WO2008069337A1 - 覚醒度判定装置及び覚醒度判定方法 - Google Patents
覚醒度判定装置及び覚醒度判定方法 Download PDFInfo
- Publication number
- WO2008069337A1 WO2008069337A1 PCT/JP2007/073742 JP2007073742W WO2008069337A1 WO 2008069337 A1 WO2008069337 A1 WO 2008069337A1 JP 2007073742 W JP2007073742 W JP 2007073742W WO 2008069337 A1 WO2008069337 A1 WO 2008069337A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- frequency component
- heartbeat
- arousal level
- stimulus
- timing
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/06—Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/347—Detecting the frequency distribution of signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4029—Detecting, measuring or recording for evaluating the nervous system for evaluating the peripheral nervous systems
- A61B5/4035—Evaluating the autonomic nervous system
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4809—Sleep detection, i.e. determining whether a subject is asleep or not
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
- A61B5/352—Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60K—ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES; ARRANGEMENT OR MOUNTING OF PLURAL DIVERSE PRIME-MOVERS IN VEHICLES; AUXILIARY DRIVES FOR VEHICLES; INSTRUMENTATION OR DASHBOARDS FOR VEHICLES; ARRANGEMENTS IN CONNECTION WITH COOLING, AIR INTAKE, GAS EXHAUST OR FUEL SUPPLY OF PROPULSION UNITS IN VEHICLES
- B60K28/00—Safety devices for propulsion-unit control, specially adapted for, or arranged in, vehicles, e.g. preventing fuel supply or ignition in the event of potentially dangerous conditions
- B60K28/02—Safety devices for propulsion-unit control, specially adapted for, or arranged in, vehicles, e.g. preventing fuel supply or ignition in the event of potentially dangerous conditions responsive to conditions relating to the driver
- B60K28/06—Safety devices for propulsion-unit control, specially adapted for, or arranged in, vehicles, e.g. preventing fuel supply or ignition in the event of potentially dangerous conditions responsive to conditions relating to the driver responsive to incapacity of driver
Definitions
- the present invention relates to an arousal level determination device and an arousal level determination method.
- a technique for detecting a person's arousal level has been devised.
- a human eyeball is imaged, and the pupil region of the eyeball is extracted from the captured image, and the blinking time and frequency are equal to or greater than a predetermined value based on the extracted pupil shape change.
- Japanese Patent Application Laid-Open No. 6-27071 1 discloses a technique for determining that the degree of arousal is reduced when it is recognized that this is the case.
- the state of wakefulness is judged from the time and frequency of blinking. Therefore, if the degree of wakefulness does not drop significantly (ie, if the sleepiness is not very strong), the degree of wakefulness is detected. Can not do it. Therefore, a technology that can detect weak sleepiness of active people has been desired.
- the present invention has been made to solve the above-described problems, and an object thereof is to provide a wakefulness determination device and a wakefulness determination method that can more reliably detect weak sleepiness of an active person. To do.
- the arousal level determination device includes a signal acquisition means for acquiring a biological signal of an active person and a physiological index indicating the strength of a state against sleepiness from the biological signal acquired by the signal acquisition means. It is characterized by comprising: an index acquisition means for acquiring; and a determination means for determining a person's arousal level based on the physiological index acquired by the index acquisition means.
- the method for determining the arousal level according to this work includes a signal acquisition step for acquiring a biological signal of an active person, and a state against drowsiness from the biological signal acquired in the signal acquisition step. Written in that it includes an index step for acquiring a physiological index indicating the strength of the W state, and a determination step for determining a person's arousal level based on the physiological index acquired in the index acquisition step.
- arousal level determination device or the arousal level determination method According to the arousal level determination device or the arousal level determination method according to the present invention, a physiological index indicating the strength of a state against drowsiness is acquired from a biological signal of an active person, and the arousal level is based on the physiological index. Therefore, weak sleepiness (decreased arousal level) of an active person can be detected more reliably.
- the physiological index having a correlation with sympathetic nerve activity it is preferable to use the physiological index having a correlation with sympathetic nerve activity.
- the state that resists sleepiness is a state in which an active person is stressed, and as a result, the activity of the sympathetic nervous system is activated in such a state.
- the wakefulness determination device or the wakefulness determination method according to the present invention since a physiological index for determining the wakefulness is correlated with sympathetic nerve activity, the state of sympathetic nerve activity, that is, against sleepiness It is possible to more reliably detect the state of being.
- a heartbeat signal as the biological signal and an amplitude spectrum power of a heartbeat fluctuation low frequency component acquired from the heartbeat signal as the physiological index.
- the amplitude spectrum of the heartbeat fluctuation low-frequency component obtained from the heartbeat signal is correlated with the sympathetic nerve activity
- the amplitude spectrum power of the heartbeat fluctuation low-frequency component is used as a physiological index to determine arousal level. It is possible to properly detect the state of nerve activity, that is, the state of being resistant to sleepiness.
- the determination means includes a heartbeat fluctuation low frequency component. It is preferable to determine that the arousal level is lowered when the amplitude spectrum power is larger than a predetermined value.
- the arousal level determination method in the determination step, when the amplitude spectral power of the heartbeat fluctuation low frequency component is larger than a predetermined value, it is determined that the arousal level is decreased. Is preferred.
- the amplitude spectrum power of the heartbeat fluctuation low frequency component correlates with the sympathetic nerve activity, and the magnitude of the amplitude spectrum power is high enough to resist drowsiness. There is a correlation. Therefore, as the degree of arousal decreases and the degree of resistance to sleepiness increases, the sympathetic nerve activity increases and the amplitude spectrum power of the heartbeat fluctuation low frequency component increases. According to the wakefulness determination device or the wakefulness determination method according to the present invention, when the amplitude spectral power of the heartbeat fluctuation low frequency component is larger than the predetermined value, it is determined that the wakefulness is lowered. It is possible to reliably detect a state in which the arousal level is lowered while preventing false detection.
- the determination means may determine that the wakefulness level is lower than when the amplitude spectrum power of the heartbeat fluctuation low frequency component is large, compared to a case where the amplitude spectral power is small. preferable.
- the wakefulness level determination method in the above determination step, when the amplitude spectrum power of the heartbeat fluctuation low frequency component is large, the wakefulness level is lower than when the amplitude spectrum power is small. Is preferably determined.
- the sympathetic nerve activity is activated and the amplitude spectral power of the heartbeat fluctuation low-frequency component increases.
