JPH10272960A - Awake degree estimating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily absorb individuality and estimate an awake degree with high accuracy by obtaining reference time of a wink on the basis of wink time while awake, detecting a long wink according to wink evaluation time set on the basis of the reference time, and evaluating the awake degree of a driver on the basis of a long wink occurrence ratio. SOLUTION: A wink reference time measuring part 14 obtains the reference time of a wink peculiar to a driver at the time of being awake, on the basis of wink time detected by a wink time computing part 12 when the driver is considered to be in a sufficiently awake state. A long wink detecting part 16 detects a wink, exceeding wink evaluation time set by a long wink time setting part 15 on the basis of reference time, as a long wink from wink time successively detected by the wink time computing part 12, and a long wink occurrence ratio computing part 17 obtains an occurrence ratio. Sleepiness prediction value is obtained according to a prediction model showing the relation between the occurrence ratio and the awake degree, and the sleepiness prediction value is evaluated to judge lowering of the awake degree of the driver.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention focuses on the time of a driver's blinking, detects the driver's arousal level, and furthermore, detects a decrease in the arousal level in a simple and reliable manner, issues a warning, and alerts the driver's attention. The present invention relates to a wakefulness estimating device suitable for the present invention.

[0002]

[Related Background Art] Recently, various systems for estimating a driver's arousal level based on various information and issuing a warning when a decrease in the arousal level is detected, for example, to arouse driving attention have been developed. Have been. One of the methods to estimate this kind of arousal
Finally, there is a method in which a driver's blink is used as an evaluation index. For example, Japanese Patent Application Laid-Open No. 61-175129 discloses a method of counting the number of blinks per unit time and determining a decrease in arousal level. I have. However, when the number of blinks per unit time is used as an index for arousal level evaluation, errors due to individual differences in blinks are likely to occur, and the accuracy of estimation cannot be improved.

Therefore, the present applicant has previously filed Japanese Patent Application No. 8-913.
No. 24, and a paper [52.
A method of estimating a decrease in arousal level by focusing on the driver's blink time was proposed as announced as a method for evaluating arousal level during driving (9632415). A method of estimating the degree of awakening focusing on the blink time is based on a frequency distribution of the blink time, and a threshold (blink time) for determining a long blink from a standard blink time distribution width and a center time of the distribution width. ) Is set, and the ratio of the total number of blinks within a predetermined period to the number of occurrences of long blinks exceeding the above-described threshold is obtained, and a decrease in arousal level is determined by evaluating this ratio.

According to such a method, there is an advantage that it is possible to absorb individual differences such as a blink time and a blink frequency and to accurately evaluate the arousal level.

[0005]

However, when the frequency distribution of the blinking time is determined, and a determination threshold value for a long blink is set in accordance with the frequency distribution and the arousal level is estimated as described above, it is assumed that the same person is present. However, since the frequency of blinking changes depending on the physical condition, the surrounding environment, and the like, it is necessary to collect blink information over a certain long period in order to improve the estimation accuracy. Moreover, it is unavoidable that the processing load in obtaining the frequency distribution from the collected blink information and analyzing the frequency distribution is large. For this reason, there is a strong demand that the arousal level can be estimated more easily and with high accuracy, and that a system (arousal level estimation device) mounted on a vehicle is simplified.

The present invention has been made in view of such circumstances, and has as its object the purpose of being able to easily and efficiently detect a decrease in arousal level and reliably alert the driver to driving, Moreover, it is an object of the present invention to provide a waking degree estimating apparatus that can simplify the system configuration.

