CIS Seminar - Erzsébet Merényi: Hyperspectral Eyeing of Heavenly Bodies – a Machine Intelligence Approach

Erzsébet Merényi is a Research Professor in the Department of Electrical and Computer Engineering
Rice University, Houston, Texas.

Erzsébet Merényi (http://www.ece.rice.edu/~erzsebet) earned an MS in mathematics
(1975) and PhD in computational science (1980) from Szeged (Attila József) University,
Hungary. She is a research professor in the Electrical and Computer Engineering
Department of Rice University, Houston, Texas. She was formerly a staff scientist at the
Lunar and Planetary Laboratory, University of Arizona, and a research associate in the
Cosmic Physics Department at the Central Research Institute for Physics of the Hungarian
Academy of Sciences. She worked on numerical modeling of charged particle transfer in the
Heliosphere (1980–1990), and on analyzing images of the nucleus of P/Halley from the
(then) Russian Vega mission. Her mathematical custom restorations of the severely
corrupted, once-in-a-lifetime images were published by the European Space Agency, and
are in international archives (PDS, IHW). 
From 1991 Erzsébet has been focusing on analyzing spectral data for resource
mapping and knowledge discovery from space missions and terrestrial remote sensing
projects, including data from Clementine, the Imager for Mars Pathfinder, the Mars
Exploration Rovers, telescopic measurements, and airborne hyperspectral sensors such as
AVIRIS. Most recently she has been collaborating on inference of latent parameters (such as
surface temperature and grain size) from high-resolution spectra, in preparation for the
Pluto-Charon encounter by the New Horizon mission. 
Her overarching research interest is in neural machine intelligence for manifold
learning, structure discovery and precise classification, in complex, high-dimensional, highly
structured data such as hyperspectral imagery and medical data. Her research has been
funded by several NASA programs, and by the Baylor College of Medicine. At Rice she also
co-leads the machine learning effort in a large DARPA – Rice compiler development effort.