HOLOGRAPHIC IMPLEMENTATION OF A BINARY ASSOCIATIVE MEMORY FOR IMPROVED RECOGNITION

Neural network associative memory has found wide applications in patte rn recognition techniques. We propose an associative memory model for binary character recognition. The interconnection strengths of the mem ory are binary valued. The concept of sparse coding is used to enhance the storage efficiency of the model. The question of imposed precondi tioning of pattern vectors, which is inherent in a sparsely coded conv entional memory, is eliminated by using a multistep correlation techni que and the ability of correct association is enhanced in a real-time application. A potential optoelectronic implementation of the proposed associative memory is also described. The learning and recall is poss ible by using digital optical matrix-vector multiplication, where full use of parallelism and connectivity of optics is made. A hologram is used in the experiment as a longterm memory (LTM) for storing all inpu t information. The short-term memory or the interconnection weight mat rix required during the recall process is configured by retrieving the necessary information from the holographic LTM. (C) 1998 Society of P hoto-Optical Instrumentation Engineers. [S0091-3286(98)02003-0].