Deep learning based classification of time series of Chen and Rössler chaotic systems over their graphic images

Abstract

In this study, the graphic images of time series of different chaotic systems are classified with deep learning methods for the first time in the literature. For the classification, a datasetcontains images of time series of Chen and Rossler chaotic systems for different parameter values, initial conditions, step size and time length are generated. Then, high accuracy classifications are performed with transfer learning methods. The used transfer learning methods are SqueezeNet, VGG-19, AlexNet, ResNet50, ResNet-101, DenseNet-201, ShuffleNet, and GoogLeNet. According to the problem, classifications accuracy is varying between 89% and 99.7% in this study. Thus, this study shows that identifying a chaotic system from its graphic image of time series is possible.