A driving fatigue warning method based on road markings

Abstract

The invention discloses a driving fatigue early warning method based on road markings, using a normal distribution method to divide the lateral position of the vehicle into regions in steps, and selecting regional data to form a Markov prediction model library, and finally realizing driving fatigue The early warning method includes the following steps: S1: select the time research object and data research object; S2: obtain the distance between the current vehicle and the left marking line of the driving lane through the video analysis of the driving recorder, and use the obtained data to The characteristics of the normal distribution determine the distribution range of the vehicle's lateral position; S3: divide the lateral distribution area determined in S2 in a stepwise manner, and use it as the standard for state division, and divide it into m states in total. At the same time, through the selection of data Composition Markov model library. The driving fatigue warning method based on road markings proposed by the present invention can warn drivers in advance to improve their attention, ensure the driver's property safety and provide the ability to monitor traffic safety.

Description

technical field [0001] The invention relates to the technical field of driving fatigue early warning methods, in particular to a driving fatigue early warning method based on road markings. Background technique [0002] With the development of our country's economy and the improvement of people's living standards, the number of vehicles has also increased sharply, which has caused frequent traffic accidents, and the main causes of traffic accidents are people (mainly drivers), vehicles, and roads. And the system coordination imbalance composed of factors such as the traffic environment and the traffic environment, among which the fatigue driving element is particularly prominent. [0003] The so-called driving fatigue refers to the phenomenon that the driver's psychological function and physiological function are out of balance after driving continuously for a long time, resulting in blurred vision, unresponsiveness, dull movements, and a decline in driving performance. Dri...

AI Technical Summary

Problems solved by technology

This model is a probabilistic model that describes the statistical characteristics of random processes based on parameter representation and has good data modeling capabilities. However, this model is realized by indirectly estimating the running state of the vehicle and cannot directly describe the running state of the vehicle. It increases the difficulty and error of state prediction, and is only applicable to relatively small data sets, which has certain limitations

In addition, when passing through curves with various curve radii, the vehicle's driving trajectory is different, but the driving state of the vehicle may be safe. The hidden Markov prediction method currently used cannot distinguish the vehicle behavior on the curve. state, which is not conducive to the general prediction of the future driving state of the vehicle

[0005] To sum up, most of the existing public fatigue detection methods will cause more or less interference to the driver, and there are many limiting factors, which will have a certain impact on the accuracy of the prediction results

Images