Research Interests

Computational Imaging, Inverse Problems, Statistical Signal Processing, Machine Learning

Education

Ph.D. - Electrical and Computer Engineering, Purdue University (Advisor: Charles A. Bouman)

Master of Science - Electrical and Computer Engineering, Purdue University

Bachelor of Technology - Electronics and Communication Engineering, NIT Trichy

Google Scholar

ORNL Researcher Profile

GitHub (Link1, Link2)

Research Summary

At a high level, my current research aims to aid the development of novel imaging systems (microscopes, cameras, CT scanners, ultrasound systems etc.) by the design of algorithms. The ability to leverage powerful computing platforms using algorithms is greatly expanding the capability of existing imaging systems and will facilitate the design of the next generation of imagers.

My research is focused on the development of algorithms for Computational Imaging systems. In particular, my work has used model-based image reconstruction (MBIR) approaches that combines statistical models for the data and low-dimensional signal models for the underlying quantity to be measured to get the most out of the measurement. The algorithms draw on ideas from diverse fields such as statistical signal processing, image processing, inverse problems and machine learning.

Working closely with inter-disciplinary teams of experimental scientists and image processors, we have demonstrated how these approaches can dramatically improve the imaging capability of electron, X-ray and neutron microscopes without any additional hardware investments to these expensive instruments. For more information on model-based imaging and the specific applications, please see the respective pages.