United States Department of Transportation
i
Advanced Road Scene Image Segmentation and Pavement Evaluation Using Neural Networks
-
2010-01-01
-
[PDF-1.02 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
TRIS Online Accession Number:01450866
-
Edition:Final report.
-
Corporate Publisher:
-
NTL Classification:NTL-HIGHWAY/ROAD TRANSPORTATION-Pavement Management and Performance;NTL-HIGHWAY/ROAD TRANSPORTATION-Construction and Maintenance;
-
Abstract:The current project, funded by MIOH-UTC for the period 9/1/2009-8/31/2010, continues our efforts in designing an image processing based pavement inspection system for the assessment of highway surface conditions. One of the most important tasks in pavement maintenance is pavement surface condition evaluation distress measurement. In order to eliminate the tedious and unreliable manual inspection of pavement surface evaluation, image processing and pattern recognition techniques are used to increase the efficiency and accuracy and decrease the costs of pavement distress measurements. Existing systems for automated pavement defect detection commonly require special devices such as lights, lasers, etc, which dramatically increase the cost and limit the system to certain applications. Therefore, in this report, a low cost automatic pavement distress evaluation approach is presented. This method can provide real-time pavement distress detection as well as evaluation results based on color images captured from a camera installed on a survey vehicle. The entire process consists of two main parts: pavement surface extraction followed by pavement distress detection and classification. In the first part, a novel color segmentation method based on a feed forward neural network is applied to separate the road surface from the background. In the second part, a segmentation technique based on probabilistic relaxation is utilized to separate distress areas from the road surface. The geometrical parameters obtained from the detected distresses are then fed to a neural network based pavement distress classifier in which the defects are classified into different types. Simulation results are given to show that the scheme presented in this report is both effective and reliable on a variety of pavement images.
-
Format:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
