KR102021932B1 - Driver sleepiness detection method - Google Patents

Abstract

The present invention relates to a driver's drowsiness detection method, and more particularly, to a driver's drowsiness detection method for accurately detecting a driver's drowsiness and informing a driver of a drowsiness occurrence state when the driver's drowsiness is detected. The configuration of the present invention comprises the steps of: a) obtaining heart rate data by measuring heart rate variability (HRV); b) measuring heart rate and respiratory volume; c) calculating a low frequency (LF) and a high frequency (HF) by performing a Fast Fourier Transform (FFT) on the heart rate data when the measured heart rate and respiratory rate are maintained or decreased; d) obtaining a sum of the low and high frequencies calculated according to a fast Fourier transform; And e) calculating a ratio of high frequency sum to low frequency sum (LF / HF) to detect drowsiness. In step e), when drowsiness is detected, the drowsiness warning notification is activated. To provide a driver drowsiness detection method.

Description

Driver drowsiness detection method {DRIVER SLEEPINESS DETECTION METHOD}

The present invention relates to a driver's drowsiness detection method, and more particularly, to a driver's drowsiness detection method for accurately detecting a driver's drowsiness and informing a driver of a drowsiness occurrence state when the driver's drowsiness is detected.

Accidents caused by drowsy driving of a driver while driving a car are a big social problem, and the longer the driving, the higher the probability of drowsy driving.

Accordingly, various devices and methods have been developed to warn drowsiness of a driver while driving or to prevent drowsiness.

For example, a technology for determining sleepiness by photographing a driver's facial expression has been developed. Specifically, the drowsiness detection technology using a camera was driven by an alarm ringing when the camera installed in front of the driver's seat observes the pupil and the eyelid is closed for a long time or the pupil disappears from the camera. However, this method has a problem in that, if the pupil of the driver is out of a predetermined position, there is a high possibility of malfunction, and the installation cost is expensive.

In addition, the method of detecting the drowsiness of the driver by measuring the biosignal of the driver using an electrode or a wearable device has a problem of making the driver feel uncomfortable while driving because it is mounted directly on the driver's body.

Accordingly, there is a need for a driver drowsiness detection method that does not require a driver to wear a separate wearable device, detects a biosignal as soon as the driver sits in a driver's seat, and accurately detects whether the driver is sleepy.

Korea Patent Registration 10-1421057

An object of the present invention for solving the above problems is more specifically to provide a driver drowsiness detection method for accurately detecting the driver's drowsiness, when the driver's drowsiness is detected to inform the driver of the drowsiness dangerous state.

The technical problem to be achieved by the present invention is not limited to the technical problem mentioned above, and other technical problems not mentioned above may be clearly understood by those skilled in the art from the following description. There will be.

The configuration of the present invention for achieving the above object comprises the steps of: a) obtaining heart rate data by measuring HRV; b) measuring heart rate and respiratory volume; c) calculating a low frequency (LF) and a high frequency (HF) by performing a Fast Fourier Transform (FFT) on the heart rate data when the measured heart rate and respiratory rate are maintained or decreased; d) obtaining a sum of the low and high frequencies calculated according to a fast Fourier transform; And e) calculating a ratio of high frequency sum to low frequency sum (LF / HF) to detect drowsiness. In step e), when drowsiness is detected, the drowsiness warning notification is activated. To provide a driver drowsiness detection method.

In an embodiment of the present invention, in the step a), the heartbeat data may be accumulated and updated according to a first-in first-out (FIFO) method every second.

In an embodiment of the present invention, in the step c), the fast Fourier transform may be performed on 10 to 20 heart rate data most recently accumulated and updated.

In an embodiment of the present invention, after step a), the method may further include correcting the heart rate data.

In an embodiment of the present invention, the step e) comprises: e1) calculating a ratio of the high frequency sum to the low frequency sum; e2) accumulating drowsiness detection points when the ratio of the high frequency sum to the low frequency sum is less than 0.74; And e3) determining that drowsiness is detected when the drowsiness detection point is accumulated more than seven times within 10 seconds.

The configuration of the present invention for achieving the above object provides a vehicle to which the driver drowsiness detection method is applied.

In an embodiment of the present invention, the contactless ballistic ballistic sensor may be mounted on the driver's seat of the vehicle to which the driver's drowsiness detection method is applied.