[0007]

SUMMARY OF THE INVENTION In order to achieve the above-mentioned object, a waking degree estimating apparatus according to the present invention uses a driver's blink image (blink duration time) based on a driver's face image captured by a camera, for example. Blinking time detecting means for detecting t), reference time calculating means for calculating a blinking reference time [To] specific to the driver based on the blinking time [t] detected when the driver wakes up, Evaluation time setting means for increasing the reference time [To] by a predetermined ratio [r%] to set a blink evaluation time [Ts] for evaluating a long blink, and according to the set blink evaluation time [Ts]. A long blink detection means for detecting a long blink from the blink time [t], and a long blink occurrence ratio [L] based on the total number of blinks [Ntotal] and the number of long blinks [Nlong] within a predetermined time.
rate], and an occurrence ratio alertness determining unit for evaluating the driver's alertness [Y1] based on the long blink occurrence rate [Lrate].

[0008] In particular, in the occurrence ratio alertness determination means,
Based on the following prediction formula Y1 = A1 + B1 · Lrate (A1 and B1 are prediction coefficients) obtained by regression analysis of a sleepiness prediction model showing the relationship between the arousal level [Y1] and the occurrence ratio of long blinks [Lrate]. The sleepiness prediction value [Y1] is calculated by using the threshold value, and the sleepiness prediction value is discriminated by a predetermined threshold value to evaluate the arousal level.

The awakening degree estimating apparatus according to the present invention, in addition to the above-described configuration, further includes a blink average time calculating means for calculating an average [BLdur] of the blink time within a predetermined time, and an average blink time [BLdur]. BLdur], an average time arousal level determining means for evaluating the driver's arousal level [Y2], and comprehensively determining each of the determination results of the average time arousal level determining means and the occurrence ratio arousal level determining means. A low alertness detecting means for detecting a low alertness of a person.

[0010] In particular, in the average time awakening degree determination means,
A drowsiness prediction value based on a prediction formula Y2 = A2 + B2 · BLdur (A2, B2 is a prediction coefficient) indicating a relationship between a degree of arousal obtained by regression analysis using a drowsiness prediction model and an average blink time [BLdur] [ Y2] is calculated, and the drowsiness predicted value is discriminated by a predetermined threshold to evaluate the arousal level,
In the alertness reduction detecting means, the sleepiness predicted value [Y
1] and [Y2] are comprehensively evaluated to accurately (reliably) estimate a decrease in the driver's arousal level.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an awakening degree estimating apparatus according to the present invention will be described with reference to the drawings. FIG.
FIG. 1 conceptually shows the configuration of an embodiment device mounted on the vehicle 1. In FIG. 2, reference numeral 2 denotes a TV camera for imaging the face of the driver D, particularly the eye area. In the figure, reference numeral 3 denotes a display (multiplex information display device) for displaying various information as images and presenting it to the driver D, and reference numeral 4 denotes a speaker for outputting a voice message, an alarm sound and the like. The TV camera 2, the display 3, and the speaker 4 are incorporated in, for example, an instrument panel in front of a driver's seat.

The awakening degree estimating apparatus according to this embodiment has a TV
A blink of the driver is detected from the face image of the driver captured by the camera 2 to estimate the degree of awakening of the driver. When the degree of awakening decreases, a message is displayed via the display 3 when the degree of awakening is reduced. Plays the role of alerting and driving the driver's attention. This device is realized by, for example, an electronic control unit (ECU) mainly composed of a microprocessor, and is schematically configured as shown in FIG.

That is, as shown in the functional block diagram of FIG. 2, the embodiment apparatus recognizes a driver's face image captured and input by the TV camera 2 in the image processing unit 10 and performs, for example, a predetermined process. The partial image of the eye region is extracted in the cycle of. The blink detection unit 11 detects blinking by detecting, particularly, the opening and closing of the eyelids from the temporal change of the image. Each time a blink is detected by the blink detection unit 11, the blink time calculation unit 12 measures a blink time (blink time) [t] indicated by an eye closing time from the start to the end of closing the eye (the blink time). Blink time detecting means). Such blink time detection processing is repeatedly executed over a predetermined period under the management of the timer 13.

