United States Department of Transportation
i
Automation of Pavement Surface Distress Survey Through Parallel Processing
-
2007-01-16
[PDF-1.89 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
TRIS Online Accession Number:01042402
-
Edition:Final Report
-
Corporate Publisher:
-
Abstract:After a number of years of research and development on highway information systems, the research team at the Department of Civil Engineering, University of Arkansas, has made substantial progress and produced a full-digital Next-Generation Highway Data Vehicle (NGV) and related technology called the Multimedia-based Highway Information System (MMHIS). NGV and MMHIS can be used for a number of highway infrastructure management needs, including collecting high-resolution digital images of pavement surface for distress survey. However, the analysis of pavement surface distress, or cracks, is still a manual process that requires skilled technicians to visually examine pavement surface and identify and classify distress. This process is labor-intensive, prone to errors, and very slow. As a large portion of the National Highway System (NHS) in the U.S. enters mid-age or late age before major rehabilitation, there is an urgent need to find an automated approach for network level distress survey. For several decades, numerous attempts were made at automating this process. Unfortunately, none of the attempts was quite successful, even though a small number of highway agencies have been using semi-automated methods for distress survey. This proposal presents a Phase I study to implement an automated design based upon the framework and infrastructure of the newly developed NGV vehicle and MMHIS. It is believed that a low-cost parallel processing approach can be efficiently implemented to speed up the image processing algorithms.
-
Format:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
