Monitoring drivers’ sleepy status at night based on machine vision

Driver fatigue is a chief cause of traffic accidents. For this reason, it is essential to develop a monitoring system for drivers’ level of fatigue. In recent years, driver fatigue monitoring technology based on machine vision has become a research hotspot, but most research focuses on driver fatigue detection during the day. This paper presents a night monitoring system for real-time fatigue driving detection, which makes up for the deficiencies of fatigue driving detection technology at night. First, we use infrared imaging to capture a driver’s image at night, and then we design an algorithm to detect the driver’s face. Second, we propose a new eye-detection algorithm that combines a Gabor filter with template matching to locate the position of the corners of the eye, and add an eye-validation process to increase the accuracy of the detection rate. Third, we use a spline function to fit the eyelid curve. After extracting eye fatigue features, we use eye blinking parameters to evaluate fatigue. Our system has been tested on the IMM Face Database, which contains more than 200 faces, as well as in a real-time test. The experimental results show that the system has good accuracy and robustness.