system in 2017. For each 0.1-mile subsection on a flexible pavement, the vendor’s 3D automated cracking data reported in LA-PMS consists of various cracking amounts in terms of the alligator, longitudinal, and transverse cracks at different severity levels (e.g., low, medium, and high). The objectives of this study include two-folds: (1) to evaluate and assess the accuracy and precision of the 3D automated cracking data on flexible pavements through comparison with manual measurements on high-resolution pavement images; (2) to develop an image analysis application in pavement cracking identification on high-resolution pavement images collected by the Louisiana Transportation Research Center (LTRC)’s high-speed data vehicle. To achieve these objectives, a comprehensive manual cracking survey based on the DOTD’s distress identification protocol was conducted on twenty-three flexible pavement sections (totally 28.6 miles long) and nine 0.5-mile calibration sites, using the vendor’s high-resolution pavement images collected during the 2017 DOTD’s pavement condition data collection cycle. By directly — 2 — comparing the manual measurements and the 2017 reported cracking data in LA-PMS, results indicated that the automated cracking measurements based on 0.1-mile long sections tend to over-estimate the medium severity level of cracking amounts for all flexible pavement crack types (alligator, longitudinal, and transverse). However, when the automated cracking measurements were re-grouped based on 50-ft. long subsections, the overall estimation errors without differentiating the cracking severity levels could be significantly reduced due to a smaller standard deviation of the measurement errors and a shorter section length. Based on 50-ft. subsections, false positive errors produced by the automated system were found to be 8.5%, 9.8%, and 8.8% for alligator, longitudinal, and transverse cracking, respectively and the corresponding missed crack errors were 5.0%, 7.9%, and 1.4% respectively in this study. Statistical tests based on the mean measurement errors and equality of variance were conducted to qualitatively evaluate the accuracy and precision of the collected cracking data. In general, based on 50-ft. subsections, the automated system could produce significantly accurate results for high severity transverse cracking and significantly precise results for low severity alligator cracking. On the other hand, based on 0.1-mile subsections, the automated system was found not able to produce significantly accurate estimation of pavement cracking at different severity levels but provide significantly precise cracking measurements at low severity levels for all crack types. Since the overall results indicated that the 3D automated cracking measurements were statistically different from the manual measurements, which possibly led to a smaller cracking index estimation and different treatment selection, an artificial neural network (ANN) model was developed using the cracking measurements in this study. The ANN model aimed at adjusting the automated cracking measurements towards the manual cracking measurements at a 0.1-mile interval used in the LA-PMS, specifically for the flexible pavement cracking measurement data found in 2017 LA-PMS database. Finally, a MATLAB-based imaging analysis computer program was developed to generate an automated cracking report from high-resolution 2D pavement images collected by LTRC.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
