Laboratory Mission Statement:

"To go boldly where no reductionists have gone before, bringing the distilled wisdom of the physicist to bear on problems far too complex for pure reductionists."

Faculty and Postdocs:
Stephen J. Daunt, J. Stewart Hager

Graduate Students:
James Wicker, Ph.D. Candidate
John Meyer, M.S. Candidate
Forrest Hoffman, M.S. Candidate

Opportunities for Graduate Students

The research areas listed below cover many interest areas in the laboratory. At present we have no GRA support but would be pleased to discuss your interests with you. Not all the projects below are presently active but could be encouraged to come alive for the right student Hyperspectral imaging studies in data processing and interpretation of hyperspectral data in the visible and infrared using cameras created by David Glenar, NASA/GSFC and John Hillman, NASA/GSFC (retired) and University of Maryland, Astronomy Department. These projects are open-ended and in progress or planned to commence in the near future. The projects include imaging the Star Spangled Banner in the National Museum of American History (Smithsonian), imaging of paintings for studies of origin and authentication, analysis of the SL-9 comet crash into Jupiter, and others.

Studies of hydrocarbon species found in the atmospheres of the outer planets

Besides, the energy level studies of gas phase molecules described above, the absolute intensity (cross section) of molecular transitions of many light hydrocarbon species found in planetary atmospheres is studied using tunable laser spectroscopy (in-house) coupled with spectra taken at NOAO's Kitt Peak National Observatory and the National Solar Observatory using the NSO's McMath Telescope Fourier Transform Spectroneter. These studies are in close collaboration with scientists at NASA/GSFC's Laboratory for Extraterrestrial Physics.

Resolution enhancement in one dimensional systems

This field involves recovering instrumental resolution in high resolution spectra with computational techniques. The recoveries are based sometimes on phenomenological models and in others on rigorous physical modeling of the instrumental processes resulting in band-limiting of the recorded data.

Resonances in molecular vibration-rotation systems

Analysis and interpretation of very high resolution spectra of gas phase molecular spectra are crucial in the study and understanding of atmospheres, both the earth's and the outer planets. Resonances-- both accidental and essential-- complicate the analysis and interpretation of such spectra and are a principal area of research. Presently, target systems are ethane, ethylene and methyl cyanide and the associated spectra in the region of the 10 micrometer atmospheric window.

Image enhancement for Hubble Space Telescope class problems

Using neural computational systems, image enhancement for HST class problems typified by the degraded resolution of the original Wide Field Planetary Camera on the Hubble spacecraft has been successfully carried out. The field involves the discovery of a novel solution to the inverse point spread function problem. A practicable numerical solution of Fredholm integral equation of the first kind has been discovered. This discovery has important applications in the deconvolution or deblurring of data acquired using scientific instruments as well as other areas where the integral equations of the first kind form an analytical base model for the instrument or related apparatus. The solution devolves from operations performed on the instrumental response function(also known as the kernel or point spread function) with the direct result of producing an effective inverse depends upon the observational data. It is therefore proposed that the prevailing concept of a scientific instrument be refined to include the practical implementation of this newly discovered effective inverse response function as a final stage of a complex scientific observing instrument.

Positron Emission Tomography (PET Imaging)

An outgrowth of the high function image enhancement program and the neural network program has been the development of a collaboration with the Paul Scherrer Institute in Villigen, Switzerland involving state of the art enhancements of PET image reconstruction from the observed sinograms. Both neural network approaches, and CBM image processing methods are being used to improve the reconstruction of PET images of biological targets.

Application of Neural Computational Paradigms to real-world physics problems

Making use of the power of neural networks and the emergence of the image paradigm in physics modeling, studies focused on real-world physics problems are pursued. None of the problems are easy or readily solved, but significant progress has been made. Typical problems include the recovery of true images from speckle interferograms and resolution enhancement of phase contrast electron micrographs beyond the highest existing resolutions in the world.

Complex Systems: Computational and Experimental Studies Including the Experimental Control of Chaos

In collaboration with John Hillman and Gordon Chin of the Laboratory for Extraterrestrial Physics at Goddard Space Flight Center, the most recent complex systems study involves the control of chaotic frequency emission of lead salt stripe geometry tunable infrared lasers. Chaotic frequency emission has been demonstrated and controlled, and present efforts involve development of a useful, enhanced resolution spectroscopic instrument using the chaos controlled laser diodes.