JP4910547B2 - Sleepiness determination apparatus and sleepiness determination program - Google Patents

Description

  The present invention relates to a drowsiness determination device and a drowsiness determination program for determining drowsiness with high accuracy.

In order to make a driver of a vehicle perform safe driving, an apparatus for determining a driver's sleepiness has been developed. As a drowsiness determination device, for example, there is a device that extracts drowsiness by extracting a feature amount that changes according to drowsiness from heartbeat, brain waves, blinks, and the like, and compares the feature amount with a threshold value. In the apparatus shown in Patent Document 1, fluctuation (change) is extracted from a physiological index of a subject, a characteristic change of fluctuation of the physiological index is extracted, and a significant characteristic in which the occurrence frequency of the characteristic change exceeds a threshold value is extracted. Sleepiness is determined based on the change.
Japanese Patent Application No. 2005-188656

  In some cases, such as long-distance driving or traffic jams, the driver may drive while taking sleepiness. When you are enduring drowsiness, the drowsiness becomes deeper, or the sleepiness peak changes repeatedly as the drowsiness deepens, or when you awaken from drowsiness with a stimulus such as a curve road or relapse. (Hunting) may occur. Thus, when the patient is drowsy, the physical condition such as heart rate and brain wave may be affected, and the heart rate and brain wave may change even at the same level of sleepiness at the peak. That is, even when the peak of drowsiness of the same strength is repeated, the heartbeat and brain wave at the peak change. For this reason, if sleepiness is determined with a uniform threshold as in the conventional sleepiness determination device, it may not be possible to determine that there is sleepiness with the same strength, and sleepiness determination accuracy may be reduced.

  Therefore, an object of the present invention is to provide a drowsiness determination device and a drowsiness determination program for determining drowsiness with high accuracy.

A drowsiness determination device according to the present invention is a drowsiness determination device that determines drowsiness of a subject, a feature amount extraction unit that extracts a feature amount that changes according to the sleepiness of the subject, and a feature amount that is extracted by the feature amount extraction unit The sleepiness determination means for determining the sleepiness of the subject based on the memory and the storage means for storing the sleepiness determination history of the subject based on the determination result of the sleepiness determination means, the sleepiness determination means includes the sleepiness stored in the storage means When it is determined that there is drowsiness in the past based on the determination history, it is easier to determine that there is drowsiness compared to the case where it is not determined that there is drowsiness.

  In this drowsiness determination device, a feature amount that changes in accordance with the subject's sleepiness is extracted by the feature amount extraction means. Feature quantities that change according to sleepiness include, for example, feature quantities extracted from heartbeats, feature quantities extracted from brain waves, feature quantities extracted from breathing, feature quantities extracted from blink, feature quantities extracted from lip movement, This is a feature amount extracted from the movement of the body (back stretch, body twist, etc.). In the sleepiness determination apparatus, the sleepiness determination unit determines the sleepiness of the subject based on the feature amount, and the past sleepiness determination history of the subject is stored in the storage unit based on the sleepiness determination result. If the subject has drowsiness and the intensity of drowsiness has changed due to repeated drowsiness peaks, the subject will accumulate fatigue due to drowsiness, and the accumulated fatigue may be related to the physical condition of the subject (for example, heart rate , Brain waves, movement of each part of the body). It is thought that this accumulated fatigue will remain unless the body is rested and the physical condition will not recover. Therefore, even when a drowsiness peak has occurred in the past and a drowsiness peak of the same strength has occurred again, the feature amount extracted from the physical condition of the subject is predicted to change from the previous peak. Therefore, in the drowsiness determination device, when the drowsiness determination means performs the drowsiness determination, it is determined whether or not drowsiness has been determined in the past from the drowsiness determination history, and is not determined when it is determined that there is drowsiness. This makes it easier to determine that there is drowsiness. In this way, the drowsiness determination device makes it easy to determine drowsiness after determining drowsiness, so that the second time even in a situation where the intensity of drowsiness is changing (when the peak of drowsiness is repeated), The drowsiness at the third peak can also be reliably determined, and the drowsiness determination accuracy is improved.