The blink reference time calculation unit 14 has a strong sense of "starting (starting)" as in, for example, the initial stage of driving, and it is determined that the driver D is sufficiently awake. When it can be considered, based on the blink time [t] obtained over a predetermined period by the blink time detecting means (blink time calculation unit 12), the reference time of the blink at the time of awakening unique to the driver D [To] is obtained (reference time calculating means). The reference time [To] is calculated as the average value of the blink time [t] in the predetermined period, for example, 5 minutes at the start of operation.

The long blink time setting section 15 sets a blink evaluation time [Ts] for detecting a long blink from the driver D's blink, particularly with a decrease in arousal level. By increasing [To] by a predetermined ratio [r (%)], the evaluation time [Ts] is determined as Ts = To + To · r / 100 (evaluation time setting means). The above ratio is, for example, 5%, 10%, 15%, 2
It is given as 0%, and is actually set to 10% or the like based on the results of a simulation experiment described later. That is, the long blink time setting unit 15 considers that when the degree of awakening decreases, generally a long-time blink tends to occur. A blink evaluation time [Ts] for detecting blink is set.

The long blink detecting section 16 calculates the blink evaluation time from the blink time [t] of the driver D sequentially detected by the blink time detecting means in accordance with the blink evaluation time [Ts] set as described above. Blinks exceeding [Ts] are detected as long blinks. In the long blink occurrence ratio calculation unit 17, for example, the total number of blinks [Ntotal] evaluated by the long blink detection unit 16 and the number of long blinks [Nlong] obtained as a detection result by the long blink detection unit 16 Are counted respectively. Then, based on the total number of blinks [Ntotal] in a predetermined period and the number of long blinks [Nlong] in the blink, a long blink occurrence ratio [Lrate (%)] is obtained as Lrate = 100 · Nlong / Ntotal. ing.

The number of long blinks [Nlong] and the number of other standard blinks [Nnormal] are counted, and the rate of occurrence of long blinks [Lrate = 100 · Nlong / (Nlong + Nnormal)] Lrate] may be obtained.

The occurrence ratio arousal level determination unit 18 determines the arousal level and the occurrence ratio of the long blink [Lrate (%)] obtained by regression analysis using a drowsiness prediction model as described later.
The drowsiness predicted value [Y1] is calculated based on the prediction formula Y1 = A1 + B1 · Lrate (A1, B1 is a prediction coefficient) showing the relationship with
For example, the prediction coefficients A1 and B1 are defined as five levels of arousal level as follows, and the rate of increase [r] in setting the above-described blink evaluation time [Ts] based on various simulation experiment results. %] Is set to 10% as described later, for example, given as A1 = 1.238, B1 = 0.046. That is, the prediction formula is given as Y1 = [1.238] + [0.046] Lrate, and the long blink occurrence ratio [Lrate (%)] during the blink of the driver D is calculated as described above. The drowsiness predicted value [Y1] is obtained by calculating the prediction formula.

The above five levels of arousal level are, for example, level 1 ... seemingly sleepless (movement of the line of sight is fast and frequent; blinking is stable and movement is active) level 2 ... slightly sleepy (line of sight) Slow movement. Lips open.) Level 3 ... sleepy (blinks are slow and frequent. There is movement of the mouth.) Level 4 ... quite sleepy (conscious blinks, blinks and gaze are slow. .) Level 5: Set to be very sleepy (closes eyelids, head tilts back and forth). Therefore, when the drowsiness predicted value [Y1] exceeds, for example, [3], it is estimated (determined) that the awakening degree is low and the user is likely to sleep, as shown below.

The predicted drowsiness value [Y1] obtained based on the occurrence ratio [Lrate] of long blinks by the occurrence ratio arousal level determination unit 18 is given to the arousal level decrease determination unit 19 and It is used to determine the decrease. When the alertness decrease determination unit 19 detects a decrease in alertness, the above-described warning using the display 3 and the speaker 4 is issued, and the driver D is alerted to driving attention. become.

On the other hand, the blink average time calculation section 21 calculates the average value of the blink time [t] in a predetermined period, that is, the average, based on the blink time [t] of the driver D sequentially detected by the blink time detecting means. Blink time [BLdur] is required. Since the blink average time calculation unit 21 performs substantially the same operation as the blink reference time calculation unit 14, the blink average time calculation unit 21 can be realized as a function of the blink reference time calculation unit 14.

