Home

I am currently a Director at Maxar Technologies (formerly DigitalGlobe) in their Product R&D group, which does research toward next generation products in high-resolution satellite imaging. I have worked on a wide variety of projects including image analysis, machine learning, data science, large-scale inverse problems in bathymetry and atmospheric compensation, and parallel computations for large geodetic networks.

Research Interests

- Satellite image analysis and spectral science
- Tensor factorizations applied to data mining
- Optimization algorithms
- Numerical linear algebra
- C/C++/Matlab code optimization
- Applications in imaging science, text analysis, machine learning, and modeling & simulation

Selected Papers

- Tensor Decompositions and Applications (with T.G. Kolda), SIREV, 51(3):455-500, Sep. 2009
- Algorithm 862: MATLAB Tensor Classes for Fast Algorithm Prototyping (with T.G. Kolda), ACM TOMS, 32(4):635-653, Dec. 2006
- See complete list in Bibliography

Selected Presentations

- Bathymetry using WorldView-2 imagery (SIAM Annual Meeting, 2013)
- Algebraic Techniques for Multilingual Document Clustering, presented at Google, January 25, 2011 (Google Tech Talk posted on YouTube)

Software

- MATLAB Tensor Toolbox - Higher-order operations and algorithms for multidimensional arrays
- Trilinos - A collection of packages for high performance computations
- NOX - A C++ nonlinear solver package (in Trilinos)

Patents

- Automated concrete/asphalt detection based on sensor time delay, US Patent App.
- Automated process for dynamic material classification in remotely sensed imagery, US Patent App.
- Cloud detection from satellite imagery, 2021
- Automated metric information network, 2016 & 2018
- Bathymetric techniques using satellite imagery, 2016 & 2017
- Technique for information retrieval using enhanced latent semantic analysis generating rank approximation matrix by factorizing the weighted morpheme-by-document matrix, 2012