A Novel Vision Sensor for Remote Measurement of Bridge Displacement

This project will develop a camera-based vision sensor for accurate remote, multi-point measurement of bridge displacements enabled by a robust target tracking algorithm, an advanced image distortion filter, and a vibration cancellation technique. A prototype system will be built and its efficacy demonstrated in bridge field tests. The project is being carried out in two stages. Work in Stage 1 focuses on developing three novel algorithms and techniques to improve the accuracy of the remote measurement of multi-point bridge displacements in outdoor environments without artificial tracking markers. (1) Robust template matching algorithm based on orientation code matching (OCM): The novel OCM will be further developed for multi-point displacement measurement in outdoor field environments. The goal is to track the existing “natural” markers on the bridge surface and completely eliminate the requirement of installing artificial target panels at the measurement points. Various commonly-available structural surface textures and objects will be tested in the laboratory. (2) Digital heat haze filter technique: A novel digital filter will be developed to eliminate random blurring and distortion of the video images caused by heat haze, which is critical for accurate and stable displacement measurements from a long distance. Alternatively, a statistical correlation analysis technique will also be investigated. The performance of the digital heat haze filter and statistical approach will be evaluated through laboratory tests using simulated heat haze, by comparing the displacement measurement results with and without the heat haze and with and without the digital filtering technique. (3) Camera vibration cancellation technique: The multi-point measurement-based technique will be developed and evaluated in terms of its ability to cancel the measurement errors caused by camera vibration (due to wind, traffic, and ambient vibration). Shake-table tests will be carried out at the Columbia University’s Carlton Laboratory, using a sample structure and two shaking tables. Finally, the OCM algorithm will be integrated with the heat haze filtering and camera vibration cancellation to systematically address all sources of environmental noise affecting measurement accuracy of vision-based sensors. Laboratory tests will be carried out to evaluate the limitation of the measurement points under simulated environmental noise such as shadowing, heat haze, change in illuminating light, and background conditions. Work in the final stage will focus on the development and bridge field evaluation tests of a prototype vision sensor system. The developed algorithms/techniques will be integrated into a software package. Advanced computational techniques will be applied to reduce the image processing time to realize real-time online measurement. The performance of the prototype vision sensor system will be further evaluated on two bridges (the short-span Jamboree Overcrossing concrete bridge in California and the long-span Manhattan Bridge in New York). Technology transfer efforts will be focused on engaging end users, such as NYC Department of Transportation (DOT) and California Department of Transportation (Caltrans), in the bridge field tests to build their awareness of capabilities of the technology. The final report will provide all data and findings along with guidelines on implementing the developed technique for measuring bridge displacement by state DOTs.

Language

  • English

Project

  • Status: Active
  • Funding: $135810
  • Contract Numbers:

    Project 20-30, IDEA 189

  • Sponsor Organizations:

    Safety Innovations Deserving Exploratory Analysis (IDEA)

    Transportation Research Board
    500 Fifth Street, NW
    Washington, DC  United States  20001

    National Cooperative Highway Research Program

    Transportation Research Board
    500 Fifth Street, NW
    Washington, DC  United States  20001

    American Association of State Highway and Transportation Officials (AASHTO)

    444 North Capitol Street, NW
    Washington, DC  United States  20001

    Federal Highway Administration

    1200 New Jersey Avenue, SE
    Washington, DC  United States  20590
  • Project Managers:

    Jawed, Inam

  • Performing Organizations:

    Columbia University

    610 SW Mudd
    500W 120th Street
    New York, New York  United States  10027
  • Principal Investigators:

    Feng, Maria

  • Start Date: 20160218
  • Expected Completion Date: 0
  • Actual Completion Date: 0
  • Source Data: RiP Project 40612

Subject/Index Terms

Filing Info

  • Accession Number: 01590440
  • Record Type: Research project
  • Source Agency: Transportation Research Board
  • Contract Numbers: Project 20-30, IDEA 189
  • Files: TRB, RiP
  • Created Date: Feb 18 2016 1:00AM