According to the effect of the present invention according to the above configuration, first, since the non-contact type ballistic sensor is mounted on the driver's seat, the drowsiness detection can be started as soon as the driver is seated in the driver's seat, and the sensor is the driver's body. Because it is not mounted directly on the driver, the driver can use it without feeling inconvenience.

In addition, according to the present invention, it is possible to determine whether drowsiness occurs primarily by measuring heart rate and respiratory rate, and secondly, accurate accuracy of drowsiness detection is detected by detecting drowsiness through analysis of heart rate variability.

The effects of the present invention are not limited to the above-described effects, but should be understood to include all the effects deduced from the configuration of the invention described in the detailed description or claims of the present invention.

1 is a flow chart of a driver drowsiness detection method according to an embodiment of the present invention.
2 is a flowchart illustrating steps of detecting whether the driver is drowsy in the driver drowsiness detection method according to an embodiment of the present invention.
3 is a graph showing the ratio of the high frequency sum to the low frequency sum over time according to an embodiment of the present invention.
Figure 4 is a graph showing the drowsiness detection test results according to an embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings will be described the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and like reference numerals designate like parts throughout the specification.

Throughout the specification, when a part is said to be "connected (connected, contacted, coupled)" with another part, it is not only "directly connected" but also "indirectly connected" with another member in between. "Includes the case. In addition, when a part is said to "include" a certain component, this means that it may further include other components, without excluding the other components unless otherwise stated.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. As used herein, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, action, component, part, or combination thereof described on the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a flow chart of a driver drowsiness detection method according to an embodiment of the present invention.

As illustrated in FIG. 1, the driver drowsiness detection method may first perform a step (S10) of obtaining heart rate data by measuring HRV.

In operation S10 of obtaining heart rate data by measuring a heart rate variability, the heart rate data may be accumulated and updated according to a first-in first-out (FIFO) method every second. Specifically, the heart rate data measured by the heart rate variability may be measured once per second. The measured heart rate data may be accumulated and updated up to a predetermined number according to a first-in first-out method. Here, the predetermined number may be 10 to 20, but is not limited thereto.

For example, if 16 heart rate data are measured and accumulated for 16 seconds, when new heart rate data is measured after 1 second, the longest heart rate data is deleted, and new heart rate data is accumulated and updated. Can be.

However, the measurement speed of the heart rate data and the number of heart rate data are not limited to one embodiment, and may be adjusted according to circumstances.

After the step of obtaining the heart rate data by measuring the degree of variance (S10), the step of correcting the heart rate data (S20) may be performed.

In the correcting of the heart rate data (S20), the abnormal data may be replaced with the most recent valid value except for the abnormal heart rate data. Specifically, the heart rate data of the heart rate variability is measured continuously in real time. At this time, the signal of the sensor is too strong or too weak may result in an abnormal value such that the value of the heart rate data is zero. In the step of correcting the heart rate data (S20), when the heartbeat data has an abnormal value as described above, the heartbeat data may be input with a valid value measured most recently except for the abnormal value.

For example, when the value of the heart rate data is 0 and the heart rate data value measured immediately before is a valid value, the value of the heart rate data measured this time may be input as the same as the heart rate data measured immediately before.

After correcting the heart rate data (S20), a step (S30) of measuring heart rate and respiratory volume may be performed.

In the step of measuring the heart rate and the respiratory rate (S30), by measuring the heart rate and respiratory volume of the driver in real time, it is possible to detect the amount of change in the heart rate and respiratory volume. In addition, when the driver's heart rate and respiratory rate appear to remain constant or decrease for a predetermined time, the next step may be determined by determining that there is a risk of drowsiness. On the other hand, if the driver's heart rate and breathing volume appear to increase over a predetermined time period, it is determined that there is no risk of drowsiness, and the driver's heart rate and breathing volume can be continuously measured.

Here, the heart rate and the breathing volume may be taken as a reference by measuring the heart rate and the breathing volume for a predetermined time from the first time the driver is seated in the driver's seat. Alternatively, the heart rate and the breath volume when the driver is in a normal state may be input as a reference.

After the step of measuring the heart rate and respiratory rate (S30), if the measured heart rate and respiratory rate is maintained or lowered, by performing a fast Fourier transform (FFT) on the heart rate data low frequency (LF) and high frequency (HF) Computing step (S40) may be performed.

