Open Access Open Access  Restricted Access Subscription or Fee Access

Evaluation of using Structure from Motion Optical Techniques for Structural Health Monitoring of Railroad Track



Routinely, civil, mechanical, and aerospace structures continue to be used despite that they are approaching or have already exceeded their design life. Structural monitoring techniques have the ability to indicate the onset of or avoid failure. Conventional sensing techniques like visual inspection and sensors such as accelerometers, strain gages, and acoustic transducers produce results at only a discrete number of points. Moreover, they can be labor intensive to install, have wire and power constraints, and the correlation between the sensor signal and structural integrity is not always straightforward. In the last few years, achievements made in camera technology, optical sensors, and image-processing algorithms allowed the development of a new generation of non-contact measuring methods. One of the most promising is Structure from Motion (SfM), a photogrammetric range imaging technique which makes it possible to obtain three-dimensional (3D) renderings from a cloud of two-dimensional images. The SfM approach has been widely used for large-area aerial inspections in the field of agriculture, geosciences, environmental disaster management, and heritage documentation. What is still missing is an evaluation of its feasibility for being used as a quantitative inspection technique for Structural Health Monitoring (SHM) purposes. In this study, an estimation of the accuracy of this non-contact, optically based measurement technique is experimentally performed. In particular, a number of laboratory experiments are executed on a real-size wood model of a railroad track to determine the capability of SfM in measuring crosstie vertical displacements, structural deformation, and geometry profiles. To validate the accuracy of the SfM measurements, a back-to-back comparison is made with Digital Image Correlation measurements. The strengths and limitations of the SfM approach for SHM are discussed and the measurement accuracy is quantified


Full Text:



  • There are currently no refbacks.