Enhanced multidimensional neo-fuzzy classification system and its learning for the video classification task

Abstract

A novel hybrid neo-fuzzy system for video classification, which includes multidimensional neo-fuzzy components with adjustable synaptic weights and kernel membership functions, is proposed. This system combines the strengths of extended neo-fuzzy neurons (ENFN) and neo-fuzzy units (NFU) with nonlinear activation functions. By integrating extended nonlinear synapses (ENS) and leveraging the neuro-fuzzy Takagi-Sugeno-Kang inference system, proposed architecture enhances the approximating capabilities of traditional models. This allows the system to effectively address the task of image recognition, including real-time video stream classification, while maintaining a high level of accuracy, as demonstrated by computational experiment.