Optical high-performance computing: introduction to the JOSA A and Applied Optics feature.

The feature issues in both Applied Optics and the Journal of the Optical Society of America A focus on topics of immediate relevance to the community working in the area of optical high-performance computing.

Emmett Leith: a personal perspective.

"You can observe a lot by watching"American baseball hero and muddled sayings ?author Yogi Berra. Greatness observed is a privilege for the observer. I knew, admired, and appreciated Emmett Leith for many years and observed a lot about him. Here I propose to share some of what I have learned by watching Emmett. He had only one true peer--Yuri Denisyuk--who shared the adventure of inventing modern holography and was his good friend. Both were great and beloved men, but Emmett was funnier. I miss them both very much.

Interference fringe analysis based on centroid detection.

Phase measurement of interference fringes is an integral part of several fields in optics. Using simple straight sinusoidal fringe patterns, we describe the relationship between fringe position or phase to the centroid position when these fringes are incident on a position sensitive detector. With detailed descriptions and some experimental results, we show that a phenomenal sensitivity is possible in principle with what we believe is a new approach of phase measurement, and excellent sensitivity is readily achieved.

Discrimination among similar looking noisy color patches using Margin Setting.

The goal of discrimination of one color from many other similar-appearing colors even when the colored objects show substantial variation or noise is of obvious import. We show how to accomplish that using a technique called Margin Setting. It is possible not only to have very low error rates but also to have some control over the types of errors that do occur. Robust spectral filtering prior to spatial pattern recognition allows subsequent filtering processes to be based on conventional coherent optical correlation that can be done monochromatically.

Simple online recognition of optical data strings based on conservative optical logic.

Optical packet switching relies on the ability of a system to recognize header information on an optical signal. Unless the headers are very short with large Hamming distances, optical correlation fails and optical logic becomes attractive because it can handle long headers with Hamming distances as low as 1. Unfortunately, the only optical logic gates fast enough to keep up with current communication speeds involve semiconductor optical amplifiers and do not lend themselves to the incorporation of large numbers of elements for header recognition and would consume a lot of power as well. The ideal system would operate at any bandwidth with no power consumption. We describe how to design and build such a system by using passive optical logic. This too leads to practical problems that we discuss. We show theoretically various ways to use optical interferometric logic for reliable recognition of long data streams such as headers in optical communication. In addition, we demonstrate one particularly simple experimental approach using interferometric coinc gates.

Achieving stabilization in interferometric logic operations.

Interferometric systems with amplitude beam splitters can implement reversible operations that, on detection, become Boolean operators. Being passive, they consume no energy, do not limit the operating bandwidth, and have negligible latency. Unfortunately, conventional interferometric systems are notoriously sensitive to uncontrolled disturbances. Here the use of polarization in a common-path interferometric logic gate with and without polarization beam splitters is explored as an attractive alternative to overcome those difficulties. Two of three device configurations considered offer significant stability and lower drive modulator voltage as advantages over the previous systems. The first experimental tests of such a system are reported. Common-path interferometry lends itself to even more stability and robustness by compatibility with no-air-gap, solid optics.

Application of supergeneralized matched filters to target classification.

The matched filter (MF) is the optimum linear operator for distinguishing between a fixed signal and noise, given the noise statistics. A generalized matched filter (GMF) is a linear filter that can handle the more difficult problem of a multiple-example signal set, and it reduces to a MF when the signal set has only one member. A supergeneralized matched filter (SGMF) is a set of GMFs and a procedure to combine their results nonlinearly to handle the multisignal problem even better. Obviously the SGMF contains the GMF as a special case. An algorithm for training SGMFs is presented, and it is shown that the algorithm performs quite well even for extremely difficult classification problems.

Quantum-implementable selective reconstruction of high-resolution images.

This paper, written for interdisciplinary audience, presents computational image reconstruction implementable by quantum optics. The input-triggered selection of a high-resolution image among many stored ones, and its reconstruction if the input is occluded or noisy, has been successfully simulated. The original algorithm, based on the Hopfield associative neural net, was transformed in order to enable its quantum-wave implementation based on holography. The main limitations of the classical Hopfield net are much reduced with the simulated new quantum-optical implementation.

A neural bridge from syntactic to statistical pattern recognition.

A simple means of differentiating between syntactic and statistical pattern recognition is that, while both deal with feature sets, syntactic methods also include relationships among the features. In this paper we describe a pulse-coupled neural network (PCNN), which operates syntactically to produce icons that are ideal for statistical pattern recognition.