The physiological state recognition and driving safety early warning system based on iPPG technology is proposed under the premise that for a long time in the future, smart cars based on SAE grade L2-L3 will be the mainstream of development, and people will still be the dominant driver. This project will measure the driver's facial data through non-contact image photoelectric fusion pulse wave tracing method, and realize the real-time detection of physiological data such as heart rate, body temperature, blood oxygen saturation, blood pressure and so on of the driver and vehicle personnel by the change of optical signal from photosensitive sensor. Through data analysis and scene judgment of fuzzy comprehensive evaluation algorithm, the physiological data will be innovatively analyzed and utilized, and the driving environment and physiological quality of the driver will be judged through data analysis, to distinguish the dangerous environment such as fatigue driving, sudden illness and so on, and complete the quantitative risk assessment. The on-board computer realizes active obstacle avoidance and auxiliary driving function adjustment to ensure the driver's driving safety. At the same time, the driving service APP software is also designed separately to provide drivers with targeted services and personalized interactive assistance based on data analysis and scene determination. It breaks the current situation of human-led and passive response of equipment, and creates a new intelligent car service idea of "active" human-vehicle interaction by car-led and on-board systems.