When the measured heart rate and respiratory rate are maintained or lowered, the fast Fourier transform is performed at step S40 of performing a fast Fourier transform (FFT) on the heart rate data to calculate a low frequency (LF) and a high frequency (HF). The calculation may be performed on the 10 to 20 heart rate data most recently accumulated and updated.

For example, when the number of accumulating and updating the heart rate data is set to 16, the fast Fourier transform is performed on the 16 most recently accumulated and updated heart rate data so that low and high frequencies may be calculated. Here, the frequency of the low frequency may be 0.04 Hz or more and less than 0.15 Hz, and the frequency of the high frequency may be 0.15 or more and less than 0.40 Hz.

When the measured heart rate and respiratory rate are maintained or decreased, after performing a fast Fourier transform (FFT) on the heart rate data to calculate a low frequency (LF) and a high frequency (HF) (S40), the fast Fourier transform Obtaining the sum of the low frequency and the high frequency calculated according to (S50) may be performed.

Specifically, obtaining the sum of the low frequency and the high frequency calculated according to the fast Fourier transform (S50), if the measured heart rate and respiratory rate is maintained or lowered, fast Fourier transform (FFT) for the heart rate data In operation S40 of calculating the low frequency LF and the high frequency HF, the calculated values may be classified into the low frequency and the high frequency according to the frequency. The low frequency sum may be calculated by summing all classified low frequency values, and the high frequency sum may be calculated by summing all classified high frequency values.

After the step S50 of obtaining the sum of the low frequency and the high frequency calculated according to the fast Fourier transform, the step S60 of calculating the ratio (LF / HF) of the sum of the high frequency to the low frequency sum may be performed (S60). Can be.

2 is a flowchart illustrating steps of detecting whether the driver is drowsy in the driver drowsiness detection method according to an embodiment of the present invention.

As shown in FIG. 2, the step (S60) of detecting the drowsiness by calculating the ratio (LF / HF) of the high frequency sum to the low frequency sum (S61) first includes calculating the ratio of the high frequency sum to the low frequency sum (S61). Can be performed.

Computing the ratio of the high frequency sum to the low frequency sum (S61), using the low frequency sum and the high frequency sum calculated in the step (S50) of obtaining the sum of the low frequency and the high frequency calculated according to the fast Fourier transform, The ratio (LF / HF) of the high frequency sum to the sum can be calculated.

After calculating the ratio of the high frequency sum to the low frequency sum (S61), when the ratio of the high frequency sum to the low frequency sum is less than 0.74, the step of accumulating drowsiness detection points (S62) may be performed.

When the ratio of the high frequency sum to the low frequency sum is less than 0.74, the step of accumulating the drowsiness detection point (S62) includes the ratio of the low frequency sum to the high frequency sum calculated in the step S61 of calculating the ratio of the high frequency sum to the low frequency sum 0.74. If less, the drowsiness detection point can be accumulated.

When the ratio of the high frequency sum to the low frequency sum is less than 0.74, the step of accumulating the drowsiness detection point (S62) may be performed in real time according to the newly updated heartbeat data at a predetermined time interval, and may be accumulated and updated by a predetermined number. .

For example, when the heart rate data is accumulated and updated at one second intervals, whether the drowsiness detection point is accumulated once per second may be determined.

When the ratio of the high frequency sum to the low frequency sum is less than 0.74, after accumulating the drowsiness detection point (S62), when the drowsiness detection point is accumulated seven or more times within 10 seconds, determining that the drowsiness is detected (S63). Can be performed.

When the drowsiness detection point is accumulated more than 7 times in 10 seconds, in step S63, if the drowsiness detection point is accumulated more than 7 times in the last 10 seconds, the driver's drowsiness is detected It can be judged that.

Specifically, the drowsiness detection point, when the ratio of the high frequency total to the low frequency total is less than 0.74, in the step of accumulating the drowsiness detection point (S62), it is determined whether to accumulate in a unit of 1 second, whether or not accumulated for up to 10 seconds The setting can be made to be determined.

That is, when the ratio of the drowsy detection point is less than 0.74, the sum of the high frequency sum to the low frequency sum, in the step of accumulating the drowsy detection point (S62), it may be determined whether the drowsy detection point is accumulated once per second for 10 seconds. As such, when the drowsiness detection point is accumulated more than seven times in the last 10 seconds, it may be determined that the drowsiness of the driver is detected.