- the arousal level determination device or the arousal level determination method according to the present invention it is determined that the arousal level is further lowered when the amplitude spectral power of the heartbeat fluctuation low frequency component is large as compared with the case where the amplitude spectral power is small. Therefore, it is possible to determine the magnitude of the decrease in arousal level.
- the arousal level determination device is provided with a stimulus that gives a stimulus to a person to increase the arousal level. And a timing setting means for setting a timing for giving the stimulus, and the timing giving means sets the timing for giving the stimulus based on a physiological index indicating the strength of the state against drowsiness. preferable.
- the arousal level determination method further includes a stimulus applying step for giving a stimulus for raising a person's arousal level, and a timing setting step for setting a timing for giving the stimulus.
- the timing giving step It is preferable to set the timing for applying the stimulus based on a physiological index indicating the strength of the state to resist. It is preferable that the appropriate timing for giving a stimulus for increasing the degree of arousal coincides with or slightly before the timing at which a person resisting sleepiness wants a stimulus.
- the timing at which a person who resists sleepiness wants stimulation correlates with the strength of the state against sleepiness.
- the timing for applying a stimulus is set based on a physiological index indicating the strength of a state against drowsiness. It becomes possible.
- the arousal level determination device further includes a stimulus applying unit that gives a stimulus for raising a person's arousal level, and a timing setting unit that sets a timing to give the stimulus, and the timing giving unit reduces heart rate fluctuation. It is preferable to set a timing to give a stimulus during a predetermined time after the amplitude spectrum power of the frequency component becomes larger than a predetermined value.
- the arousal level determination method further includes a stimulus applying step for giving a stimulus for raising a person's arousal level, and a timing setting step for setting a timing for giving the stimulus.
- a stimulus applying step for giving a stimulus for raising a person's arousal level
- a timing setting step for setting a timing for giving the stimulus.
- heart rate fluctuation It is preferable to set the timing for applying the stimulus from when the amplitude spectral power of the low frequency component becomes larger than a predetermined value until the predetermined time elapses.
- the arousal level determination device further includes a stimulus applying unit that gives a stimulus for raising a person's arousal level, and a timing setting unit that sets a timing to give the stimulus, and the timing giving unit reduces heart rate fluctuation. It is preferable to set the timing for applying the stimulus after the amplitude spectrum power of the frequency component exceeds a predetermined value until the latest minimum value is obtained.
- the arousal level determination method further includes a stimulus applying step for applying a stimulus for increasing a person's arousal level, and a timing setting step for setting a timing for applying the stimulus.
- a stimulus applying step for applying a stimulus for increasing a person's arousal level
- a timing setting step for setting a timing for applying the stimulus.
- heart rate fluctuation It is preferable to set the timing for applying the stimulus after the amplitude spectrum power of the low frequency component exceeds the predetermined value until the latest minimum value is obtained.
- the stimulus is given until the nearest minimum value is reached, that is, when an active person resists sleepiness. Therefore, stimulation can be given at an appropriate timing.
- the physiological index that has a correlation with sympathetic nerve activity and that has a correlation with parasympathetic activity.
- sympathetic nervous system activity is activated in the state of resistance to sleepiness.
- parasympathetic nervous system activity usually decreases.
- the activity of the parasympathetic nervous system may increase simultaneously with the increase of the activity of the sympathetic nervous system.
- the physiological index correlated with the parasympathetic nerve activity is further considered. This makes it possible to accurately detect a state that is truly resistant to sleepiness.
- a heartbeat signal is used as the biological signal, and a heartbeat signal is used as the physiological index. It is preferable to use the amplitude spectrum power of the heartbeat fluctuation low frequency component acquired from the above and the amplitude spectrum power of the heartbeat fluctuation high frequency component.
- the amplitude spectrum power of the heartbeat fluctuation high-frequency component obtained from the heartbeat signal correlates with the parasympathetic nerve activity. Therefore, by using the amplitude spectrum power of the heartbeat fluctuation high-frequency component as a physiological index, It is possible to exclude typical cases.
- the wakefulness level determination device when the determination means has an amplitude spectrum power of the heartbeat fluctuation low frequency component larger than the amplitude spectrum power of the heartbeat fluctuation high frequency component, the wakefulness level decreases. It is preferable to determine whether or not.
- the arousal level determination method in the determination step, when the amplitude spectral power of the heartbeat fluctuation low frequency component is larger than the amplitude spectral power of the heartbeat fluctuation high frequency component, the arousal level It is preferable to determine whether or not the value has decreased.
- An arousal level determination apparatus includes threshold setting means for setting a threshold value for determining whether or not a person's arousal level is lowered, and the determination means includes a heartbeat fluctuation low frequency When the amplitude spectral power of the component is larger than the threshold value, it is preferable to determine that the arousal level is lowered.
- the arousal level determination method includes a threshold for setting a threshold value for determining whether or not a person's arousal level is lowered, and a value setting step.
- a threshold for setting a threshold value for determining whether or not a person's arousal level is lowered
- a value setting step In the above determination step, When the amplitude spectral power of the heartbeat fluctuation low frequency component is larger than the threshold value, it is preferable to determine that the arousal level is lowered.
- the threshold for determining whether or not the arousal level is decreasing is, for example, individual differences, Even if it is the same person, it is preferable to change according to the physical condition change of the day.
- the threshold is set by the threshold setting means or the threshold setting step. Then, when the amplitude spectrum power of the heartbeat fluctuation low frequency component is larger than the threshold value, it is determined that the arousal level is lowered. Therefore, drowsiness detection can be performed according to the characteristics of each person, and the accuracy of drowsiness detection can be improved.
- the threshold setting means sets the threshold based on the heart rate signal and the amplitude spectrum power of the heart rate fluctuation low frequency component.
- the threshold value is set based on the heartbeat signal and the amplitude spectrum power of the heartbeat fluctuation low frequency component in the threshold setting step.
- the threshold value based on the heart rate signal and the amplitude spectrum power of the heart rate fluctuation low frequency component, it becomes possible to set the threshold value according to each person's physiological characteristics. .
- the physiological index indicating the strength of the state against drowsiness is acquired from the biological signal of an active person, and the arousal level is determined based on the physiological index. It becomes possible to detect weak sleepiness of the person of the present more reliably.
- FIG. 1 is a block diagram showing the overall configuration of the arousal level determination device according to the first embodiment.