Thus, the average time awakening degree determination unit 22 that inputs the average blink time [BLdur] obtained by the blink average time calculation unit 21 is obtained by regression analysis using a drowsiness prediction model as described later. A drowsiness prediction value [Y2] is calculated based on a prediction formula Y2 = A2 + B2 · BLdur (A2, B2 is a prediction coefficient) indicating a relationship between the arousal level and the average blink time [BLdur (sec / time)]. . The prediction coefficients A2 and B2 are, for example, the arousal level of 5 as described above.
When defined in stages, based on various simulation experiment results, it is given as, for example, A2 = -4.378, B2 = 0.029.

In this manner, the average time arousal level determination section 22
The sleepiness predicted value [Y2] obtained based on the average blink time [BLdur] is also provided to the arousal level decrease determination unit 19 to be used for determining the arousal level decrease. When the arousal level decrease determination unit 19 detects a decrease in the arousal level, the above-described warning using the display 3 and the speaker 4 is issued, and the driver D is alerted to driving attention. become.

In particular, the awakening degree decrease determination section 19 predicts a drowsiness value [Y1] obtained based on the long blink occurrence rate [Lrate] and an average blink time [BLdur].
The drowsiness prediction value [Y2] obtained based on the above is comprehensively determined as described later to detect a decrease in the arousal level of the driver D, and a warning is issued based on the detection result.

Note that the processing for determining a decrease in arousal level shown in the above-described functional blocks is actually executed under a microprocessor according to a control routine shown in FIG. That is, at the beginning of the operation start, the blinking time [t] is measured for, for example, 5 minutes (steps S1, S2, S3), and the average of the blinking time is calculated as the reference time [To] (step S4). Thereafter, one unit of measurement period is set to 5 minutes, and the subsequent blink time [t] is measured (steps S5 and S6).

Thereafter, the reference time [To] is set to 1 according to the blink time [t] detected every 5 minutes as described above.
The occurrence rate [Lrate] of a long blink [Long10] detected under the blink evaluation time [Ts] set to be 0% longer is calculated (step S8), and the average blink time [B] at that time is calculated.
Ldur] is calculated (step S9). Then, when the occurrence ratio [Lrate] and the average blink time [BLdur] are obtained, the sleepiness prediction value [Y1],
[Y2] are calculated (steps S10 and S1).
1) For example, the drowsiness predicted value (arousal level) obtained thereby is displayed on the display 3 (step S12).
The display of the predicted drowsiness value (degree of arousal) is performed by, for example, displaying the degree of arousal in a bar graph or changing the display color of the information in accordance with the five levels set as described above.

Then, the sleepiness predicted values [Y1] and [Y2] obtained as described above are evaluated (step S13). For example, when the level (predicted value) exceeds [3],
A warning is issued to awaken the driver D to encourage driving attention (step S14). If the level of the arousal level is equal to or less than [3], the processing from step S5 described above is repeatedly executed, whereby the estimating process of the arousal level based on the blink information for the next five minutes is executed again.

Now, the calculation of the sleepiness predicted values [Y1] and [Y2] based on the occurrence ratio [Lrate] of the long blink [Long10] and the average blink time [BLdur] and the evaluation thereof will be described in more detail. As the evaluation index of the arousal level, for example, a temporal change of a physiological index such as an electroencephalogram, an electrocardiogram, or respiration, and a temporal change of a performance index such as a steering operation characteristic indicated by a steering wheel angle are used. Then, when the relationship between each of these indices and the drowsiness of the driver D, which was objectively evaluated by a third party at that time, was examined, a simulation experiment result as shown in FIG. 4 was obtained.

