Design and analysis of driver sign safety monitoring system based on multi-algorithm combination

Design and analysis of driver sign safety monitoring system based on multi-algorithm combination

연구 분야: Verification

학회: EITCE '23: Proceedings of the 2023 7th International Conference on Electronic Information Technology and Computer Engineering

The purpose of this study is to explore the key points of dangerous driving behavior and driver health monitoring. For this purpose, a smart steering wheel capable of monitoring dangerous driving behavior and a steering wheel cover capable of monitoring physical health have been designed. The real-time face image acquisition mainly adopts MTCNN model. This paper studies driver safety monitoring from three perspectives: cockpit environment parameters, driver body information and driver behavior state, and puts forward a design scheme of driver safety monitoring system combining artificial intelligence and Internet of Things technology. Image recognition is used to detect the driver's behavior state. The fusion algorithm based on LSTM integrates PPG and electrocardiogram signals to collect the relevant data information obtained by various sensors. Through testing and analysis, the data is accurate and effective, and the system realizes effective monitoring and alarm of the dangerous factors existing in the driver's driving process. To a certain extent, it can reduce the road traffic accidents caused by the driver's own problems, and has certain application value. The system has the ability of real-time monitoring of driving behavior and health management at the same time.