- FIG. 2 is a flowchart illustrating a processing procedure of a wakefulness level reduction detection process performed by the wakefulness level determination device according to the first embodiment.
- FIG. 3 is a flowchart showing a processing procedure of heartbeat signal preprocessing in the arousal level reduction detection processing.
- Figure 4 shows the flow of the feature extraction process in the arousal level detection process It is a chaat.
- FIG. 5 is a diagram illustrating an example of a heartbeat signal.
- FIG. 6 is a schematic diagram for explaining the binarization processing of the heartbeat signal.
- FIG. 7 is a diagram showing a binarized signal obtained by binarizing the heartbeat signal.
- FIG. 8 is a schematic diagram for explaining the heartbeat cycle calculation process.
- FIG. 9 is a schematic diagram for explaining the interpolation processing of the cardiac cycle.
- FIG. 10 is a schematic diagram for explaining the FFT processing of the cardiac cycle.
- Fig. 11 is a schematic diagram for explaining the integration of the power spectrum.
- Fig. 12 is a chart showing the change over time of the amplitude spectral power.
- Fig. 13 is a diagram for explaining a method for detecting drowsiness (a method for determining a decrease in arousal level).
- FIG. 14 is a diagram for explaining a method for setting the timing for applying stimulus.
- Fig. 15 is a table showing criteria for sensory evaluation level (drowsiness level).
- FIG. 16 is a diagram showing an example of the drowsiness detection result.
- FIG. 17 is a diagram showing another example of the drowsiness detection result.
- FIG. 18 is a diagram illustrating an example of the timing for applying a stimulus.
- FIG. 19 is a diagram illustrating another example of the timing for applying a stimulus.
- FIG. 20 is a block diagram showing the overall configuration of the arousal level determination device according to the second embodiment.
- FIG. 21 is a flowchart showing the processing procedure of the arousal level reduction detection process by the arousal level determination device according to the second embodiment.
- FIG. 22 is a flowchart showing the procedure of the feature amount extraction process in the arousal level reduction detection process.
- Figure 23 shows the relationship between the heart rate, the low frequency component of heart rate fluctuation, and the set threshold.
- Figure 24 shows the method for determining the increase / decrease of the low frequency component of heart rate and heart rate fluctuation.
- FIG. 25 is a diagram showing an example of the drowsiness detection result when the high-frequency component power of heartbeat fluctuation is not taken into account.
- FIG. 26 is a diagram showing an example of the drowsiness detection result in consideration of the high-frequency component power of heartbeat fluctuation.
- FIG. 27 is a diagram showing the sleepiness detection result when the sleepiness determination threshold is fixed.
- FIG. 28 shows the result of sleepiness detection when the sleepiness determination threshold is varied.
- FIG. 29 is a diagram illustrating an example of a change result of the sleepiness determination threshold value.
- FIG. 1 is a block diagram showing the overall configuration of the arousal level determination device 1.
- arousal level determination device 1 is mounted on a vehicle and a decrease in the arousal level of the driver of the vehicle is detected will be described as an example.
- the arousal level determination device 1 detects whether or not the driver's arousal level is decreasing based on a heartbeat signal obtained from the driver, and presents the detection result to the driver or a stimulus that increases the arousal level. Is to give.
- the arousal level determination device 1 includes a heart rate sensor 10 and an electronic control unit (hereinafter referred to as “ECU”) 20.
- the ECU 20 includes a heart rate signal preprocessing unit 2 1 and a feature amount extraction unit 2 2. , Drowsiness detection unit 2 3, Drowsiness occurrence notification unit 2 4. Stimulus timing setting unit 25, drowsiness removal stimulus output unit 26.
- the heart rate sensor 10 is a potential heart rate sensor that detects a pulsed voltage (cardiac potential) generated during myocardial contraction, and acquires a driver's heart rate signal.
- the heart rate sensor 10 detects the cardiac potential from an electrode attached to, for example, a vehicle steering wheel.
- the heart rate sensor 10 outputs the acquired heart rate signal to the ECU 20.
- the heart rate sensor 10 functions as a biological signal acquisition unit described in the claims, and executes a biological signal acquisition step.
- the ECU 20 is stored in a microprocessor that performs calculations, a ROM that stores programs for causing the microprocessor to execute each process, and a RAM and battery that stores various data such as calculation results. It is composed of a pack-up RAM and the like.
- the heartbeat signal preprocessing unit 21 reads the heartbeat signal from the heartbeat sensor 10 at regular intervals, and acquires the heartbeat cycle (RR interval) time series from the heartbeat signal.
- RR interval heartbeat cycle
- the heartbeat cycle time series acquired by the heartbeat signal preprocessing unit 21 is output to the feature amount extraction unit 22.
- the feature quantity extraction unit 22 uses the amplitude spectrum power of the heartbeat fluctuation low frequency component, which is a physiological index indicating the strength of the state against sleepiness, from the heartbeat cycle time series acquired by the heartbeat signal preprocessing unit 21 ( Time series). More specifically, first, the time-series data of the heartbeat cycle is subjected to FFT processing to obtain an amplitude spectrum that is a frequency component of heartbeat fluctuation. Next, the frequency band of the low frequency component is designated for this amplitude spectrum, and the amplitude spectrum of the band is integrated. By repeating this process, the amplitude spectrum power time series of the heartbeat fluctuation low frequency component is acquired. As described above, the heartbeat signal preprocessing unit 21 and the feature amount extraction unit 22 are provided with the fingers described in the claims. It functions as a mark acquisition means and executes an index acquisition step. The amplitude spectrum power (time series) of the heartbeat fluctuation low frequency component acquired by the feature quantity extraction unit 22 is output to the sleepiness detection unit 23.
- the drowsiness detection unit 2 3 detects whether or not the driver has drowsiness based on the amplitude spectrum power (time series) of the heartbeat fluctuation low frequency component acquired by the feature amount extraction unit 2 2 (that is, the driver Is determined). More specifically, when the amplitude spectrum power of the heartbeat fluctuation low frequency component (hereinafter referred to as “heartbeat fluctuation low frequency component power”) is equal to or less than the first sleepiness judgment threshold D 1, there is no sleepiness. And judged. If the heartbeat fluctuation low-frequency component power is greater than the first sleepiness discrimination threshold D 1 and less than or equal to the second sleepiness discrimination threshold D 2, there is weak sleepiness (the degree of arousal is slightly reduced) Is determined).