  In the sleepiness determination apparatus of the present invention, the sleepiness determination means may be configured to relax the sleepiness determination threshold when it is determined that there is sleepiness in the past based on the sleepiness determination history stored in the storage means.

  In this drowsiness determination device, the drowsiness determination means determines whether or not there is drowsiness by comparing the feature amount with a drowsiness determination threshold. When making this determination, the drowsiness determination means relaxes the drowsiness determination threshold when the sleepiness determination history has been determined to have drowsiness in the past, compared to the case where drowsiness has not been determined. Make it easier to determine that there is. As described above, in the drowsiness determination device, the drowsiness determination criterion is changed by a very simple method of changing the drowsiness determination threshold.

  In the drowsiness determination device of the present invention, it is preferable that the drowsiness determination history is a history from the start of drowsiness determination to the subject until the present day.

  In this drowsiness determination device, the drowsiness determination history is stored in the storage means based on the determination results from the start of drowsiness determination to the subject until the present. As described above, when the test subject endures sleepiness and fatigue due to endurance of sleepiness accumulates, the physical condition of the test subject changes, and the characteristic amount changes with respect to the peak of sleepiness of the same strength. However, if you take a rest and get rid of your fatigue, you will recover. Therefore, if the drowsiness determination is finished once and drowsiness determination is started again, even if the patient dwells at the time of the previous drowsiness determination and fatigue has accumulated, the fatigue is removed while the determination is not being made. Therefore, there is no need to reflect the result of the previous sleepiness determination in the current sleepiness determination. On the other hand, if the drowsiness determination has been started and continues, if drowsiness occurs after the drowsiness determination starts (once it is determined that there is drowsiness), there is a possibility that the patient may endure drowsiness and accumulate fatigue Therefore, it is necessary to reflect the determination result after the start of the determination in the current sleepiness determination. Therefore, only the determination results from the start of sleepiness determination until the present time are reflected in the sleepiness determination history.

A drowsiness determination program according to the present invention is a drowsiness determination program for determining a subject's drowsiness, and the computer extracts a feature amount extraction function that extracts a feature amount that changes according to the subject's sleepiness, and a feature amount extraction function If the subject's sleepiness is determined based on the feature amount extracted in step, and if it is determined that there is sleepiness in the past based on the sleepiness determination history stored in the storage means when performing the determination, it is determined that there is sleepiness A sleepiness determination function that makes it easier to determine that there is drowsiness compared to a case where the sleepiness is not performed is realized. According to the drowsiness determination program, the computer can execute the program to achieve the operation and effect of the drowsiness determination device.

  The present invention makes it easy to determine that there is drowsiness after determining that there is drowsiness, so that drowsiness can be reliably determined even in a situation where the intensity of drowsiness is changing, and drowsiness determination accuracy is improved.

  Hereinafter, embodiments of a drowsiness determination device and a drowsiness determination program according to the present invention will be described with reference to the drawings.

  In the present embodiment, the present invention is applied to a dozing detection device that is mounted on a vehicle and detects a driver's dozing. The dozing detection device according to the present invention measures an index representing the physical condition of the driver, detects a dozing based on a feature amount extracted from the index, and if the dozing state is detected, the driver is a dozing state Announce that.

  With reference to FIGS. 1-3, the dozing detection apparatus which concerns on this Embodiment is demonstrated. FIG. 1 is a configuration diagram of a snoozing detection apparatus according to the present embodiment. FIG. 2 is a graph showing the relationship between sleepiness intensity, a feature amount for dozing determination, and a threshold for dozing determination. FIG. 3 is a graph showing a relationship between a heartbeat feature amount and a threshold for dozing determination when a heartbeat is used as an example of dozing detection.