In the step (S60) of detecting the drowsiness by calculating the ratio (LF / HF) of the high frequency total to the low frequency total, when the drowsiness is detected in the same manner as above, the drowsiness warning notification may be activated to enable the driver to experience drowsiness. You can tell that there is.

3 is a graph showing the ratio of the high frequency sum to the low frequency sum over time according to an embodiment of the present invention.

As shown in FIG. 3, as time passes by the driver becomes drowsy, the ratio of the low frequency sum to the high frequency sum gradually decreases. And, as shown by c, when the ratio of the high-frequency total to the low-frequency total to less than 0.74 from 931 seconds, the drowsiness warning notification may start from 937 seconds. At 937 seconds, there are seven drowsiness detection points within the last 10 seconds. Accordingly, the drowsiness warning notification may be made at 937 seconds.

Figure 4 is a graph showing the drowsiness detection test results according to an embodiment of the present invention.

The pink line illustrated in FIG. 4 refers to the drowsiness detection point, and a refers to the drowsiness warning notification point where drowsiness is detected to generate a drowsiness warning. And b refers to the point where the actual drowsiness of the driver begins.

As shown in FIG. 4, the present invention enables more accurate drowsiness detection by informing the user that drowsiness is detected within 3 minutes before the actual drowsiness starts.

The sensor of the driver drowsiness detection method is a contactless ballistic ballistic sensor, it may be mounted on the driver seat. Since the driver's seat is equipped with a non-contact type ballistic sensor, the driver can start to detect drowsiness immediately after sitting in the driver's seat, and the driver is not mounted directly on the driver's body so that the driver feels uncomfortable. This is possible. Here, the sensor may be a Murata ultra low noise MEMS-based ballistic sensor. However, the type of the sensor is not limited thereto.

In addition, according to the present invention, it is possible to determine whether drowsiness occurs primarily by measuring heart rate and respiratory rate, and secondly, accurate accuracy of drowsiness detection is detected by detecting drowsiness through analysis of heart rate variability.

The foregoing description of the present invention is intended for illustration, and it will be understood by those skilled in the art that the present invention may be easily modified in other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is represented by the following claims, and it should be construed that all changes or modifications derived from the meaning and scope of the claims and their equivalents are included in the scope of the present invention.

Claims (7)

a) measuring heart rate variability (HRV) to obtain heart rate data;
b) measuring the driver's heart rate and respiratory volume to detect a change in heart rate and respiratory volume by continuously measuring the driver's heart rate and respiratory volume in real time;
c) The heart rate and the respiratory volume for a preset time from when the driver is seated in the driver's seat is measured or based on the previously input values of the heart rate and the respiratory volume when the driver is in a normal state, and the measured heart rate and respiratory volume are measured. If it is maintained or lowered than this criterion, it is determined that there is a risk of drowsiness, and a fast Fourier transform (FFT) is performed on the most recently accumulated and updated 10 to 20 heart rate data among the low heart rate (LF) and Calculating a high frequency (HF);
d) classifying the low frequency and high frequency values calculated according to the fast Fourier transform to obtain a sum of the high frequency sum of all the low frequencies and the high frequency sum of the high frequency values; And
e) calculating the ratio of high frequency sum to low frequency sum (LF / HF) to detect drowsiness;
In step e), when the drowsiness is detected, the drowsiness warning notification is activated to inform the driver of the possibility of drowsiness,
In the step a), the heart rate data is accumulated and updated up to a predetermined number according to a first-in first-out (FIFO) method every second,
After the step a), except for the abnormal heart rate data, further comprising the step of correcting the heart rate data by replacing the abnormal data with the most valid valid value measured,
In the step a), the driver's heart rate data is not directly mounted on the driver's body, but is measured by the contactless cardiograph sensor provided in the driver's seat so that the driver's heart rate is measured by immediately measuring the driver's heart rate as soon as the driver is seated in the driver's seat. Measured by
Step e),
e1) calculating a ratio of the high frequency sum to the low frequency sum;
e2) accumulating and updating the drowsiness detection point by a predetermined number in real time according to the newly updated heartbeat data at a predetermined time interval when the ratio of the high frequency sum to the low frequency sum is less than 0.74; And
e3) when the drowsiness detection point has accumulated more than 7 times in 10 seconds, determining that drowsiness is detected. delete delete delete delete An automobile to which the driver drowsiness detection method according to claim 1 is applied. delete

Images