PATTERN RECOGNITION OF FACE IMAGES USING BIDIRECTIONAL ASSOCITIVE MEMORY

Abstract

The focus of the article is the exploration and application of Bidirectional Associative Memory (BAM), a form of recurrent neural network, for the purpose of recognizing human facial images. The article presents a novel algorithmic framework rooted in Hebb's rule, tailored specifically for the recognition of human faces. This pattern recognition system is characterized by its compact size and efficiency. The output signal undergoes testing in conjunction with electronic devices and notification speakers, serving as a reliable indicator for correct and incorrect pattern recognitions. It is adaptable for ON/OFF control, making it well-suited for various applications, particularly in the realm of small and medium-sized systems such as smart homes and their equivalents.