Applicability of Classical and Hybrid Neural Network Algorithms in Problems of Recognition of Biometric Patterns
Vladimir I. Vasilyev, Alexey Е. Sulavko, Grigory A. Fofanov, Danil P. Inivatov · 2018 XIV International Scientific-Technical Conference on Actual Problems of Electronics Instrument Engineering (APEIE) · 2018
To date, there is a wide variety of methods and algorithms for pattern recognition in a neural network logical basis (“deep” convolutional, evolutionary, hybrid, small neural networks). However, not all of these methods are applicable in problems of biometric authentication, identification, as well as recognition of states (resource, functional, emotional) of human condition. The article is devoted to research of achievements in the field of construction of highly reliable pattern recognition automatons in order to identify the most promising of them for application in the specified problems.