  The dozing detection device 1 compares the feature amount that changes according to the driver's sleepiness with a threshold for dozing determination, and determines that the feature amount is larger than the threshold value, so that the dozing state is determined. In particular, in the snoozing detection apparatus 1, in order to determine a dozing state with high accuracy even in a hunting situation in which a peak of sleepiness occurs repeatedly, the threshold for dozing detection is made variable. For this purpose, the dozing detection apparatus 1 includes an index measurement unit 2, an output unit 3, and an ECU [Electronic Control Unit] 4. The dozing occurrence history storage buffer 10, a feature amount extraction unit 11, a dozing occurrence history determination unit 12, and a doze are included in the ECU 4. A detection unit 13, a hunting support unit 14, a dozing detection presence / absence determination unit 15, and a dozing output unit 16 are configured.

  In the present embodiment, the dozing occurrence history storage buffer 10 corresponds to the storage means described in the claims, the feature quantity extraction unit 11 corresponds to the feature quantity extraction means described in the claims, and the dozing occurrence history. The determination unit 12, the dozing detection unit 13, and the hunting support unit 14 correspond to drowsiness determination means described in the claims.

  The index measuring means 2 is a means for measuring various indices representing the physical condition for judging the driver's sleepiness. The index measuring means 2 measures each index and transmits an index signal indicating the measured index to the ECU 4. The index measuring means 2 is, for example, a heartbeat sensor, an electroencephalogram sensor, a pulse wave sensor, a respiration sensor, a camera that images the driver's face or body, and an image processing device that processes the captured image. The means using the camera detects the blink of the driver's face, the movement of the lips, etc. from the captured image, or detects the driver's body stretching, twisting of the body, and the like.

  The output means 3 is a means for notifying the output target that the driver is asleep or prompting the driver to rest. When the output means 3 receives an output signal from the ECU 4, the output means 3 outputs according to each means. As the output means 3, for example, means for notifying by sound (buzzer, audio, radio, horn), means for notifying by light (meter lighting, room lighting), means for notifying by tactile sense or hot / cold sense (embedded in a sheet) Vibration device, air conditioner wind and temperature changes), smell notification means (fragrance injection), command output to the system. Examples of the output target include a driver, an occupant sitting outside the driver's seat, a manager who manages the operation of a business vehicle such as a truck or a taxi, and a vehicle control system.

  The ECU 4 includes a CPU [Central Processing Unit], a ROM [Read Only Memory], a RAM [Random Access Memory], and the like, and comprehensively controls the dozing detection device 1. When the ECU 4 is activated, it takes an index signal from the index measuring means 2 at regular time intervals, stores the index signal index in each buffer, and holds it as time-series data at regular time intervals. Moreover, in ECU4, each part 11-16 is comprised by running each program stored in ROM by CPU, and the process of each part 11-16 is performed for every fixed time.

The dozing occurrence history storage buffer 10 is configured in a predetermined area of the RAM, and is a buffer for storing a dozing occurrence history. Dozing occurrence history is a history of data when doze detecting device 1 detects a doze state until now since the start time and characteristic amount when the detected threshold TH ₂ beyond have time etc. It consists of data.

The feature quantity extraction unit 11 extracts a feature quantity for determining a doze state from each index measured by the index measuring means 2. The extracted feature amount is a parameter that changes according to sleepiness, and is set so that the value increases as sleepiness increases. Further, the feature extraction unit 11 obtains the basic value of the threshold TH ₂ for determining the dozing state from the feature of each indicator.