- the drowsiness detection unit 23 functions as a determination unit described in the claims, and executes a determination step.
- a method of setting the first drowsiness determination threshold value D 1 and the second drowsiness determination threshold value D 2 will be described.
- An example is a method of quantifying sleepiness by other methods to obtain a correlation between sleepiness and heartbeat fluctuation low frequency component power.
- the method for setting the first drowsiness discrimination threshold D 1 and the second drowsiness discrimination threshold D 2 is not limited to this method.
- learning is performed in consideration of the health condition of the driver such as blood pressure and heart rate. You may set by.
- the drowsiness occurrence notifying unit 24 provides the driver with information on arousal level reduction when a flag indicating that the drowsiness detection unit 23 is in a weak drowsiness state is input.
- the drowsiness generation notification unit 24 presents the arousal level reduction information using a display for displaying character information and video information, a speaker for reproducing voice information, and the like.
- the stimulus timing setting unit 25 sets the timing to give the driver a stimulus to remove drowsiness based on the heart rate fluctuation low frequency component power (time series) input from the sleepiness detection unit 23. To do. More specifically, as shown in FIG. 14, the stimulation timing setting unit 25 has a heartbeat fluctuation low-frequency component power of a predetermined value D x (first drowsiness determination threshold D 1 or second drowsiness determination.
- the timing for applying the stimulus is set until the most recent minimum value is reached (between time T s and time Tm). That is, the stimulation timing setting unit 25 functions as a timing setting unit described in the claims and executes a timing setting step. In addition, the stimulation timing setting unit 25 outputs a drowsiness removal stimulation signal to the drowsiness removal stimulation output unit 26 at the set timing.
- a predetermined time e.g, 60 seconds
- the stimulation timing setting unit 25 outputs a drowsiness removal stimulation signal to the drowsiness removal stimulation output unit 26 at the set timing.
- the stimulus application timing is set in this way for the following reason. That is, it is preferable that the timing of giving the stimulus coincides with or slightly before the timing when the driver who resists sleepiness wants the stimulus. Giving a stimulus at the timing when the driver wants the stimulus is in response to the driver's request and is considered to be highly effective in removing drowsiness.
- the timing Tu at which the driver wants a stimulus is determined by the heartbeat fluctuation low frequency component power exceeding the predetermined value (first sleepiness determination threshold D 1 or second sleepiness determination threshold D 2). After the time Tm at which the first local minimum value was taken, or after the heartbeat fluctuation low frequency component power exceeded the predetermined value, it was found that it was within about 60 3 ⁇ 4 ⁇ >.
- the timing for applying the stimulus is from the time T s when the heartbeat fluctuation low frequency component power exceeds a predetermined value to the time Te when the power falls below the predetermined value (that is, the driver fights sleepiness). Or the time TP when the low frequency component power is maximized (that is, when the driver's sleepiness is strongest) is preferable.
- the drowsiness removal stimulus output unit 26 gives the driver a stimulus for removing drowsiness based on the drowsiness removal stimulus signal input from the stimulus timing setting unit 25.
- the following stimuli can be given to the driver. That is, for example, for buzzer, audio, sound stimulation using horn, meter lighting, light stimulation using room lighting, vibration device embedded in handle or seat, tactile / hot / cold sense using air conditioning Examples include irritation and irritation caused by the smell of spraying fragrances.
- the drowsiness removal stimulus output unit 26 functions as the stimulus applying means described in the claims, and executes the stimulus applying step. Next, the operation of the wakefulness determination device 1 and the method of determining wakefulness will be described using FIGS. 2 to 4 and FIGS. 5 to 14 together.
- FIG. 2 is a flowchart showing a processing procedure of the arousal level decrease detection process by the arousal level determination device 1.
- 3 and 4 are flowcharts showing the processing procedures of the heartbeat signal preprocessing and the feature quantity extraction processing in the arousal level reduction detection processing. These processes are performed by the ECU 20. From the time the ECU 20 is turned on until the power is turned off, the ECU 20 is repeatedly executed at a predetermined timing.
- step S 1 0 the heart rate signal is read from the heart rate sensor 10.
- step S 1 0 2 pre-processing is performed on the heartbeat signal read in step S 1 0 0.
- this heartbeat signal pre-processing will be described with reference to FIG. 3.
- step S 2 0 the time-series data of the heartbeat signal is processed by the pan-pass filter, so that the time-series data of the heartbeat signal is processed.
- the components in the passband 0.1 ⁇ ⁇ to 30 ⁇ ⁇ are extracted.
- Figure 5 shows the result of this process.
- step S 2 0 2 as shown in FIG. 5, a threshold value TH for detecting heartbeat timing from time-series data of heartbeat signals.
- the above waveform portion is cut out.
- the timing at which the extracted waveform portion is maximized is set to 1, and the other timings are set to 0 and binarized.
- a series of heartbeat timings are obtained as shown in FIG.
- step S 2 0 4 the time (seconds) from each heartbeat timing t 1 to the next heartbeat timing t 2 is obtained, and the obtained time (t 2-t 1) Is added to each heartbeat timing t 1 to obtain information on the heartbeat cycle.
- step S 2 06 as shown in FIG. 9, the heartbeat period information is completed and the heartbeat period curve C is obtained, whereby time-series data of the heartbeat period is obtained. Thereafter, the process proceeds to step S 1 0 4 shown in FIG.
- step S 10 04 a feature amount extraction process for acquiring heart rate fluctuation low-frequency component power (time series) from the time series data of the heart cycle acquired in step S 2 06 is executed.
- the feature amount extraction processing will be described with reference to FIG.
- step S 3 0 0 as shown in Fig. 10, for the time series data of the cardiac cycle in the analysis unit interval width T term (seconds) before the reference time T, which is an arbitrary time stamp, Fast Fourier transform (FFT) processing is performed.
- step S 3 0 2 as shown in Fig. 11, in the power spectrum obtained for each analysis unit interval by FFT processing, the frequency band of the low frequency component (0.1 Hz) The amplitude spectrum is integrated for (near).
- step S 3 0 4 every time a fixed time elapses and the reference time is reached, FFT processing is performed on the time-series data of the heartbeat period in the analysis unit interval width T term to obtain a power spectrum.
- the integrated process is repeated.
- time-series data of the amplitude spectrum power in the frequency band of the low frequency component is acquired.
