Human Image Recognition System with an Artificial Neural Network Working in Three-Valued Logic

Abstract

The paper presents a method for a biometric image recognition of the human using the vector distance measure. The basis of the proposed method of recognition is a biometric characteristic picture of a man (the image of the eye, hands, etc.). To this end, the classification structure was presented of the biometric image. An important part of this study is to present the theoretical basis of the functioning of the proposed pattern recognition method and the neural network operation algorithm. The neural network engineered is a convenient research tool used in the image analysis. The structure of the image recognition system was illustrated and described. An important part of this system is a module for determining the coordinates of singular points (the image scanning module) and the recognition module that uses a SBM type neural network known as the Simile-Based Methods (SBM), working in three-valued logic.