Here, a case will be described in which feature amounts are extracted from heartbeats measured by a heartbeat sensor as an example of extracting feature amounts. First, the feature amount extraction unit 11 performs a band pass filter process on the time series data of the heartbeat, and extracts a component of a predetermined pass band (for example, 0.1 Hz to 30 Hz) from the time series data of the heartbeat. Next, the feature amount extraction unit 11 cuts out a waveform portion that is equal to or greater than the threshold value TH ₀ for heartbeat timing detection from the heartbeat time-series data. Then, the feature quantity extraction unit 11 binarizes with the timing when the cut-out waveform portion is maximized as 1 and the other timing as 0. Next, the feature amount extraction unit 11 obtains the time from each heartbeat timing t _{i that} is binarized 1 to the next heartbeat timing t _{i + 1} that becomes 1, and sets the time interval (t _{i + 1} −t _i ) for each time interval. It is applied to the heart rate timing _{t i.} Time information given to the each heartbeat timing t _i becomes the information of the cardiac cycle. Next, the feature amount extraction unit 11 obtains a heartbeat period curve by complementing the heartbeat period information, and obtains time-series data of the heartbeat period.

  Next, the feature quantity extraction unit 11 performs a fast Fourier transform process on the time-series data of the heartbeat period in the analysis unit interval width Tterm before the reference time T that is an arbitrary time stamp. Next, the feature amount extraction unit 11 performs an integration process on the amplitude power spectrum for each preset frequency band in the power spectrum obtained for each analysis unit interval width Tterm by fast Fourier transform. The frequency band may be a frequency band in which fluctuations (changes) appear strongly with respect to the measured heartbeat. Then, the feature amount extraction unit 11 performs fast Fourier transform processing on the time-series data of the heartbeat period in the analysis unit section width Tterm every time a fixed time elapses and reaches the reference time T, thereby obtaining the analysis unit section width Tterm. Every time, the integration processing is performed on the amplitude power spectrum. The time series data of the amplitude spectrum power obtained here is the time series data of heartbeat fluctuation.

Next, the feature amount extraction unit 11 performs a differentiation process on the time series data of heartbeat fluctuation. Next, the feature extracting unit 11 sets an analysis section width A to terminate td before time from the current time t _0. Then, the feature extracting unit 11 calculates an average value mean and the standard deviation sd of the time-series data of the differential value of the heartbeat fluctuation of the analysis section width A, and calculates the threshold value TH ₁ by the equation (1).

By setting the threshold TH ₁ to a value that is three times the standard deviation sd from the mean value mean, it is unlikely that the heartbeat fluctuation differential value exceeds the threshold TH ₁ and there is a statistically significant difference. A characteristic change of heartbeat fluctuation can be detected.

Next, the feature amount extraction unit 11 determines whether the differential value of the heartbeat fluctuation exceeds a threshold value TH ₁ for each time. Constant time prior to the current time t ₀ in the past, if the differential value of the heartbeat fluctuation does not exceed the threshold value TH _1, the feature amount extraction unit 11, and updates the threshold value TH ₁ at the next time t _{0 + 1.} On the other hand, a certain time prior to the current time t ₀ in the past, if the differential value of the heartbeat fluctuation exceeds the threshold value TH _1, the feature extracting unit 11, without updating the threshold TH ₁ at the next time t _{0 + 1} In addition, the threshold value TH ₁ at the current time t ₀ is maintained as the threshold value TH ₁ . This prevents the use of the differential value of the heartbeat fluctuation exceeding the threshold value TH ₁ to the threshold setting, it becomes possible to accurately determine the change in the driver drowsiness.

Next, the feature extracting unit 11, for each frequency band, determines whether the differential value of the heartbeat fluctuation exceeds a threshold value TH _1. Then, the feature extracting unit 11, a predetermined time interval, counting the characteristic change of the heartbeat fluctuation exceeding the threshold value TH _1, obtaining the time-series data of the above threshold density. The time-series data of the density exceeding the threshold is the occurrence frequency of the characteristic change of the heartbeat fluctuation, and is the final feature amount for determining the dozing state.

Next, the feature amount extraction unit 11 extracts time-series data of the specific section A in the first half of the time-series data from the time-series data having the density exceeding the threshold. Then, the feature amount extraction unit 11 extracts the above threshold density maximum value MM from time-series data in a specific section A, calculates a threshold value TH ₂ by the equation (2). Threshold TH ₂ obtained by the formula (2) is a basic value of the threshold value for determining the dozing state.

  scale is an arbitrary real number, param is an arbitrary variable (example 1: MM, example 2: standard deviation of density exceeding threshold in specific interval A), and the values of scale and param are statistically significant fluctuations Adjustments may be made so that a characteristic change is detected.

