A novel associative memory model applied to handwritten character recognition

In this paper, we propose a novel associative memory model used for setting up a handwritten Alpha character recognition system. This special kind of associative memory model, is of the simple algorithm operations needed for training and recall. It is also easy to understand and is very fast, even in simulations on sequential computers. Furthermore, due to the binary weights of synapses, it is very suitable for implementations using digital VLSI technology.