Machine vision based fatigue driving monitoring method and system

Abstract

The invention discloses a machine vision based fatigue driving monitoring method and a machine vision based fatigue driving monitoring system. The method comprises the following steps: A1, acquiring a face image of a driver; B1, positioning a human face area according to a pre-trained human face feature classifier based on the face image; C1, positioning a human eyes area according to a pre-trained human eyes feature classifier based on the human face area; D1, acquiring an iris image in the human eyes area; E1, analyzing an eyes closing state based on the iris image; and F1, comparing the eyes closing state and a fatigue standard to judge whether the driver is in a fatigue state. The method and the system detect the human face and the human eyes through the human face classifier and the human eyes classifier respectively, reduce the influence of individual difference on a detection result, reduce the influence of lighting and human face gestures on the detection result, and improve the accuracy of eyes positioning and eyes closing judgment of the driver so as to improve the accuracy of fatigue judgment.

Description

【Technical field】 [0001] The invention relates to the technical field of safe driving of vehicles, in particular to the technical field of fatigue driving monitoring. 【Background technique】 [0002] Fatigue driving is one of the important hidden dangers of today's traffic safety. When the driver is in a state of fatigue, the ability to perceive the surrounding environment, the ability to judge the situation, and the ability to control the vehicle are all reduced to varying degrees, and traffic accidents are prone to occur. Therefore, the research and development of high-performance real-time monitoring and early warning technology of driver fatigue status can effectively reduce the hidden dangers caused by fatigue driving, so as to achieve the purpose of ensuring the personal safety of drivers and the safety of related personnel around them. [0003] The driver fatigue detection system refers to the sampling of information such as driver physiological signals, driver physio...

AI Technical Summary

Problems solved by technology

[0006] 1. The positioning of the driver's face and eyes is not accurate

Due to the large individual differences in driver's face and eye features, when the existing monitoring method is used to locate the face and eyes of various drivers, the accuracy is very low, and the position of the feature points of the upper and lower eyelids and the position of the eyebrows Individual differences are also large, which affects the accuracy of judging the degree of eye closure, and ultimately affects the accuracy of judging the fatigue state

[0007] 2. Image processing efficiency is not high

The processing speed of the collected image is slow, and it takes a few seconds to judge a frame of image, which makes it impossible to monitor the driver's fatigue state in time

In the driving process, the driver often appears drowsy (such as dozing off) or distracted in a short period of time. If the driver cannot be monitored and reminded in time, the significance of fatigue monitoring will be greatly reduced.

[0008] 3. In the image processing process, the existing methods (such as the existing active shape model method) are greatly affected by light, and the accuracy of judgment is low

Images