  The dozing occurrence history determination unit 12 determines, based on the dozing occurrence history, whether or not a dozing state has been detected since the start of the dozing detection device 1. If there is no dozing state detection data in the dozing occurrence history, the dozing occurrence history determination unit 12 causes the dozing detection unit 13 to execute the process without executing the processing in the hunting support unit 14. On the other hand, when there is data of dozing state detection in the dozing occurrence history, the dozing occurrence history determination unit 12 causes the hunting handling unit 14 to execute the process and then causes the dozing detection unit 13 to execute the process.

Doze detecting section 13 determines whether the feature quantity of each indicator extracted by the feature amount extraction unit 11 exceeds the threshold value TH _2. As the threshold TH ₂ , when the processing in the hunting handling unit 14 is being executed, the threshold TH ₂ (correction value) obtained by the hunting handling unit 14 is used to execute the processing in the hunting handling unit 14. If not, the threshold TH ₂ (basic value) obtained by the feature quantity extraction unit 11 is used. When the characteristic amount does not exceed the threshold value TH _2, doze detecting section 13 determines that not the doze state. When the characteristic amount exceeds the threshold value TH _2, doze detecting section 13 measures the time that has exceeded the threshold value TH _2, if the measured time exceeds the time threshold TH ₃ determines that the dozing state.

Here, as an example of detecting the dozing state, a case where the dozing state is detected using the above-described heartbeat feature amount (time-series data of density exceeding the threshold) will be described. The dozing detection unit 13 extracts time series data after the specific section A in the first half of the time series data from the time series data of the density exceeding the threshold value of the heart rate obtained by the feature quantity extraction unit 11. Then, in the doze detecting section 13 determines whether or exceeds the threshold value TH ₂ at the time series data after specific section A. If it does not exceed the threshold value TH _2, the doze detecting section 13 determines that not doze state. On the other hand, if the threshold is exceeded TH _2, the doze detecting section 13 measures the time that has exceeded the threshold value TH _2. Data exceeds this threshold TH ₂ is a characteristic change in the statistically a significant difference fluctuation indicates that drowsiness is strong. Then, it is determined whether the doze detecting section 13, the time exceeds the threshold value TH ₂ exceeds the time threshold TH _3. If it does not exceed the time threshold TH _3, the doze detecting section 13 determines that not doze state. On the other hand, if it exceeds the time threshold value TH _3, the doze detecting section 13, the driver determines that doze state. The time exceeding the threshold value TH _2, the may be the time that has exceeded continuously, or may be integrated time of time that exceeds discretely within a predetermined time.

The hunting handling unit 14 subtracts the threshold correction amount B from the basic value of the threshold TH ₂ obtained by the feature amount extraction unit 11, and sets the subtracted value as the correction value of the threshold TH ₂ . The threshold correction amount B is set in advance from the relationship between the change in the intensity of sleepiness and the change in the magnitude of the feature amount of each index, based on experiments with a plurality of subjects. Thus, when detecting a doze state after starting the doze detecting device 1, in order to correspond to hunting situation repeating or weakened or growing drowsiness, by reducing the threshold value TH _2, the detection of the first time of dozing state Make it easy to detect a dozing state.

Here, the case of obtaining the threshold value TH ₂ by using the feature amount of the heartbeat described above as an example of Hainchingu corresponding. The hunting corresponding section 14, by using the threshold value correction amount B set for above threshold density maximum value MM and heart in a specific section A, calculates a threshold value TH ₂ by the equation (3). Threshold TH ₂ obtained by the formula (3) is a correction value of the threshold value for determining the dozing state.

