The Extended Generalized Neo-Fuzzy Network and Its Online Learning in Image Recognition Problem
Nonna Ye. Kulishova, Yevgeniy V. Bodyanskiy, Iryna Pliss · 2019
Real-time pattern recognition is required when solving many important practical problems. This paper proposes an extended generalized neo-fuzzy network that provides high recognition accuracy. The network learning algorithm based on the use of the cross-entropy criterion makes it possible to achieve the required speed important for real-time applications. The effectiveness of the network is confirmed for the human emotions recognition.