Automating Bridge Inspection Procedures: Real-Time UAS-Based Detection and Tracking of Concrete Bridge Element
Abstract
Bridge inspections are necessary to maintain the safety, health, and welfare of the
public. All bridges in the United States are federally mandated to undergo routine
evaluations to confirm their structural integrity throughout their lifetime. The traditional
process implements a bridge inspection team to conduct the inspection, heavily relying
on visual measurements and subjective estimates of the existing state of the structure.
Conducting unmanned automated bridge inspections would allow for a more efficient,
accurate, and safer alternative to traditional bridge inspection procedures. Optimizing
bridge inspections in this manner would enable frequent inspections in order to
comprehensively monitor the health of bridges and quickly recognize minor problems
which could be easily corrected before turning into more critical issues. In order to
create an unmanned data acquisition procedure, unmanned aerial vehicles with high-resolution
cameras will be employed to collect videos of the bridge under inspection. To
automate a bridge inspection procedure employing machine learning methods, such as
neural networks, and machine vision methods, such as Hough transform and Canny edge
detection, will assist in identifying the entire beam. These methods along with future
work in damage detection and assessment will be the main steps to create an unmanned
automated bridge inspection.
Subject
Machine LearningComputer Vision
Region-based convolutional neural networks
automated bridge inspections
concrete beam element
unmanned aerial system
unmanned aerial vehicle
Citation
Miller, Emily Elizabeth (2018). Automating Bridge Inspection Procedures: Real-Time UAS-Based Detection and Tracking of Concrete Bridge Element. Master's thesis, Texas A & M University. Available electronically from https : / /hdl .handle .net /1969 .1 /174517.