Driver fatigue detection method based on blink rule recognition

Abstract

The invention discloses a driver fatigue detection method based on blink rule recognition, and the method comprises the steps: firstly extracting eye images, building a training set, and then traininga model; recognizing the eye images of a driver, and if the continuous multiple frames are recognized as eye closing, displaying that the driver is in a fatigue state. According to the method, multiple algorithms are combined to form a whole, the eyes can be finely positioned, and the eye opening and closing state of an eye part matrix can be judged more accurately; and when the scene changes drastically, the false detection rate of eye opening and closing is low, so that the recognition precision of fatigue detection through blinking rules is relatively high.

Description

technical field [0001] The invention belongs to the technical field of machine vision, and in particular relates to a driver fatigue detection method based on the recognition of blinking rules. Background technique [0002] 1) With the improvement of living standards in our country, people's clothing, food, housing and transportation have been greatly improved, and there has been a qualitative leap in transportation. The well-connected roads and convenient transportation have greatly shortened the distance between people, but traffic accidents have also become more frequent. Among them, fatigue driving is gradually becoming one of the main causes of traffic accidents. Fatigue driving refers to the decline in the driver's response ability due to lack of sleep or long-term continuous driving. This decline is manifested in inattention, slow response, and improper operation. and other issues, which directly lead to the occurrence of traffic accidents. If the driver is reminded...

AI Technical Summary

Problems solved by technology

[0007] (3) The way of using the feature points extracted from the human face, although the detected eye feature points are hardly affected by factors such as light intensity, scene changes, human eye size, and human eye shape when distinguishing between eye opening and closing. influence, and achieve better accuracy, however, if the person lowers his head during eye blink detection, in most cases it will appear as figure 2 As shown, the eyebrows are misjudged as human eyes, which greatly reduces the recognition accuracy in this case, which further affects the overall recognition accuracy of the algorithm.

Images