Adaptive neuro-fuzzy model for traffic signs recognition

Traffic sign recognition is a very important component of Intelligent Transport Systems (ITS), which is largely based on the application of artificial intelligence today. The aim of this study is to explore the ability to recognize traffic signs on an image using an adaptive neuro-fuzzy inference system (ANFIS) model. To date, many studies with the area of concern related to the recognition of traffic signs have been published. However, the application of the ANFIS model as a possible solution has not been sufficiently explored. The methodology presented in this paper uses the geometric properties of symbols on a traffic sign as input ANFIS variables. It is proposed to develop five independent models that should categorize the sign presented. The final decision is made based on the majority of the outputs of the ANFIS model, and the method showed a high level of recognition accuracy and adaptability.