Research on Smart Car Safety Based on YOLOv3 Target Detection Algorithm

In recent years, smart cars have achieved rapid development. Environmental awareness technology is the basis of behavior prediction, decision-making and control of moving targets around smart cars, which largely determines the intelligence level of cars. In the complicated and changeable road traffic environment, the negligence and inattention of vehicle drivers may cause irreversible losses. Real-time classification and location of surrounding objects is a key link to ensure road traffic safety during vehicle driving. Among them, the lane line detection and lane departure warning system based on vision is a hot topic in the research of smart cars at present, and it is also one of the key technologies to realize the safe driving assistance function of automobiles. Therefore, this article combines the improved YOLOv3 algorithm with target object detection and distance measurement. The program outputs 6 categories of target object classification and location information and 4 categories of longitudinal distance between the target object and the camera. The video detection speed reaches 29.8 frames per second, meeting the real-time requirements. This method can provide a reference for vehicle-assisted driving in natural road traffic scenarios.