COMPACT HOLOGRAPHIC OPTICAL NEURAL-NETWORK SYSTEM FOR REAL-TIME PATTERN-RECOGNITION

One of the important characteristics of artificial neural networks is their capability for massive interconnection and parallel processing. Recently, specialized electronic neural network processors and VLSI ne ural chips have been introduced in the commercial market. The number o f parallel channels they can handle is limited because of the limited parallel interconnections that can be implemented with one-dimensional electronic wires. High-resolution pattern recognition problems can re quire a large number of neurons for parallel processing of an image, T his paper describes a holographic optical neural network (HONN) that i s based on high-resolution Volume holographic materials and is capable of performing massive 3-D parallel interconnection of tens of thousan ds of neurons. A HONN with more than 16,000 neurons packaged in an att ache case has been developed. Rotation-shift-scale-invariant pattern r ecognition operations have been demonstrated with this system. System parameters such as the signal-to-noise ratio, dynamic range, and proce ssing speed are discussed. (C) 1996 Society of Photo-Optical Instrumen tation Engineers.