Density-Sorting-Based Convolutional Fuzzy Min-Max Neural Network for Image Classification

Traditional image classification methods mostly use offline learning mode, which takes a lot of time when data is updated. In this paper, we propose a density-sorting-based convolutional fuzzy min-max neural network (DCFMNN) for image classification to solve this problem. DCFMNN is realized based on convolutional Neural Network (CNN) and density-sorting-based fuzzy min-max neural network. CNN is applied for image feature extraction. Density-sorting-based fuzzy min-max neural network is used for classification, which includes density-based sorting part and fuzzy min-max (FMM) neural network part. In the part of density-based sorting, patterns are sorted according to the points with the highest density in the same class and two densest points are considered for selection. The purpose is to overcome the influence of the pattern input order in the original FMM on the creation of the hyperbox. In the part of FMM, the fuzzy set classification method is used to enable online learning. Diverse CNN architectures are applied to DCFMNN. The benchmark image datasets were employed for evaluation on DCFMNN. Experimental results show that DCFMNN has high classification accuracy and less network complexity, and its online learning ability reduces the training time.

I. Introduction

Image classification is an image processing method that distinguishes image information according to its different features. Neural networks are often used for image classification. Artificial Neural Network (ANN) is a mathematical model that simulates the processing of complex information by human brain [1]. It can be applied to many fields, such as financial economics [2], healthcare [3], power systems [4], satellite [5], fault detection [6], voice recognition [7], pattern recognition [8], etc. However, at present, many ANN structures, such as Multi-Layer Perceptron (MLP) networks [9] and Hopfield network [10], will produce a relatively serious catastrophic forgetting problem [11] with the continuous entry of new pattern, resulting in the loss of original information. The problem of catastrophic forgetting is also known as the stability plasticity dilemma, which means that after learning new knowledge, the previously learned knowledge is almost completely forgotten. Convolutional Neural Network (CNN) is one of the commonly used image classification methods. Offline learning is based on CNN scratch training to learn new knowledge. Whenever we need to learn new data, we need to learn the learned data and new data together again, which greatly increases the training time. In order to solve this problem, we need online learning to reduce training time.