Over the past decade, several systems have been developed to collect digital imagery for evaluating the condition of paved surfaces. The techlogy has w expanded to include other road components. A pilot program recently completed in the San Francisco Bay Area demonstrates that off-the-shelf, GPS-enabled cameras and cellphones can be used to collect digital imagery that is adequate for identifying, evaluating, and creating georeferenced datasets for a wide variety of road assets. The purpose of this program was to evaluate multiple road asset types, including pavement, delineations, guard rails, signage, and signals. With a single pass in each lane direction at rmal driving speeds, the program collected imagery using a hood-mounted, GPS-enabled digital camera. The collected data was then processed using proprietary artificial intelligence coding to identify and georeference the type, dimensions, condition, and various other attributes of each asset. The evaluated datasets were visually field checked to validate the results from the AI processing. Although different AI algorithms were needed for specific components, the process was cost-effective and provided consistent results. Because of the demonstrated GPS-enabled aspect of the process and results, the evaluations will be repeatable and useable by public agencies and providing a very promising advancement in the public works community. After this session, participants will be able to:
• Develop an increased ability to identify, compile, and evaluate a variety of assets using a single-pass drive-by method.
• Implement a long-range strategic plan to inventory and assess the performance of the more than 2.8 million lane miles of paved roads, lane delineation, guard rails, fencing, signage, traffic signals, and other road system components.
• Optimize the planning, designing, and programming of most all aspects of public road infrastructure.

Contributor/Source

Justin Lindeman

Claim CEUs

Lorem ipsum dolor sit amet consectetur adipiscing elit, mattis condimentum justo velit convallis taciti faucibus, egestas elementum vitae vestibulum cum fames