VIVA: Automated Analysis of InSAR images

---

Project Info

Purpose

Early detection of potential transportation hazards to improve safety and reduce maintenance costs.

Discussion

This project, sponsored by the U.S. Department of Transportation Research and Innovative Technology Administration, investigates the applicability of new commercial remote sensing technology to three transportation-related problems. First, the project addresses the high-risk, high-reward problem of early detection of sinkholes. Then, the same satellite based technology will be used to monitor landslides and bridge settlement. The proposed solutions are fueled by the combination of new millimeter-radar (InSAR) and novel image analysis algorithms developed at the University of Virginia (UVA). A partnership consisting of UVA, the Virginia Transportation Research Council (VTRC), and TRE Canada (the supplier of the data), has been forged to tackle these important transportation application.

Fig.1 Sinkhole development

In the project, the feasibility of using the remotely sensed data to detect and monitor sinkholes, settling bridges and landslides will be assessed. Automated methods for analyzing the images will be developed and tested in a selected region of Virginia within the I-81 Interstate corridor. The end product of the research will be a suite of software tools that can be used by the state departments of transportation across the U.S. to automatically detect and monitor potential sinkholes, bridge settlement and landslide movement from satellite imagery. It is anticipated that the automated processing tools, in combination with the newly available commercial remote sensing data, will lead to multi-million dollar cost savings in the highway repairs, significant reduction in highway closures and enhanced safety of the traveling public.

Fig.2 Bridge settlement detection

Collaborators

• Virginia Center for Transportation Innovation & Research (VCTIR)
• Tele-Rilevamento Europa (TRE) - Canada

Publications

• Graph Cut Segmentation of Sparsely Sampled Images with Application to InSAR-measured Changes in Elevation

  Abstract - In this paper, we outline an algorithm for the automatic segmentation of sparse data in order to detect possible terrain-deformation phenomena. Segmentation is accomplished through a graph cut technique. In the graph structure, for each edge, we derive a unique energy by combining multiple independent energies tailored toward accurately locating the boundaries of spot-like, subsiding regions. We then find the series of cuts with minimum total energy and fit splines to these cuts for smooth segment boundaries. The segmentation approach is applied to the problem of localizing sinkholes in karst regions. Test results indicate efficacy for a sufficient density of InSAR features.

  Stuecheli, M.; Vaccari, A.; Acton, S.T.; , "Graph cut segmentation of sparsely sampled images with application to InSAR-measured changes in elevation," Image Analysis and Interpretation (SSIAI), 2012 IEEE Southwest Symposium on , vol., no., pp.149-152, 22-24 April 2012. doi: 10.1109/SSIAI.2012.6202475

Workshops

• Transportation Research Board - 92nd Annual Meeting - January 13-17, 2013

  Workshop: 135 - Sensing Technologies for Transportation Applications

  Acton, S.T.; "Model for Sinkhole Detection by InSAR" (P13-6215) [.pdf 2MB]

• Transportation Research Board - 91st Annual Meeting - January 22-26, 2012

  Workshop: 184 - Sensing Technologies for Transportation Applications

  Acton, S.T.; "Investigating Interferometric Synthetic Aperture Radar for Transporation Infrastructure Monitoring" (P12-6176) [.pdf 2MB]

Links

• Surface Reconstruction [.zip]

  Matlab implementation of mosaicked, thin-plate smoothing spline surface reconstruction. Example included.

• GIS Export [.zip]

  Matlab routines which export polygon risk information into Google Earth and ArcGIS. ArcGIS export requires the mapping toolbox.