Design of fatigue driving automatic detection system based on OpenCV and Raspberry PI
The traffic safety accidents caused by the decline of judgment and reaction ability during fatigue driving will cause great harm to personal life and property safety and society.The traditional fatigue driving detection system has some problems,such as high production cost,low data accuracy and reliability,complex system process,low detection feedback lag and efficiency.This paper proposes a design scheme of fatigue driving automatic detection system based on OpenCV and Raspberry PI with high effi-ciency and low cost.Firstly,the Eigenfaces for Recognition algorithm of OpenCV Dlib library was used to collect 2000 face samples for machine deep learning training,and then the AdaBoost iterative algorithm was used to establish different weak classifier models to train strong classifiers.Secondly,the line coordinates of the eyes,mouth and other facial regions were established as reference in-dicators,and the blink frequency of the eyes and the degree of mouth closure were collected.The Raspberry PI 4B was used as the main control to drive the two-dimensional steering gimbal,and the PID algorithm was used to automatically detect and track the driver.Finally,the correct recognition detection rate of the fatigue driving automatic detection system is more than 90%.
万方数据
