Degree: PhD or MSc (by research)

Key-words: Structural monitoring, video cameras, deflection measurement

Entry: BSc(Hons), preferably 1st class, or MSc, in Engineering, Mathematics, Photogrammetry, Geodesy or Physics.

Supervisor(s): Dr Derek Lichti

Project Funding: Curtin University of Technology

Student Funding: Student required to win ASA, IPRS or other scholarship

Resources: Video cameras and asynchronous frame grabber, bundle adjustment and image sequence capture software, camera calibration facilities

Starting Date: Unrestricted

Knowledge of how structures such as bridges respond to dynamic loading conditions is critical information for civil engineers seeking to evaluate structural integrity and for model verification. Conventional monitoring methods use contact probes like linear-variable-differential transformers (LVDTs). Whilst very precise, LVDTs provide only one-dimensional information and are best suited to laboratory use. Additionally, LVDT accuracy degrades significantly outside its linear range, which restricts the magnitude of deflection that can be measured. Traditional surveying methods and laser scanning are immune to these problems but are too slow for dynamic event measurement. Photogrammetry offers both high spatial and temporal resolution (50 Hz or greater) through the use of video cameras.

The Curtin video monitoring-system has been successfully tested in lab conditions for static measurement of deflections in reinforced and pre-stressed concrete beams. Several beams were loaded to failure to test their shear and flexural strength. The next step in system application is capture of dynamic events. This project will be an investigation into the use of video cameras for deflection measurement of bridges in response to dynamic loading conditions.

The proposed project will involve research into one or more of the following:

1. Dynamic measurement. Software has been developed at Curtin to capture sequences of imagery from a pair of video cameras at a rate of 50 Hz (field rate). The software stores each field in the host PC’s RAM and writes the imagery to disk after the entire sequence has been captured. This aspect of the project will address issues such as target and network design requirements to meet the sub-millimetre accuracy required for structural monitoring.

 

2. Target measurement algorithms. Centroid-based algorithms have become the standard for close-range photogrammetric image point measurement. Under appropriate lighting conditions, circular, retro-reflective targets are imaged as bright ellipses covering several pixels in the imagery. Both the contrast and pixel coverage are exploited by centroid algorithms to obtain sub-pixel target measurements which, through collinearity, are transformed into three-dimensional object point co-ordinates. This aspect of the project will seek to automate centroiding algorithms and optimise bundle adjustment processing for off-line processing of image sequences of moving targets.

Fraser, C.S. and B. Riedel (2000) Monitoring the thermal deformation of steel beams via vision metrology. ISPRS Journal of Photogrammetry & Remote Sensing, 55 (4): 268-276.

Fraser, C.S. and M.R. Shortis (1995) Metric exploitation of still video imagery. Photogrammetric Record, 15 (85): 107-121.

Mills, J.P., I. Newton and G.C. Peirson (2001) Pavement deformation monitoring in a rolling load facility. Photogrammetric Record, 17(97): 7-24.

Olaszek, P (1999). Investigation of the dynamic characteristics of bridge structures using a computer vision method. Measurement, 25 (3): 227-236.

Shortis, M R and W L Snow (1997) Videometric tracking of wind tunnel aerospace models at NASA Langley research centre. Photogrammetric Record, 15 (89): 673-689.

Woodhouse, N.G., S. Robson and R. Eyre (1999) Vision metrology and three dimensional visualization in structural testing and monitoring. Photogrammetric Record, 16 (94): 625-641.