Dr. Thomas Oommen

Assistant Professor, Geological and Mining Engineering and Sciences

Thomas Oommen Dr. Oommen’s research is focused on the application of remote sensing for geohazard characterization. His research group uses machine learning and pattern recognition algorithms to efficiently classify or cluster remote sensing data to characterize hazards. One area of his research evaluates the use of crowd-sourcing to analyze earthquake-induced damages in remotely sensed imagery. The advent of Web 2.0 technologies and the ubiquity of free remote-sensing images possess three components of high resolution -- spatial (measure of the smallest object that can be identified), spectral (the specific wavelengths that a sensor can record), and temporal (how often a sensor can obtain imagery) -- have made crowd-sourced damage assessment possible. The benefit has already been demonstrated -- delivery of damage estimates two to three times faster than previously. However, detailed ground assessment also highlighted the limitation of this approach and the variability in accuracy of damage assessment with varying expertise of the crowd volunteers. To overcome this limitation, Dr. Oommen proposes automating key tasks using state-of-the-art machine learning and image processing and shifting the role of the volunteer crowd from manual image interpretation to supervised guidance of the automated tasks for mapping and classifying damage. This study is conducted in collaboration with Dr. Umaa. D. Rebbapragada, a Machine Learning Scientist at the Jet Propulsion Laboratory, California Institute of Technology.

Another area of research that Dr. Oommen’s group is evaluating is, the use of a variety of remote-sensing techniques, from Optical Photogrammetry (OP) to Light Detection and Ranging (LiDAR) to Interferometric Synthetic Aperture Radar (InSAR) to manage geotechnical assets along the transportation corridor. The one common theme for US transportation infrastructure is that it is supported by the ground, constructed over it, through it or next to it. Scientists call these natural bases geotechnical assets. The problem, however, is that we do not have a comprehensive program in the US to monitor the geotechnical assets of our transportation corridors.

Dr. Oommen says that it is highly likely that cameras mounted on unpiloted aerial vehicles (UAVs), cars and trains will monitor entire transportation networks automatically, sending remotely sensed imagery back for analysis of potential instability. Co-investigators on this research include Colin Brooks, senior research scientist at the Michigan Tech Research Institute (MTRI); Pasi Lautala, assistant professor in civil and environmental engineering; Stan Vitton, associate professor in civil and environmental engineering; and Keith Cunningham, a research professor from the University of Alaska Fairbanks.

For more information, please visit Dr. Oommen' website.

Recent publications

01 A Study of the Impacts of Freeze-Thaw on Cliff Recession at the Calvert Cliffs in Calvert County, Maryland
B. Zwissler, T. Oommen, S. Vitton
Geotechnical and Geological Engineering, vol. 32, p. 1133 (2014)
02 Hazard Assessment of Rainfall-Induced Landslides: A Case Study of San Vicente Volcano in Central El Salvador
D. M. Smith, T. Oommen, L. J. Bowman, J. S. Gierke, S. J. Vitton
Natural Hazards (2014)
03 Numerical Modeling of Volcanic Slope Instability and Related Hazards at Pacaya Volcano, Guatemala
L. N. Schaefer, T. Oommen, C. Corazzato, A. Tibaldi, R. Escobar-Wolf, W. I. Rose
Bulletin of Volcanology, vol. 75, p. 720 (2013)
04 Documenting Earthquake-Induced Liquefaction using Satellite Remote Sensing Image Transformations
T. Oommen, L. G. Baise, R. Gens, A. Prakash, R. P. Gupta
Environmental & Engineering Geoscience, vol. 19, p. 303 (2013)
05 An Objective Analysis of Support Vector Machine Based Classification for Remote Sensing
T. Oommen, D. Misra, N. K. C. Twarakavi, A. Prakash, B. Sahoo, S. Bandopadhyay
Mathematical Geoscience, vol. 40, p. 409 (2008)