Demetri Psaltis

TL;DR: Optical image correlators are presented that use acousto-optic and charge-coupled devices as the input and output transducers respectively and the applicability to pattern recognition of a non-linear pseudo-correlation that can also be conveniently computed with these processors is discussed.

TL;DR: This report presents a system that uses holograms to extract spatial and color information (4-D imaging) of a specimen and project it on to a 2-D space on the detector and makes the case that a direct interface between an optical memory and a chip integrating detectors and logic circuitry can better utilize the high parallelism inherent in holographic modules.

TL;DR: This approach enables arbitrarily programmable simultaneous unicast and multicast connectivity, eliminating the need for optical splitters that hinder scalability due to optical power loss, and uses phase modulation in multiple layers, tailored to implement any multicast connectivity map.

TL;DR: In this paper, an experimental comparison of optical tomography techniques: Learning Tomography (LT), an iterative method based on beam propagation model and diffraction tomography (DT) based Born and Rytov approximation is presented.