A driver fatigue driving detection system and detection method based on machine vision

Abstract

The invention proposes a driver fatigue driving detection system and detection method based on machine vision, which belongs to the technical field of machine vision and machine learning. The driver's normal driving, and the equipment is low in price and small in size. It is only necessary to install a Bluetooth camera in the car and an app software in the mobile phone to realize the driver's fatigue detection; this system is convenient and easy to collect information, and only It can be adapted to any vehicle type and road conditions without an additional camera, and has consistent fatigue judgment standards and high fatigue judgment accuracy; this system integrates the fatigue characteristics of eyes, mouth and face, and improves fatigue judgment in complex driving environments The accuracy rate, combined with machine learning to quickly update the system's own parameters according to the driver's feedback to adapt to the different characteristics of different drivers, the system has short training time, fast calculation speed, and strong real-time performance.

Description

technical field [0001] The invention belongs to the technical fields of machine vision and machine learning, and in particular relates to a driver fatigue driving detection system and detection method based on machine vision. Background technique [0002] Fatigue driving is one of the main causes of traffic accidents. Around the world, many car accidents are caused by drowsy driving every year. According to the survey statistics of the U.S. National Highway Safety Administration (NHTSA), there are 100,000 traffic accidents caused by fatigue driving every year, accounting for more than 16% of the total traffic accidents. The fatigue driving detection system can detect the fatigue state of the driver in real time and send an alarm to the occupants in time when the fatigue driving occurs. Cost losses play an important role. [0003] There are mainly three types of information based on the existing fatigue driving detection system, which are the driver's physiological informa...

AI Technical Summary

Problems solved by technology

The fatigue detection system based on the driver's physiological information requires corresponding information collection equipment, such as electrocardiographs, electroencephalographs, etc., to directly contact the driver, which seriously interferes with the driver's normal driving

At the same time, due to the expensive price and huge volume of these devices, the popularization of this type of system has been further hindered.

For the fatigue detection system based on vehicle behavior information, due to the differences between different models and different road conditions, it is difficult to collect vehicle behavior information such as steering wheel angle, accelerator pedal force, distance between the vehicle and the road centerline, etc.

And because different drivers have different driving habits, it is difficult to unify the fatigue judgment standard, resulting in low fatigue judgment accuracy of this type of system

The fatigue detection system based on the driver's facial image can judge the fatigue state of the driver by extracting the fatigue features in the image, such as eye features (blink frequency), head features (head up and down) and mouth features (breathing) , but the existing systems of this type have three major defects: first, the existing systems mainly judge the fatigue state through a single facial feature, which leads to their poor adaptability in complex driving environments; second, the existing The system cannot be self-improved to adapt to the characteristics of different drivers; third, the existing system is not real-time due to insufficient calculation speed

Images