Monitoring Dynamic Structural Tests Using Image Deblurring Techniques

The Photogrammetric Record

2014

Unmanned aerial vehicles (UAVs) have become an interesting and active research topic for photogrammetry. Current research is based on images acquired by UAVs which have a high ground resolution and good spectral resolution due to low flight altitudes combined with a high-resolution camera. One of the main problems preventing full automation of data processing of UAV imagery is the unknown degradation effect of blur caused by camera movement during image acquisition. The purpose of this paper is to analyse the influence of blur on photogrammetric image processing. Images with precisely known motion blur were produced to determine the effect. It was found that even small blurs affect normal photogrammetric processes significantly. Although operator intervention might be time consuming, it can ensure that the results are still of acceptable accuracy.

“…Although blur might disturb most image processing procedures it can be also exploited for some applications, as identified by Boracchi () and McCarthy et al. ().…”

Section: Discussionmentioning

confidence: 99%

Motion blur disturbs – the influence of motion‐blurred images in photogrammetry

Sieberth

Wackrow

The Photogrammetric Record

2014

“…A one-dimensional shaker table was used to test and demonstrate the proposed approach. The APS "Electro-Seis" 400 is a long-stroke shaker capable of carrying out programmed one-dimensional motions up to 150 mm as described in McCarthy et al (2013McCarthy et al ( , 2014. In its normal use, models are fitted to the 455 mm 9 455 mm table bed which is programmed with vibrations of varying frequencies and amplitudes.…”

Section: Practical Tests and Resultsmentioning

confidence: 99%

“…This is explained and tested after the importance of structural testing and prior work is reviewed. The approach has built on previous developments (McCarthy et al., , ) and the results of enhanced case studies are included here.…”

Section: Introductionmentioning

confidence: 99%

Monitoring 3D Vibrations in Structures using High‐resolution Blurred Imagery

McCarthy

The Photogrammetric Record

Palmeri

2016

Self Cite

Photogrammetry has been used in the past to monitor the laboratory testing of civil engineering structures using multiple image‐based sensors. This has been successful, but detecting vibrations during dynamic structural tests has proved more challenging because they usually depend on high‐speed cameras which often results in lower image resolutions and reduced accuracy. To overcome this limitation, a novel approach has been devised to take measurements from blurred images in long‐exposure photographs. The motion of the structure is captured in individual motion‐blurred images without dependence on imaging speed. A bespoke algorithm then determines each measurement point's motion. Using photogrammetric techniques, a model structure's motion with respect to different excitation frequencies is captured and its vibration envelope recreated in 3D. The approach is tested and used to identify changes in the model's vibration response.

“…Instead of using high-speed imaging to capture the instantaneous position at each epoch, long-exposure images are used in which the localised image of the object becomes blurred (McCarthy, 2013). This paper describes tests where the approach has been further developed to demonstrating 3D.…”

Section: Methodsmentioning

confidence: 99%

3D Case Studies of Monitoring Dynamic Structural Tests using Long Exposure Imagery

McCarthy

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

Palmeri

2014

ABSTRACT:Structural health monitoring uses non-destructive testing programmes to detect long-term degradation phenomena in civil engineering structures. Structural testing may also be carried out to assess a structure's integrity following a potentially damaging event. Such investigations are increasingly carried out with vibration techniques, in which the structural response to artificial or natural excitations is recorded and analysed from a number of monitoring locations. Photogrammetry is of particular interest here since a very high number of monitoring locations can be measured using just a few images. To achieve the necessary imaging frequency to capture the vibration, it has been necessary to reduce the image resolution at the cost of spatial measurement accuracy. Even specialist sensors are limited by a compromise between sensor resolution and imaging frequency.To alleviate this compromise, a different approach has been developed and is described in this paper. Instead of using high-speed imaging to capture the instantaneous position at each epoch, long-exposure images are instead used, in which the localised image of the object becomes blurred. The approach has been extended to create 3D displacement vectors for each target point via multiple camera locations, which allows the simultaneous detection of transverse and torsional mode shapes. The proposed approach is frequency invariant allowing monitoring of higher modal frequencies irrespective of a sampling frequency. Since there is no requirement for imaging frequency, a higher image resolution is possible for the most accurate spatial measurement. The results of a small scale laboratory test using off-the-shelf consumer cameras are demonstrated. A larger experiment also demonstrates the scalability of the approach.