In FIG. 4, Cz (α / β) and Pz (α / β)
Is the ratio of the spectral power values of the α waves and β waves of the brain waves obtained at the top Cz region and the Pz region of the driver D. Blink-No is the number of blinks in 5 seconds, for example, obtained from the eye movements of driver D, Blink-Dur is the average blink time, HR is the heart rate of driver D per minute, and Resp is the respiration rate. It is. Speed is the running speed of the vehicle, Steer is the average value of the steering wheel angle, and Steer-SD is the deviation of the steering wheel angle. Sleepiness is the driver's D sleepiness evaluated by a third party, and MWS is the driver's D subjective sleepiness evaluation.

As shown in the simulation experiment results illustrated in FIG. 4, the driver D sleepiness (Sleepine
ss) increases with driving time, and drowsiness counts the number of blinks (Blink-No) and blinking time (Blink-Dur)
It shows that you have a strong relationship with That is, as the drowsiness increases with the passage of time, the number of blinks tends to decrease, and the blink time tends to increase. Incidentally, as the operation time elapses, the above Cz (α / β) and Pz (α / β)
Also show a tendency to increase, and HR and Resp show a tendency to decrease. The increase in drowsiness of the driver D with the passage of time is generally caused by the monotonousness and familiarity of the driving operation, and furthermore, by fatigue.

Therefore, in the present invention, in order to evaluate the arousal level (drowsiness fluctuation) without contact with the driver D, attention is paid particularly to the time of blinking, the prediction accuracy is improved, and individual differences are reduced. Investigation was proceeded. Specifically, attention was paid to long blinks that increased with a decrease in arousal level, and the relationship between the occurrence ratio of long blinks and arousal level in a predetermined period was examined. In particular, the blink evaluation time [Ts] for judging a long blink as pre-processing is defined as an average blink time at the time of awakening as a reference time [To], and an increase ratio of the blink time with respect to the reference time [To] is calculated. The blink evaluation time [Ts] was set at 5%, 10%, 15%, and 20%, respectively.

A plurality of drivers D for a predetermined time (for example, 5 minutes) under each of the blink evaluation times [Ts].
(Subject: Subj.1, ~ 12) Long blink rate [Lrat
e] were obtained as [Long5], [Long10], [Long15], and [Long20], respectively. Further, the average blink time [BLdur] of each driver D in the predetermined time was also determined at the same time. And the occurrence ratio of long blinks [Long5], [Long10], [Long1
5], [Long20] and the average blink time [BLdur], while shifting the sample center time according to the Lag value given as, for example, a 60-second cycle, so that drowsiness (degree of arousal) is obtained.
Was analyzed to determine the relation coefficient R under the condition that the relation coefficient was the largest, and the result as shown in FIG. 5 was obtained.

According to the results of the analysis of variance regarding the relation coefficient R shown in FIG. 5, a plurality of five indices consisting of the occurrence ratios of the four types of long blinks having different blink evaluation times and the average blink time indicate a plurality of indices. There is a significant difference in the average value over driver D (Subj). Especially the occurrence ratio of long blink [Long20]
And the analysis of variance with respect to the Lag value confirmed that there was no significant difference between the above indexes. However, each of these indices is data having a physiologically different meaning.

By the way, the results of the analysis of variance shown in FIG. 5 are obtained when each index (the ratio of occurrence of long blinks) is obtained as an average for 300 seconds, and the expression of sleepiness is obtained as an average for 60 seconds. Therefore, the result of the analysis of variance implies that, with respect to the time characteristic of sleepiness prediction, for example, the average of the indices from 0 to 300 seconds predicts the average of sleepiness from 120 seconds to 180 seconds. Delay is 120
Seconds. Therefore, when a Lag value indicated by a cycle of 60 seconds is given by [−1], it is actually 300
At the time when the second index is obtained, sleepiness at the time of 240 seconds is predicted, which indicates that the prediction delay is 60 seconds. However, the average period of actual sleepiness is 15
Since the delay is less than half of 0 second, practically, the change of drowsiness can be predicted almost in real time.

