Chair: Charles A. Bouman, Purdue University, USA
Wen Gao, Purdue University (U.S.A.)
Yibin Zheng, Purdue University (U.S.A.)
Peter C. Doerschuk, Purdue University (U.S.A.)
We describe measurements, models, and algorithms for several signal reconstruction problems arising in the structural biophysics of so-called spherical viruses.
Saad J. Saleh, Shell E&P Technology Company (U.S.A.)
Gary L. Syverson, Shell Continental Companies (U.S.A.)
In this paper, we give a brief overview of the problem of modeling the near-surface rock layers, which are often responsible for causing significant distortions of the travel times of seismic waves used for oil and gas exploration. The modeling scheme is a nonlinear inverse problem. In addition to the theoretical overview, we show practical results from a west Texas seismic survey obtained using the popular generalized linear inversion method for solving the modeling problem.
Timothy J. Schulz, Michigan Technological University (U.S.A.)
In this paper a nonlinear model is presented for the problem of space-object imaging through atmospheric turbulence, and a nonlinear method is discussed for forming fine-resolution images from blurred telescope data. Results from real telescope data are also presented.
Michael J. Gerry, The Ohio State University (U.S.A.)
Lee C. Potter, The Ohio State University (U.S.A.)
Randolph L. Moses, The Ohio State University (U.S.A.)
Michael A. Koets, The Ohio State University (U.S.A.)
We present a parametric model to describe radar scattering of man-made objects from synthetic aperture radar (SAR) measurements. The model is developed for high frequency scattering of objects in the frequency-angle domain, and transformed into the image domain for parameter estimation. The image-domain model is applied to SAR image segments to extract a geometrically relevant parametric description of dominant scattering behavior. The estimated parameters provide a concise description of the measured scattering, and has applications in object recognition and data compression.
Kenneth M. Hanson, Los Alamos National Laboratory (U.S.A.)
Gregory S. Cunningham, Los Alamos National Laboratory (U.S.A.)
Suhail S. Saquib, Los Alamos National Laboratory (U.S.A.)
Experimental data often can only be interpreted by means of a computational simulation that approximately models the physical situation. We will discuss techniques that facilitate application to complex, large-scale simulations of the standard approach to inversion in which gradient-based optimization is used to find the parameters that best match the data. The fundamental enabling techniques are adjoint differentiation to efficiently compute the gradient of an objective function with respect to all the variables of a simulation and relatively new gradient-based optimization algorithms. These techniques will be illustrated through the simulation of the time-dependent diffusion of infrared light through tissue, which has been used to perform optical tomography . The techniques discussed have a wide range of applicability to modeling including the optimization of models to achieve a desired design goal.
Herve Carfantan, CNRS/Supélec/UPS (France)
Ali Mohammad-Djafari, CNRS/Supélec/UPS (France)
The Bayesian approach has been proven to give a common estimation structure to existing image reconstruction and restoration methods 11]. The goal of this paper is to investigate diffraction tomography in this framework. A regularized solution to this nonlinear inverse problem is defined as the maximum a posteriori estimate, introducing prior information on the object to be reconstructed. Two equivalent formulations of this definition axe proposed which lead to solution of a constrained or an unconstrained optimization problem. From this point of view, we propose a classification of existing methods for solving this problem and new orientations to compute the defined solution.
Alfred O. Hero III, University of Michigan (U.S.A.)
Robinson Piramuthum, University of Michigan (U.S.A.)
In  a methodology for incorporating extracted MRI anatomical boundary information into penalized likelihood (PL) ECT image reconstructions and tracer uptake estimation was proposed. This methodology used quadratic penalty based on Gibbs weights which enforced smoothness constraints everywhere in the image except across the MRI-extracted boundary of the ROI. When high quality estimates of the anatomical boundary are available and MRI and ECT images are perfectly registered, the performance of this method was shown to be very close to that attainable using ideal side information, i.e. noiseless anatomical boundary estimates. However when the variance of the MRI-extracted boundary estimates becomes significant this penalty function method performs poorly. We give a modified Gibbs penalty function implemented with a set of averaged Gibbs weights, where the averaging is performed with respect to a limiting form of the posterior distribution of the MRI boundary parameters.