- This time series data of the amplitude spectrum power is the heartbeat fluctuation low frequency component power (time series).
- the process proceeds to step S 1 06 shown in FIG.
- step S 106 a determination is made as to whether or not the heartbeat fluctuation low frequency component power is greater than zero and less than or equal to the first sleepiness determination threshold D 1 (see FIG. 13).
- the heartbeat fluctuation low frequency component power is equal to or less than the first drowsiness determination threshold D1
- it is determined that there is no drowsiness and the process is temporarily exited.
- the heartbeat fluctuation low-frequency component part 1 is larger than the first drowsiness determination threshold value D1
- the process proceeds to step S1008.
- step S 1 0 8 a determination is made as to whether or not the heartbeat fluctuation low frequency component power is greater than the second drowsiness discrimination threshold D 2 (see FIG. 13).
- the heartbeat fluctuation low frequency component power is less than or equal to the second drowsiness threshold D 2
- it is determined that there is weak drowsiness a state in which the arousal level is slightly reduced
- step S Processing proceeds to 1 1 0.
- the heartbeat fluctuation low frequency component power H RV L is greater than the second drowsiness threshold D 2
- step S Processing shifts to 1 1 2.
- step S 1 1 the awakening level reduction information is presented to the driver. For example, text information or video information indicating that the degree of arousal is low is displayed on the display, or voice information indicating that the degree of arousal is low is displayed. By outputting from the peak power, the arousal level reduction information is presented. After that, the process is finished.
- step S 1 1 the driver is stimulated to remove sleepiness based on the heartbeat fluctuation low frequency component power (time series).
- the timing to give is set. More specifically, the timing at which the stimulus is applied until a predetermined time (eg, 60 seconds) elapses from the time T s when the heartbeat fluctuation low-frequency component power becomes greater than the second sleepiness discrimination threshold D 2. Is set. Or, when the heartbeat fluctuation low-frequency component power exceeds the second drowsiness detection threshold D 2, the timing for giving the stimulus is between the nearest minimum value (between time T s and time T m). It is set (see Fig. 14).
- a drowsiness removal stimulation signal is output based on the timing set in step S 1 1 2, and a stimulus for removing drowsiness is given to the driver.
- a stimulus for removing drowsiness is given to the driver.
- the heartbeat fluctuation low frequency component power that is a physiological index indicating the strength of the state against sleepiness is acquired from the heartbeat signal of the driver, and the arousal level is based on the heartbeat fluctuation low frequency component ratio. Therefore, it is possible to more reliably detect weak sleepiness (decrease in arousal level) of the driver while driving.
- Heart rate fluctuation low-frequency component power that correlates with sympathetic nerve activity is used as a physiological index to determine arousal level, so it is possible to more reliably detect the state of sympathetic nerve activity, that is, the state of fighting sleepiness. .
- the heartbeat fluctuation low-frequency component power correlates with sympathetic nerve activity, and its magnitude correlates with the activity of sympathetic nerve activity, that is, the magnitude against the drowsiness. Therefore, as the degree of arousal decreases and the degree of resistance to sleepiness increases, the sympathetic nerve activity is activated and the heartbeat fluctuation low-frequency component power increases.
- the heartbeat fluctuation low-frequency component power is larger than the first drowsiness determination threshold value D1, it is determined that the arousal level is reduced. It is possible to reliably detect a state in which the degree is lowered.
- the sympathetic nerve activity is activated and the heartbeat fluctuation increases and the low frequency component power increases.
- the heartbeat fluctuation low-frequency component power is larger than the second drowsiness determination threshold D2
- the appropriate timing for applying the stimulus for removing drowsiness coincides with or slightly before the timing when the driver who is fighting drowsiness wants the stimulus.
- the timing at which a driver who is struggling with sleepiness wants a stimulus correlates with the strength of the state that resists sleepiness.
- the timing for applying the stimulus is set based on the heartbeat fluctuation low-frequency component power, which is a physiological index indicating the strength of the state against sleepiness, so that the stimulus can be applied at an appropriate timing. It becomes ability.
- the heartbeat fluctuation low frequency component power is the second drowsiness discrimination threshold D
- the stimulus is given when the degree of arousal is reduced and the degree of resistance to sleepiness is increased, so that the stimulus can be given at an appropriate timing.
- the heartbeat fluctuation low frequency component power exceeds the second drowsiness determination threshold value D2
- the latest minimum value is obtained, that is, when the driver is fighting drowsiness. Stimulation can be given at the right time (when resisting sleepiness).
- a wakefulness decrease detection test was performed.
- the evaluation test was performed by comparing the drowsiness detection result based on the heartbeat fluctuation low-frequency component power of the subject (driver) with the drowsiness level sensory evaluation result obtained from the face image.
- the test method and test results will be described. The test was performed according to the following procedure.
- the facial image time series is evaluated based on the following levels 1 to 5, and the subject's sleepiness is divided into five levels (sensory evaluation). The number of evaluators for sensory evaluation was two.
- Level 1 Not sleepy at all (Gaze movement is fast and frequent, blinking is a stable cycle of about 2 times in 2 seconds, movement is active and accompanied by body movement)
- Level 5 Very sleepy (closes eyelids, head tilts forward, head falls back) (Source: “Human Sensory Measurement Manual, Volume 1”, P 1 46, Human Life Engineering Research Center) 3. Calculate the average value Sens of sensory evaluation of two people.
- the wakefulness determination device 1 detects the subject's sleepiness based on the heartbeat fluctuation low-frequency component power. In order to ensure fair evaluation, the evaluator was not given any information about the heartbeat fluctuation low-frequency component power obtained from the subject's heartbeat signal in advance.
- Figure 16 shows the test results of the arousal level detection test.
- the horizontal axis in Fig. 6 is the elapsed time (seconds).
- the upper part (A) shows the sleepiness level by sensory evaluation
- the lower part (B) shows sleepiness detection based on the heartbeat fluctuation low-frequency component power by the wakefulness determination device 1. Results are shown. More specifically, area I in Fig. 16 is an area determined to be drowsy by sensory evaluation, area II is an area determined to be somewhat sleepy by sensory evaluation, and area III is determined to be sleepy by sensory evaluation. Area. On the other hand, region I 'in Fig.
