|Stochastic Systems Group|
In this talk, we present some of our recent work in coherent image reconstruction. Some applications that use coherent imaging include synthetic aperture radar, holography, and medical ultrasound. One of the motivations for our work has been the increased interest in using reconstructed images in automated decision-making tasks. The success of such tasks (e.g. target recognition) depends on how well the computed images exhibit certain features of the underlying scene. Current coherent image formation techniques have no explicit means to enhance features (e.g. scatterer locations, object boundaries) that may be useful for automatic interpretation. We have developed a mathematical foundation and the associated algorithms for feature-enhanced coherent imaging to address this challenge. Our framework is based on a regularized reconstruction of the scattering field, which combines an explicit mathematical model of the data collection process with non-quadratic functionals representing prior information about the nature of the features of interest. We solve the challenging optimization problems posed in our framework by extending half-quadratic regularization methods to the complex-valued coherent imaging problem. Our technique effectively deals with the complex-valued, and potentially random-phase nature of the underlying reflectivities, which is inherent in many coherent systems. We demonstrate the performance of this method on a number of applications and data collection scenarios. Compared to conventional techniques, the method we propose produces images with increased resolution, reduced sidelobes, reduced coherent speckle artifacts, and easier-to-segment regions.
Problems with this site should be emailed to email@example.com