5.3.2. Traps and Grating Dark Decay 3288 5.3.3. Other Concerns 3288 6. New Photorefractive Materials 3289 6.1. Polymer Composites 3289 6.2. Organic Amorphous Glasses 3292 6.3. Fully Functionalized Polymers 3294 6.4. Polymer-Dispersed Liquid Crystals 3295 6.5. Other Liquid Crystal-Containing Materials 3296 6.6. Near-Infrared-Sensitive Materials 3297 6.7. Other Materials Directions 3298 6.7.1. Hybrid Organic−Inorganic Composites and Glasses 3298

Organic photorefractives: mechanisms, materials, and applications.

Artificial intelligence tasks across numerous applications require accelerators for fast and low-power execution. Optical computing systems may be able to meet these domain-specific needs but, despite half a century of research, general-purpose optical computing systems have yet to mature into a practical technology. Artificial intelligence inference, however, especially for visual computing applications, may offer opportunities for inference based on optical and photonic systems. In this Perspective, we review recent work on optical computing for artificial intelligence applications and discuss its promise and challenges.

Inference in artificial intelligence with deep optics and photonics.

It has been shown that it is possible to to generate very small lenses by melting 'islands' of photoresist on a glass substrate. The authors have made lenses with diameters ranging from 5 mu m to 750 mu m in the form of spheres, cylinders and crossed cylinders and have studied their optical properties. Lenses with numerical apertures between 0.5 and 0.2 may be made close to diffraction limited but those of lower numerical aperture can only be achieved using more complex techniques.

https://iopscience.iop.org/article/10.1088/0957-0233/1/8/016/pdf

The manufacture of microlenses by melting photoresist

Convolutional neural networks (CNNs) excel in a wide variety of computer vision applications, but their high performance also comes at a high computational cost. Despite efforts to increase efficiency both algorithmically and with specialized hardware, it remains difficult to deploy CNNs in embedded systems due to tight power budgets. Here we explore a complementary strategy that incorporates a layer of optical computing prior to electronic computing, improving performance on image classification tasks while adding minimal electronic computational cost or processing time. We propose a design for an optical convolutional layer based on an optimized diffractive optical element and test our design in two simulations: a learned optical correlator and an optoelectronic two-layer CNN. We demonstrate in simulation and with an optical prototype that the classification accuracies of our optical systems rival those of the analogous electronic implementations, while providing substantial savings on computational cost.

/pdf/hybrid-optical-electronic-convolutional-neural-networks-with-1rnvcu8ug3.pdf

Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification.

The spatial resolution capacity of human foveal retina

A set of twelve specially doped lithium niobate crystals were grown to test the effect of the dopant on holographic recording in the crystals via the photorefractive effect. The crystals were doped with Ce, Co, Cr, Cu, Fe, Mn, Ni, Rh, Tb, Fe:Ce, Fe:Cr, and Fe:Mn. The transmission spectra was measured for each crystal and holograms have been written in each of the crystals with wavelengths from 457 nm to 671 nm. The wavelength sensitivity, scattering, and stability of the holograms varied substantially among the crystals. A qualitative description of the hologram’s properties and a comparison of sensitivities between the crystals will be presented.

Holographic recording in specially doped lithium niobate crystals.

Correlation experiments for the images used in the proposed setup indicate that nonlinear compression of the joint power spectrum may be necessary to produce good correlation performance and a peak-to-sidelobe ratio of larger than unity.

Experiments on nonlinear joint transform correlator using an optically addressed spatial light modulator in the Fourier plane.

We demonstrate a simple all-optical realization of programmable edge enhancement and edge-enhanced correlation using novel photorefractive polymers. We show that the higher non-Bragg order in a two-beam coupling scheme contains the edge enhancement of the object when placed in the path of one of the incident beams. Also, this arrangement provides a scheme for writing joint transform correlation dynamic holograms, which can be read by a third beam. The correlation is edge enhanced, and the correlation peak increases with the applied bias voltage. Numerical results without and with beam fanning are presented. Theoretical predictions are reconciled with experimental results.

Edge enhancement and edge-enhanced correlation with photorefractive polymers.

A new photopolymer from Polaroid has been investigated for producing holographic lens arrays. Arrays such as this may be important elements in optical computing architectures requiring large numbers of parallel interconnections.

Design Of Photopolymer Holograms For Optical Interconnect Applications

The Army Aviation and Missile Command (AMCOM), Morgan Research Corporation, and Aegis Research Corporation are developing an SOI-based vibratory-rate z-axis MEMS gyroscope utilizing force-feedback control, and intended for wide dynamic range and harsh environment applications. Rate sensing in small diameter ballistic missile guidance units requires a rate resolution of less than 1 degree(s)/hr over a range of -3000 to +3000 degree(s)/sec, resulting in a dynamic range of 107. In addition, the devices must operate through military specifications on temperature (-55 degree(s)C to +125 degree(s)C) and vibration (1000 g at 5 - 15 kHz). This paper presents modeling, simulation, and fabrication efforts, as well as initial test data, for an SOI-based rate sensor intended for this application. The prototyped gyroscope is a single layer structure consisting of a proof mass placed in a three-fold mode-decoupled symmetric suspension. The device is fabricated in a cost-effective and highly-controllable Silicon-on-Insulator (SOI) process for in-plane inertial devices. The mechanical structure is integrated in a vacuum-sealed hermetic package with a separate CMOS readout ASIC. At the present time, the device has undergone two design iterations, with the most recent just completed.© (2001) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Tracy Dean Hudson

Papers

Holographic recording in specially doped lithium niobate crystals.

Experiments on nonlinear joint transform correlator using an optically addressed spatial light modulator in the Fourier plane.

Edge enhancement and edge-enhanced correlation with photorefractive polymers.

Design Of Photopolymer Holograms For Optical Interconnect Applications

Single-layer silicon-on-insulator MEMS gyroscope for wide dynamic range and harsh environment applications