- region 16 is the region where sleepiness is determined by wakefulness determination device 1 (0 ⁇ heart rate fluctuation low-frequency component ⁇ first sleepiness detection threshold D1), and region II' is Area determined to be drowsiness by the arousal level determination device 1 (first drowsiness threshold D 1 ⁇ heart rate fluctuation low frequency component ⁇ second drowsiness threshold D 2), area III ' This is the region determined as strong sleepiness by the device 1 (heartbeat fluctuation low frequency component power> second sleepiness discrimination threshold D 2).
- the arousal level determination device 1 determines that drowsiness is weak. In addition, when the drowsiness level based on the sensory evaluation is drowsiness level D 2 (sleepy), the arousal level determination device 1 determines that drowsiness is strong. That is, the drowsiness of the subject was successfully detected by the arousal level determination device 1, and the effectiveness of the arousal level determination device 1 was confirmed.
- Example 1 Awakening reduction detection test was performed on subjects different from Example 1. Test method described above Since this is the same as Example 1, the description is omitted here.
- Figure 17 shows the test results of this arousal level detection test. Also in this test result, when the drowsiness level by sensory evaluation is level D 1 (somewhat sleepy), the arousal level determination device 1 determines that drowsiness is weak. In addition, when the drowsiness level by sensory evaluation is drowsiness level D 2 (appears to sleep), the arousal level determination device 1 determines that drowsiness is strong. In other words, the drowsiness of the subject was successfully detected by the arousal level determination device 1, and the effectiveness of the arousal level determination device 1 was confirmed.
- a stimulus application timing test was performed as an evaluation test of the stimulus application timing of the arousal level determination device 1 described above.
- the subject driver
- Figure 18 shows the test results of the stimulus application timing test.
- the horizontal axis in Fig. 18 is the elapsed time (seconds), and the vertical axis is the heartbeat fluctuation low frequency component power.
- the timing when the subject wanted the stimulus was indicated by an arrow.
- Example 3 A stimulus application timing test was performed on subjects different from Example 3. Since the test method is the same as in Example 3 described above, description thereof is omitted here.
- Figure 19 shows the test results of this stimulus application timing test.
- the timing when the subject wanted the stimulus see the arrow in Fig. 19.
- Force Heartbeat fluctuation Low-frequency component power for a predetermined time after the second drowsiness discrimination threshold D 2 ( 60 seconds), and after the heartbeat fluctuation low-frequency component power exceeds the second drowsiness detection threshold D 2 It was confirmed that the stimulus application timing set in the degree determination device 1 is appropriate.
- FIG. 20 is a block diagram showing the overall configuration of the arousal level determination device 2. As shown in FIG. In FIG. 20, the same or similar components as those in the first embodiment are denoted by the same reference numerals.
- the arousal level determination device 2 is different from the first embodiment in that it includes an ECU 20D instead of the ECU 20 described above. More specifically, the ECU 20D replaces the feature amount extraction unit 22 and the drowsiness detection unit 23 included in the ECU 20 with a feature amount extraction unit 22
- D Drowsiness detection unit 23 This is different from ECU 20 in that it has D.
- the ECU 20D is different from the ECU 20 in that it further includes a threshold value setting unit 27 for setting a threshold value for determining whether or not the driver is drowsy.
- Other configurations are the same as or similar to those of the first embodiment described above, and thus description thereof is omitted here.
- the feature quantity extraction unit 22D calculates the spectrum power (time series) of the heartbeat fluctuation high frequency component from the heartbeat cycle time series acquired by the heartbeat signal preprocessing unit 21. It differs from the above-described feature quantity extraction unit 22 in that it is acquired. More specifically, the feature amount extraction unit 22D first performs FFT processing on the time series data of the heartbeat cycle to obtain an amplitude spectrum that is a frequency component of heartbeat fluctuation. Next, a frequency band of a high frequency component is designated for this amplitude spectrum, and the amplitude spectrum of the band is integrated. By repeating this process, the amplitude spectrum power time series of the heartbeat fluctuation high-frequency component is acquired. The method for acquiring the heartbeat fluctuation low frequency component power time series is the same as described above, and the description thereof is omitted here.
- the heartbeat signal preprocessing unit 21 and the feature amount extraction unit 22D are also It functions as the index acquisition means described in the range, and executes the index acquisition step.
- the heart rate fluctuation low frequency component power obtained by the feature amount extraction unit 2 2 D is output to the threshold value setting unit 27.
- the heartbeat fluctuation low frequency component power (time series) and the heartbeat fluctuation high frequency component power (time series) are output to the drowsiness detection unit 23 3 D.
- the threshold value setting unit 2 7 includes the heart rate (RR interval) acquired by the heart rate signal preprocessing unit 21 and the heart rate fluctuation low frequency component power (time series) acquired by the feature amount extraction unit 2 2 D. ) Based on the above, a threshold is set to determine whether the driver is drowsy. That is, the threshold value setting unit 27 functions as a threshold value setting means described in the claims. The threshold value set in the threshold value setting unit 27 is output to the drowsiness detection unit 23 3 D.
- the threshold value setting method will be described by taking as an example the case of setting the second drowsiness determination threshold value D2.
- the method for setting the first drowsiness determination threshold value D 1 is the same as or similar to the second drowsiness determination threshold value D 2, and thus the description thereof is omitted here.
- a plurality of preset thresholds are set according to the change state of the heart rate (RR interval) and the change state of the heartbeat fluctuation low frequency component power. Switching methods and computation methods. First, a method of switching a plurality of preset threshold values will be described with reference to FIG. Figure 23 is a table showing the relationship between the heart rate, the low frequency component of heart rate fluctuation, and the set threshold value. In this case, as shown in Fig. 23, four preset thresholds D 2 A, D 2 B, D 2 C and D 2 D are the heart rate change state and heart rate fluctuation low frequency component. It can be switched according to the state of change.
- the threshold D 2 A is selected when the heart rate is increasing or decreasing and the heart rate fluctuation low frequency component power is increasing, and (2) the heart rate is increasing or decreasing.
- the threshold value D 2 B is selected when there is no change or the heart rate fluctuation low frequency component parameter is unchanged or decreasing.
- the threshold D 2 C is selected, (4)
- the threshold value D 2 D is selected when the heart rate does not change and the heart rate fluctuation low frequency component power remains unchanged or is decreasing.
- the relationship between the four threshold values D 2 A, D 2 B, D 2 C, and D 2 D is “D 2 A> D 2 C> D 2 B> D 2 D”.
- Examples of methods for determining the increase or decrease in heart rate and heart rate fluctuation low-frequency component power include a method based on statistical tests and a method based on the slope of the interval data time series.
