Research on Intelligent Traffic Safety Education System Based on Facial State Recognition

Abstract

With the continuous development of the traffic industry, the continuing education of the drivers also urgently needs to be improved under the background of epidemic normalization. In order to improve the manager′s learning status of the online training of employees, we designed a traffic safety education system based on real-time facial recognition. First of all, high-precision face recognition is achieved through a lightweight face recognition network. Head posture is estimated based on the 3D rotation of the face. When the set threshold value is reached, the warning “Please drive carefully” pops up to ensure the learning effect of drivers and help managers avoid the loss and risk of traffic accidents caused by drivers′ weak safety awareness.