Now, based on the analysis results and the consideration results shown in FIG. 5, the occurrence ratio (index) of long blinks and the arousal level (sleepiness) are modeled by prediction, and a plurality of drivers D obtained by simulation obtain the simulation results. From the sample data, a prediction expression indicating the relationship is approximated by a linear expression of Y1 = A1 + B1 · Lrate Y2 = A2 + B2 · BLdur, and the prediction coefficients A1, B1, A2, B2 when the relationship is the highest. And the average of AIC = 2n logeσ + 2p (where n is the number of data and p is the number of regression coefficients). For example, the result shown in FIG. 6 was obtained.

That is, the relationship between the arousal level obtained by regression analysis using a prediction model and the occurrence ratio [Lrate] of long blinks is expressed by a prediction formula of Y1 = A1 + B1 · Lrate (A1 and B1 are prediction coefficients). The relationship between the arousal level and the blink time is expressed by a prediction formula of Y2 = A2 + B2 · Lrate (where A2 and B2 are prediction coefficients), and based on a plurality of sample data obtained by the simulation, the relation coefficient is obtained. R
, The prediction coefficients A1, B1, A2, and B2 were determined, and the relation coefficient R and the information amount reference AIC at that time were determined. As a result, in this example, the information amount reference AI
The prediction model that minimizes C is the blink evaluation time [T
s] was set as a 10% increase in the reference time [To], it was confirmed that the occurrence ratio of long blinks [Long10] was obtained. In addition, it was confirmed that the prediction model in which the largest relation coefficient R was obtained was a case where attention was paid to the blink time [Blink-Dur].

In the present invention, based on the above findings,
As described above, the occurrence rate of long blinks of driver D [Lrat
e], and a sleepiness prediction value [Y1] corresponding to the occurrence ratio [Lrate] is obtained according to a prediction formula based on a prediction model indicating the relationship between the arousal level and the occurrence ratio. By evaluating the drowsiness predicted value [Y1], a decrease in the awakening degree of the driver D is determined. Further, the average blink time [BLdur] of the driver D is determined, and the sleepiness prediction value [Y2] corresponding to the average blink time [BLdur] is determined according to a prediction formula based on a prediction model indicating a relationship between the arousal level and the average blink time. ing. Then, by evaluating the sleepiness prediction value [Y1],
It is determined that the driver D has a lower arousal level.

FIG. 7 shows a plurality of drivers D (subject: Subj.1,
~ 12) The measured value of drowsiness obtained from the drowsiness expression,
As described above, the relationship with the sleepiness predicted value [Y1] obtained from the long blink occurrence ratio [Lrate] is shown in comparison. In FIG. 7, the characteristics indicated by the thick lines indicate the drowsiness predicted value [Y1], and the characteristics indicated by the thin lines indicate the actual changes in drowsiness. As shown in FIG. 7, it can be seen that the sleepiness predicted value [Y1] obtained as described above satisfactorily predicts a change in sleepiness (arousal level) in advance of actual sleepiness. In addition, it can be said that drowsiness is predicted by absorbing the individuality of the driver D almost satisfactorily.

FIG. 8 shows the plurality of drivers D (examinee:
The relationship between the measured drowsiness value in Subj. 1 to 12) and the drowsiness predicted value [Y2] obtained from the average blink time [BLdur] as described above is shown in comparison. In FIG. 8 as well, the characteristic shown by the thick line indicates the drowsiness predicted value [Y2], and the characteristic shown by the thin line indicates the actual change in drowsiness. As shown in FIG. 7, the predicted drowsiness value [Y2] obtained based on the average blink time [BLdur] has a slightly lower prediction accuracy than the aforementioned drowsiness predicted value [Y1]. It can be seen that the change in sleepiness (arousal level) is predicted well in advance of actual sleepiness.

