Open AccessComputer modelling of statistical, structural and neural network methods of pattern recognition
Author(s) -
Н. М. Новикова
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1479/1/012041
Subject(s) - computer science , artificial neural network , artificial intelligence , pattern recognition (psychology) , hamming code , time delay neural network , relevance (law) , machine learning , data mining , algorithm , decoding methods , block code , political science , law
Pattern recognition is one of the most important tasks of both intelligent control systems and arti cial intelligence. The paper substantiates the relevance of the study of computer modelling of statistical, structural and neural network methods of pattern recognition. The article presents a comparative analysis of the quality of pattern recognition using the Hamming neural network and the statistical algorithm. The analysis shows that the use of the Hamming neural network is preferable in most cases. Computer modelling of the structural method using the Freeman code gives a description that allows us to unambiguously assign an object to its class. Based on the analysis of the results of computer modelling, the positive and negative aspects of each method are revealed. As a result, the structural method is the most optimal.