- the horizontal axis in Fig. 24 is the elapsed time t (seconds) from the detection start time, and the vertical axis is the heart rate or heart rate fluctuation low frequency component power.
- t O is the drowsiness detection start time
- t s is an arbitrary time for defining section 1.
- T n is the current time
- t b is an arbitrary time for defining interval 3 and interval 2.
- the intervals shown in Fig. 24 are compared by statistical tests to determine whether the heart rate or heart rate fluctuation low-frequency component power has increased or decreased.
- combinations of test sections include a combination of sections 1 and 3, or a combination of sections 2 and 3, and tests are performed between these combined sections. .
- the hypothesis at the time of testing is as follows. That is, the null hypothesis H 0 is the case where the data included in both sections is extracted from a group with the same distribution, and the alternative hypothesis HI is the case where the data included in both sections is extracted from a group having a different distribution. To do. Therefore, when the null hypothesis H 0 is adopted by the test, it can be determined that there is no significant difference between the two sections, so that the heart rate or the heart rate fluctuation low frequency component power is determined to be unchanged (no increase or decrease). The On the other hand, when the alternative hypothesis HI is adopted by the test, it can be determined that there is a significant difference between the two sections, so that it is determined that the heart rate or heart rate fluctuation low frequency component power is increasing or decreasing.
- the slope B is obtained by performing a single regression calculation on the interval 3, and if the absolute value of the slope B is greater than the slope threshold ⁇ (I ⁇ I> ⁇ ), it is determined that there is an increase or decrease. be able to.
- the slope threshold value ⁇ is a constant that is arbitrarily set.
- M is the interval average value of the heartbeat fluctuation low frequency component power
- SD is the interval standard deviation of the heartbeat fluctuation low frequency component power
- the first sleepiness discrimination threshold D 1 is obtained by the following equation (2).
- a is a coefficient satisfying a ⁇ b.
- the drowsiness detection unit 23D detects the driver based on the heartbeat fluctuation low frequency component power (time series) when the heartbeat fluctuation low frequency component power acquired by the feature amount extraction unit 22D is larger than the heartbeat fluctuation high frequency component power.
- the presence or absence of sleepiness ie, driver's arousal level
- heart rate fluctuation low frequency component power When the first drowsiness determination threshold D 1 set by the unit 27 is less than or equal to 1, it is determined that there is no sleepiness. If the heartbeat fluctuation low frequency component is greater than the first drowsiness threshold tDl and less than or equal to the second drowsiness threshold D 2, there is weak sleepiness (the degree of arousal decreases slightly) Is determined).
- the drowsiness detection unit 23D also functions as the determination unit described in the claims, and executes the determination step.
- FIG. 21 is a flowchart showing the processing procedure of the arousal level decrease prediction process by the arousal level determination device 2. This process is performed by the ECU 20D, and is repeatedly executed at a predetermined timing from when the power of the ECU 20D is turned on to when it is turned off.
- Steps S 400 and S 402 are the same as steps S 100 and S 102 described above, and thus a duplicate description is omitted here.
- step S 40.4 feature amount extraction processing 2 is executed for acquiring heartbeat fluctuation low frequency component power (time series) and heartbeat fluctuation high frequency component power (time series) from time series data of the heartbeat period.
- time series heartbeat fluctuation low frequency component power
- time series heartbeat fluctuation high frequency component power
- step S 500 as in step S 300 described above, the time-series data of the cardiac cycle in the analysis unit interval width T term (seconds) before the reference time T, which is an arbitrary time stamp, Fast Fourier transform (FFT) processing is performed.
- FFT Fast Fourier transform
- step S502 in the power spectrum obtained for each analysis unit interval by FFT processing, the frequency band of the low-frequency component (near 0.1 Hz) In addition, the amplitude spectrum is integrated, and the amplitude spectrum is integrated for the frequency band of high-frequency components (near 0.3 Hz).
- step S 5 0 4 every time a fixed time elapses and the reference time is reached, FFT processing is performed on the time-series data of the heartbeat period in the analysis unit interval width T term to obtain a power spectrum. The integrated process is repeated. Thereby, heartbeat fluctuation low frequency component power (time series) and heartbeat fluctuation high frequency component power (time series) are acquired. Thereafter, the processing shifts to step S 4 06 shown in FIG.
- step S 4 06 a threshold value for determining whether or not the driver is drowsy is set based on the heart rate (RR interval) and the heart rate fluctuation low frequency component power (time series). Since the threshold setting method is as described above, a duplicate description is omitted here.
- step S 400 a determination is made as to whether or not the heartbeat fluctuation low frequency component power is greater than the heartbeat fluctuation high frequency component power. If the heartbeat fluctuation low-frequency component power is larger than the heartbeat fluctuation high-frequency component power, the process proceeds to step S 4 10. On the other hand, when the heartbeat fluctuation low frequency component power is equal to or less than the heartbeat fluctuation high frequency component power, it is determined as an exceptional case, and the process is temporarily exited.
- step S 4 10 a determination is made as to whether or not the heartbeat fluctuation low-frequency component power is greater than zero and less than or equal to the first drowsiness determination threshold D 1 set in step S 4 0 6. Is done.
- the heartbeat fluctuation low frequency component power is equal to or less than the first sleepiness determination threshold D1
- it is determined that there is no sleepiness and the process is temporarily terminated.
- the heartbeat fluctuation low frequency component power is larger than the first drowsiness determination threshold value D1
- the process proceeds to step S412.
- step S 4 1 2 a determination is made as to whether or not the heartbeat fluctuation low-frequency component threshold is greater than the second sleepiness determination threshold D 2 set in step S 4 06.
- the heartbeat fluctuation low frequency component power is less than or equal to the second drowsiness threshold D2
- it is determined that there is weak sleepiness a state in which the arousal level is slightly reduced
- the process proceeds to step S 4 1 4.
- the heartbeat fluctuation low frequency component power HRVL is larger than the second sleepiness discrimination threshold D2
- it is determined that there is a strong sleepiness (a state in which the arousal level is greatly reduced)
- step S4 1 6 Processing shifts to.
- Step S 4 14 is the same as Step S 1 1 0 described above, and Steps S 4 1 6 and S 4 1 8 are the same as Steps S 1 1 2 and S 1 1 4. I will omit the description.
