|
Title: Exploiting Imagery Data Collected with Unmanned Aircraft Systems (UAS) for Bridge Inspections
Accession Number: 01658365
Record Type: Component
Abstract: The usage of Unmanned Aerial Systems (UAS) is increasing at a staggering speed in the U.S. and the world. In 2016, the American Association of State Highway and Transportation Officials found that thirty-three state Department of Transportations have or are employing or researching the use of drones for bridge inspections, surveying, clearing of accidents and others. However, the increased use of UAS systems implies an increased amount of data collected which needs to be processed and stored. Recently developed online platforms and software are available for location-based image inspections, fourth-dimensional modeling (where the fourth dimension is time) and geometric analysis using digital state models, change detection and automated damage detection using Artificial Intelligence (A.I.). These technologies accelerate the human analysis of the images by providing location-aware image selection, highlighting potential defects, and enriching two-dimensional image data into a multi-dimensional world. Researchers have found methods to build digital replicas from images with an accuracy of ±0.09 in. for 3D features measured (95% confidence). Current developments in artificial intelligence will soon function as an inspection-assistant to support defect and change detections, increase safety and provide predictive analytics. Big data combined with A.I. or pattern recognition technologies will provide a framework to develop new methods to uncover currently unknown factors that affect the overall health of our civil infrastructures. Technologies, as of 2016, were able to detect up to 70 percent of all cracks in cluttered images. However, a sharp increase in accuracy is expected with recent developments in deep learning (i.e., A.I. inspired by the structure and function of the brain).
Supplemental Notes: This paper was sponsored by TRB committee ABJ50 Standing Committee on Information Systems and Technology.
Report/Paper Numbers: 18-03134
Language: English
Authors: Salomon, Abraham LamaWells, JenniferPagination: 18p
Publication Date: 2018
Conference:
Transportation Research Board 97th Annual Meeting
Location:
Washington DC, United States Media Type: Digital/other
Features: Figures; Photos; References
TRT Terms: Subject Areas: Aviation; Bridges and other structures; Data and Information Technology; Highways; Maintenance and Preservation; Vehicles and Equipment
Source Data: Transportation Research Board Annual Meeting 2018 Paper #18-03134
Files: TRIS, TRB, ATRI
Created Date: Jan 8 2018 10:45AM
|