Here, the threshold value TH ₂ (correction value) is obtained by Expression (3), but the threshold value correction amount B may be subtracted from the threshold value TH ₂ (basic value) obtained by Expression (2).

  The dozing detection presence / absence determination unit 15 determines whether the dozing detection unit 13 has detected a dozing state. When the dozing state is not detected by the dozing detection unit 13, the dozing detection presence / absence determination unit 15 does not execute the process in the dozing output unit 16. On the other hand, when the dozing state is detected by the dozing detection unit 13, the dozing detection presence / absence determination unit 15 causes the dozing output unit 16 to execute processing.

The dozing output unit 16 transmits an output signal to the output means 3 in order to notify that it is in the dozing state. Further, the dozing output unit 16 stores the dozing state detection data (detection time, time exceeding the threshold TH ₂ , etc.) in the dozing occurrence history storage buffer 10.

FIG. 2 shows temporal changes in sleepiness intensity F (broken line), feature amount C (solid line), and threshold value TH ₂ (dashed line). The sleepiness intensity F is a sleepiness level determined by sensory evaluation of sleepiness, and is expressed in 6 stages from D0 to D5, where D0 is the weakest (wake state) and D5 is the strongest. The feature amount C is a feature amount that increases when sleepiness is strong. When the sleepiness intensity F reaches the first peak F1, the feature amount C exceeds the threshold value TH ₂ (basic value) for a predetermined time, and therefore a dozing state is detected. Thereafter, the drowsy state is detected, the threshold value TH ₂ is changed to a smaller value from the basic value (correction value). Therefore, when the sleepiness intensity F reaches the second peak F2, the feature amount C is smaller than that at the first peak F1, but the feature amount C is set to the threshold value TH ₂ (correction value) for a predetermined time. Since it exceeds, a dozing state is detected.

When the driver is drowsy, the peak of drowsiness is repeated, and the drowsiness intensity changes (hunting). The intensity of this drowsiness is almost unchanged at each peak. However, the feature amount that changes in response to sleepiness is smaller at the second sleepiness peak than at the first sleepiness peak. This is thought to be due to the fact that the patient has drowsiness, so that fatigue has accumulated and the physical condition (such as heartbeat activity) has fallen, resulting in a decrease in the feature amount. Therefore, maintaining the threshold TH ₂ remains basic value, in the second and subsequent peaks of sleepiness, because drowsiness strong despite characteristic amount does not reach the threshold value TH _2, can not be detected doze state. However, the threshold value TH ₂ by lowering from the basic value, even when the second and subsequent peaks of sleepiness, beyond feature amount threshold TH _2, it is possible to detect the doze state.

By the way, when the driver is drowsy, if he takes enough rest and rests, he will recover from accumulated fatigue and return to physical condition. Thereafter, when restarting the operation, since the doze detection from physical condition has returned status, also the threshold value TH ₂ as a basic value, it is possible to detect the doze state. Therefore, in the dozing occurrence history, the previous data is erased when the dozing detection device 1 is activated, and only data after activation is stored.

FIG. 3 shows changes over time in sleepiness intensity F (dashed line), time-series data C (solid line) of threshold excess density as a feature quantity of heartbeat, and threshold TH ₂ (dashed line). Basic value of the threshold TH ₂ is calculated by using the data of the first specific section A of the time series data C. Then, the time series data C after a certain interval A is compared to the base value of the threshold TH _2, since the time-series data C is larger than the threshold value TH 2 _(basic value) discretely predetermined time or more, and dozing state detecting Is done. Thereafter, the drowsy state is detected, the threshold value TH ₂ is changed to a smaller value from the basic value (correction value). The density exceeding the threshold is smaller in the second half of the time series data C than in the first half, but the threshold TH ₂ (correction value) is also smaller, and the threshold TH ₂ is also set in the second half of the time series data C. Since it exceeds the predetermined time discretely, a dozing state is detected.