However, as shown in FIGS. 7 and 8, the drowsiness predicted values [Y1] and [Y
2] can predict the decrease in the arousal level as a whole, but the prediction accuracy of the specific driver D (individual) may be insufficient. Specifically, in the comparative example shown in FIG.
For the driver D indicated by bj.1 and the driver D indicated by Subj.8, the sleepiness prediction value [Y1] obtained from the long blink occurrence ratio [Lrate] sufficiently predicts actual sleepiness. Absent. In the comparative example shown in FIG.
7, a drowsiness predicted value [Y obtained from the average blink time [BLdur] in the driver D represented by Subj.10 and Subj.11
2] does not fully predict actual drowsiness.

The insufficient prediction for the specific driver D is due to the specificity of the blink of the driver D. The specificity is a predicted drowsiness value obtained from a long blink occurrence rate [Lrate]. [Y1] and the average blink time [BLdu
r] shows the opposite tendency with the sleepiness prediction value [Y2]. In other words, the specificity of the blink has an adverse effect on the prediction accuracy of drowsiness based on the occurrence ratio [Lrate] of long blinks and the prediction accuracy of drowsiness based on the average blink time [BLdur]. When the prediction accuracy is low, the other prediction accuracy tends to be increased.

Accordingly, as described above, the drowsiness predicted values [Y1] and [Y2] obtained from the long blink occurrence ratio [Lrate] and the average blink time [BLdur] of the driver D are comprehensively determined and determined. According to the present apparatus for predicting a decrease in arousal level (sleeping), it is possible to determine a decrease in arousal level of the driver D with very high prediction accuracy and effectively absorbing various personalities. . In particular, as described above, the occurrence ratio of long blinks [L
rate] and the average blink time [BLdur].
The sleepiness prediction value [Y
Since [1] and [Y2] can be calculated, the processing load is light and the system configuration can be simplified. Moreover, the processing speed can be increased,
The time delay of the detection can be reduced.

That is, the prediction accuracy is improved while absorbing the individuality, and the processing is simplified and the processing speed is increased.
Further, the system configuration can be simplified and the cost can be reduced. Therefore, there is a great advantage in practical use as an awakening degree estimation device mounted on each vehicle. Note that the present invention is not limited to the above-described embodiment. For example, the reference time [To] of the blink evaluation time [Ts] for detecting a long blink
The increase ratio [r%] with respect to may be set based on more simulation experiment results and the like.
It is not limited to the example described above. That is, the blink evaluation time [Ts] may be set so as to obtain a higher relationship. Also, when realizing a simple type of arousal estimating device,
The estimation of the arousal level based on the average blink time [BLdur] described above can be omitted.

In place of the average blink time [BLdur], other behavioral characteristics of the driver, such as horizontal movement of the eyeball indicating movement of the line of sight, backward and forward leaning of the head, and yawning, sighing, and lips It is also possible to detect behavior such as relaxation of the subject, estimate the arousal level from a prediction model based on the information, and make a comprehensive judgment with the evaluation result of the arousal level based on the occurrence ratio of long blinks described above. Furthermore, by using the estimation result of the arousal level decrease described above, the vehicle brake mechanism is operated to decelerate,
It is also possible to activate an automatic driving mode based on recognition of a white line on the road or control of the distance between other vehicles. In addition, the present invention can be variously modified and implemented without departing from the gist thereof.

[0045]

As described above, according to the present invention, the blink time of the driver is detected, and the reference time of the blink specific to the driver is obtained based on the blink time at the time of awakening. A long blink of the driver is detected according to a blink evaluation time set by increasing by a predetermined ratio, and the driving is performed based on a long blink occurrence ratio obtained from the total number of blinks and the number of long blinks within a predetermined time. To evaluate the degree of awakening of the person, so it is simple and effective to absorb personality,
The degree of arousal can be estimated with high accuracy.

In particular, a sleepiness prediction value is calculated based on a regression analysis of a prediction formula based on a sleepiness prediction model showing a relationship between the degree of arousal and the occurrence ratio of long blinks. Since the evaluation is performed, the system configuration can be simplified and the processing speed can be increased while sufficiently increasing the estimation accuracy. Furthermore, as described in claim 3, means is further provided for obtaining an average of the blinking time within a predetermined time, and evaluating the degree of awakening of the driver based on the average blinking time. Since the arousal level and the arousal level obtained from the occurrence ratio of the long blinks are comprehensively determined to detect a decrease in the arousal level of the driver, high-precision awakening using two types of indicators having different physical meanings Degree estimation
Practically great effects such as easy operation can be obtained.