Here, the target level detected as the sleepiness state is the D2 level of sleepiness intensity. Therefore, unless a smaller threshold TH _2, the second part of the time series data C can not detect the D2 or stronger drowsiness. However, it is possible to detect by reducing the threshold value TH ₂ after detecting a doze state, the D2 in the strength of sleepiness in the latter part of the time series data C.

  With reference to FIG. 1, operation | movement of the dozing detection apparatus 1 is demonstrated. In particular, the processing in the ECU 4 will be described with reference to the flowchart in FIG. FIG. 4 is a flowchart showing the flow of processing in the ECU of FIG.

  When the driver starts the engine, the dozing detection device 1 is activated. In the dozing detection device 1, no data is stored in the dozing occurrence history at the time of activation. And the dozing detection apparatus 1 repeats the following operations after activation.

The index measuring means 2 measures an index from the driver and transmits the measured value to the ECU 4 as an index signal. The ECU 4 receives the index signal and stores the index data in a predetermined buffer (S1). Thereby, time-series data of the index is generated. Then, the ECU 4, extracts the feature amount from the time series data of the index to determine the basic values of the threshold TH ₂ for determining the dozing state from the feature amount (S2).

The ECU 4 determines whether or not a dozing state is detected after activation based on the dozing occurrence history (S3). If it is determined that detected a doze state after activation at S3, the ECU 4, in order to cope with the hunting of drowsiness, obtaining a correction value of the threshold TH ₂ by subtracting the threshold value correction amount B from the basic value of the threshold TH ₂ (S4).

When executing the processing in the case or S4 is determined that detects dozing state after activation at S3, it is determined whether or not the ECU 4, the feature amount exceeds a threshold value TH ₂ (S5). If that is performing the processing of S4 (case with detection dozing state), a correction value as a threshold value TH ₂ (low value), the detection level of the dozing state is relaxed than usual. (No dozing state detection) when not running the process in S4, a basic value as a threshold value TH ₂ (large value), the detection level of the dozing state is normal level. When the characteristic amount does not exceed the threshold value TH _2, ECU 4 determines that it is not the doze state (S5). If the feature amount exceeds the threshold value TH _2, the ECU 4, when the time exceeds the threshold value TH ₂ exceeds the time threshold TH ₃ is determined to dozing state detection, does not exceed the time threshold TH ₃ In this case, it is determined that the device is not in a dozing state (S5).

  The ECU 4 determines whether or not a dozing state is detected in the process at S5 (S6). If it is determined in S6 that no dozing state is detected, the ECU 4 ends the current process and performs the next process after a predetermined time has elapsed. When it is determined in S6 that the dozing state is detected, the ECU 4 transmits an output signal to the output means 3 (S7). When the output signal is received, the output means 3 performs an output for informing that the driver is in a dozing state. By this output, the driver notices that the driver is dozing, or a person other than the driver knows that the driver is dozing and alerts the driver. As a result, the sleepiness of the driver is weakened or the driver takes a rest. Further, the ECU 4 buffers the dozing state detection data in the dozing occurrence history, ends the current process (S8), and performs the next process after a predetermined time has elapsed.

According to this dozing detection device 1, by making it easier to detect the dozing state after the detection of the dozing state, it is possible to reliably determine the dozing state after the first time even in the situation where the strength of drowsiness is hunting, The drowsiness determination accuracy is improved. Further, according to the doze detecting device 1, the simple way simply by changing the threshold value TH ₂ for determination dozing, it is possible to easily detect a doze state than normal.

  As mentioned above, although embodiment which concerns on this invention was described, this invention is implemented in various forms, without being limited to the said embodiment.

  For example, in this embodiment, the present invention is applied to a drowsiness detection device that determines the drowsiness of a vehicle driver. However, the drowsiness of various people such as other vehicle drivers, various plant supervisors, and night workers is determined. May be used to

  In the present embodiment, the present invention is applied to a device for determining doze. However, the present invention is applied to a program stored in a storage medium such as a CD-ROM or a program that can be used via a network such as the Internet. It is good also as a structure which determines a dozing by running a program on a computer.