[Brief description of the drawings]

FIG. 1 is a view conceptually showing a configuration of a vehicle equipped with a waking degree estimation device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a functional block configuration of a waking degree estimation device shown in FIG. 1;

FIG. 3 is a diagram showing a flow of a wakefulness estimation process in the wakefulness estimation device shown in FIG. 2;

FIG. 4 is a diagram showing a simulation experiment result of a relationship between a plurality of arousal indices having different physical meanings and objectively evaluated drowsiness.

FIG. 5 is a graph showing the relationship between the occurrence ratio of long blinks and the relationship coefficient R when the relationship coefficient between the average blink time and drowsiness (degree of arousal) is greatest among a plurality of drivers (subjects: Subj. 1 to 12). FIG.

FIG. 6: Occurrence ratio of long blinks (index) and arousal level (drowsiness)
The figure which shows the relationship between the prediction formula at the time of making a prediction model, the average of the relationship coefficient R, and the information amount reference AIC.

FIG. 7 is a diagram showing a comparison between a measured drowsiness value of a plurality of drivers D (subjects: Subj. 1 to 12) and a drowsiness prediction value obtained from the occurrence ratio of long blinks;

FIG. 8 is a diagram showing a comparison between a measured drowsiness value of a plurality of drivers D (subjects: Subj. 1 to 12) and a drowsiness predicted value obtained from an average blink time.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Vehicle 2 TV camera 3 Display 4 Speaker 10 Image processing unit 11 Blink detection unit 12 Blink time calculation unit 13 Timer 14 Blink reference time calculation unit 15 Long blink time setting unit 16 Long blink detection unit 17 Long blink occurrence ratio calculation unit 18 Occurrence ratio arousal level determination unit 19 Arousal level decrease determination unit 21 Blink average time calculation unit 22 Average time arousal level determination unit

Claims (4)

[Claims] 1. A blink time detecting means for detecting a blink time of a driver; a reference time calculating means for calculating a reference time of a blink specific to the driver based on the blink time when the driver is awake; Evaluation time setting means for setting a blink evaluation time for evaluating a long blink by increasing a reference time by a predetermined ratio; and long blink detection means for detecting a long blink from the blink time according to the set blink evaluation time. ,
An occurrence ratio calculating means for obtaining an occurrence ratio of a long blink from the total number of blinks and the number of long blinks within a predetermined time; and an occurrence ratio arousal degree for evaluating the arousal degree of the driver based on the occurrence ratio of the long blink. A wakefulness estimating device comprising: a determination unit. 2. The method according to claim 1, wherein the arousal ratio arousal level determination unit determines a drowsiness predicted value obtained based on a regression analysis of a prediction formula based on a drowsiness prediction model indicating a relationship between the arousal level and a long blink occurrence ratio with a predetermined threshold value. The arousal level estimating device according to claim 1, wherein the arousal level is separately evaluated. 3. The awakening degree estimating device according to claim 1, wherein: a blinking average time calculating means for calculating an average of the blinking times within a predetermined time; and a driver's awakening degree based on the blinking average time. Average time arousal degree determination means to be evaluated, arousal degree decrease detection means for comprehensively determining each determination result of the average time arousal degree determination means and the occurrence ratio arousal degree determination means and detecting the driver's arousal degree decrease, A wakefulness estimation device characterized by comprising: 4. The average time arousal level determination means discriminates a drowsiness predicted value obtained based on a regression analysis of a prediction formula based on a drowsiness prediction model indicating a relationship between the arousal level and the average time of blinking with a predetermined threshold value. The arousal level estimation device according to claim 3, wherein the arousal level is evaluated.