  In the present embodiment, an example of determining a dozing state using a feature amount of a heartbeat has been described. However, it changes according to sleepiness required from brain waves, pulse waves, breathing, blinking, face and body actions, and the like. A configuration may be used in which the feature amount is used to determine the dozing state, or a combination of some of these feature amounts may be used to determine the dozing state. In addition, a method other than the method shown in the example can be applied to the method for obtaining the heartbeat feature amount.

  Also, in this embodiment, the dozing state detection data from the start of the dozing detection device to the present time is stored as the dozing occurrence history, but in a short time from the stop of the dozing detection device to the restart (In the case of a short period of time when the state of drowsiness has not recovered from the previous operation of the drowsiness detection device), the drowsiness detection data during the operation of the previous drowsiness detection device is stored as the drowsiness occurrence history You may make it do.

In the present embodiment, the threshold TH ₂ is reduced by one step as the hunting handling process. However, the threshold TH ₂ may be reduced stepwise in accordance with the number of times the dozing state is detected. .

Also, except in this embodiment has a configuration that facilitates detecting a doze state by changing the threshold value TH ₂ for the feature quantity, which changes the threshold value TH ₂ for the feature quantity as a way to easily detect a doze state also is applicable in the process, for example, may be easily detect a doze state by changing the time threshold TH ₃ versus time characteristic amount exceeds the threshold value TH _2, when determining the characteristic quantity You may make it obtain | require as a feature-value which is easily detected as a dozing state.

In the present embodiment, the threshold TH ₂ for determining the dozing state is obtained from the feature amount. However, the threshold TH ₂ may be set in advance.

It is a block diagram of the dozing detection apparatus which concerns on this Embodiment. It is a graph which shows the relationship between the drowsiness intensity, the characteristic amount for dozing determination, and the threshold for dozing determination. It is a graph which shows the relationship between the feature-value of the heartbeat at the time of using a heartbeat as an example of dozing detection, and the threshold value for dozing determination. It is a flowchart which shows the flow of the process in ECU of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Dozing detection apparatus, 2 ... Index measurement means, 3 ... Output means, 4 ... ECU, 10 ... Dozing occurrence history storage buffer, 11 ... Feature quantity extraction part, 12 ... Dozing occurrence history judgment part, 13 ... Dozing detection part, 14 ... hunting support unit, 15 ... dozing detection presence determination unit, 16 ... dozing output unit

Claims (4)

A drowsiness determination device for determining drowsiness of a subject,
A feature amount extraction means for extracting a feature amount that changes according to the subject's sleepiness;
Drowsiness determination means for determining drowsiness of the subject based on the feature amount extracted by the feature amount extraction means;
Storage means for storing the sleepiness determination history of the subject based on the determination result of the sleepiness determination means,
The drowsiness determining means determines that there is drowsiness compared to the case where it is not determined that there is drowsiness when it has been determined that there is drowsiness in the past based on the drowsiness determination history stored in the storage means. A drowsiness determination device characterized by facilitating.   The sleepiness determination unit according to claim 1, wherein the sleepiness determination unit relaxes a sleepiness determination threshold when it is determined that there is sleepiness in the past based on a sleepiness determination history stored in the storage unit. apparatus.   The sleepiness determination apparatus according to claim 1 or 2, wherein the sleepiness determination history is a history from the start of sleepiness determination to a subject until the present day. A sleepiness determination program for determining a subject's sleepiness,
On the computer,
A feature amount extraction function for extracting feature amounts that change according to the subject's sleepiness;
When the subject's sleepiness is determined based on the feature amount extracted by the feature amount extraction function , and when it is determined that there is sleepiness in the past based on the sleepiness determination history stored in the storage means when performing the determination Is a drowsiness determination program characterized by realizing a drowsiness determination function that makes it easier to determine that there is drowsiness compared to a case where it is not determined that drowsiness is present.

Images