- sympathetic nervous system activity is activated in the state of resistance to sleepiness. At that time, parasympathetic nervous system activity usually decreases. However, in exceptional cases, parasympathetic nervous system activity may increase simultaneously with increased sympathetic nervous system activity.
- the heartbeat fluctuation low frequency component which is a physiological index correlated with sympathetic nerve activity
- the heartbeat fluctuation high frequency component power which is a physiological index correlated with parasympathetic nerve activity. Since it is determined whether or not the arousal level is lowered, it is possible to accurately detect a state that is truly resistant to sleepiness by excluding the exceptional cases described above.
- the threshold value is set based on the heart rate signal and the heart rate fluctuation low frequency component power, and the arousal level is increased when the heart rate fluctuation low frequency component value is larger than the threshold value. It is determined that it has decreased. Therefore, the threshold value can be set according to the physiological characteristics of each driver, and the accuracy of sleepiness detection can be improved.
- FIG. 25 is a diagram showing an example of the drowsiness detection result when the high frequency component power of heartbeat fluctuation is not taken into consideration.
- FIG. 26 is a diagram showing an example of the drowsiness detection result when the heartbeat fluctuation high-frequency component power is considered.
- the horizontal axis in Figs. 25 and 26 is the elapsed time (seconds) from the start of detection, and the vertical axis is (A) sensory evaluation results, (B) drowsiness detection results (estimation results), (C) from the top.
- Heartbeat fluctuation low-frequency component power This is a heartbeat fluctuation high-frequency component power.
- the middle part of Figure 25 As shown in the figure, it is determined that there is strong sleepiness when the heartbeat fluctuation low frequency component power increases at an elapsed time of about 1500 seconds. However, at this time, the heartbeat fluctuation high-frequency component power is also increasing at the same time, and the driver's arousal level is not in a state of true decline. In some cases, the parasympathetic nervous system activity increases at the same time that the sympathetic nervous system activity increases. Therefore, drowsiness is not detected in the sensory evaluation shown in the upper part (A) of Figure 25. In other words, such an exceptional case cannot be excluded unless high-frequency component power of heartbeat fluctuation is taken into consideration.
- the heartbeat fluctuation high frequency component power is considered, that is, the heartbeat fluctuation low frequency component power is larger than the heartbeat fluctuation high frequency component power, and the heartbeat fluctuation low frequency component power exceeds the second drowsiness determination threshold D2.
- the heartbeat fluctuation high-frequency component power an exceptional case in which the activity of the parasympathetic nervous system increases simultaneously with the increase of the activity of the sympathetic nervous system can be eliminated. I was able to confirm. (Example 6)
- FIG. 27 is a diagram showing an example of the sleepiness detection result when the sleepiness determination threshold is fixed.
- FIG. 28 is a diagram showing an example of the sleepiness detection result when the sleepiness determination threshold is varied.
- the horizontal axis in Figs. 27 and 28 is the elapsed time (seconds) from the start of detection. From the top, the vertical axis shows (A) sensory evaluation of sleepiness, and (B) heart rate fluctuation low-frequency component power. is there.
- the sleepiness threshold is fixed, as shown in Fig. 27, when the elapsed time is around 1600 seconds and it is determined that there is strong sleepiness according to sensory evaluation (arrow in the figure) However, the heartbeat fluctuation low-frequency component power is less than the second sleepiness determination threshold D 2, and it is not determined that there is strong sleepiness. Thus, when the threshold for sleepiness determination is fixed, it is difficult to flexibly cope with individual differences and changes in physical condition of the driver.
- the drowsiness determination threshold is made variable, as shown in Fig. 28, when the elapsed time is around 1600 seconds, it is determined that there is strong sleepiness according to sensory evaluation (Fig. The heartbeat fluctuation low frequency component power exceeds the second drowsiness threshold D 2 after the change, and it is determined that there is strong sleepiness.
- the heartbeat fluctuation low frequency component power exceeds the second drowsiness threshold D 2 after the change, and it is determined that there is strong sleepiness.
- FIG. 29 is a diagram illustrating an example of a change result of the sleepiness determination threshold value.
- the horizontal axis in Fig. 29 is the elapsed time (seconds) from the start of detection, and the vertical axis is (A) heart rate fluctuation low frequency component ratio and (B) heart rate from the top.
- the threshold setting method the above-described statistical test method was used.
- the second drowsiness determination threshold D 2 is changed from 1 0 X 1 0 4 to 4.0 5 X 1 0 at the time when the interval 2 ends. Changed to 4 . In this way, it was confirmed that the sleepiness judgment threshold and the value were changed appropriately based on the heartbeat fluctuation low frequency component power and the heart rate.
- the present invention is not limited to the above embodiments, and various modifications can be made.
- the place where the arousal level determination device is installed is not limited to the vehicle, and the subject for which the reduction of the arousal level is detected is not limited to the driver of the vehicle. In other words, it can be applied to health equipment and medical equipment.
- an infrared heart rate sensor that detects the amount of reflected infrared light that periodically changes according to the heart rate, and a sensor that detects the blood pressure of the driver. Etc. can be used.
- the level of arousal level detected by the arousal level determination device 1 is divided into three stages, but it may be divided into two stages or four or more stages.
- the present invention relates to a wakefulness determination device and a wakefulness determination method, wherein a physiological index indicating the strength of a state against sleepiness is obtained from a biological signal of an active person, and the physiological index Since the degree of arousal is determined based on this, it is possible to more reliably detect weak sleepiness of an active person.
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2007800446557A CN101547643B (zh) | 2006-12-04 | 2007-12-04 | 觉醒度判定装置以及觉醒度判定方法 |
KR1020097005588A KR101384439B1 (ko) | 2006-12-04 | 2007-12-04 | 각성도 판정 장치 및 각성도 판정 방법 |
EP07850316.6A EP2092889B1 (en) | 2006-12-04 | 2007-12-04 | Device for judging degree of awakening and method for judging degree of awakening |
US12/514,692 US8311621B2 (en) | 2006-12-04 | 2007-12-04 | Device for judging degree of awakening and method for judging degree of awakening |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2006-327379 | 2006-12-04 | ||
JP2006327379 | 2006-12-04 | ||
JP2007140975A JP4697185B2 (ja) | 2006-12-04 | 2007-05-28 | 覚醒度判定装置及び覚醒度判定方法 |
JP2007-140975 | 2007-05-28 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2008069337A1 true WO2008069337A1 (ja) | 2008-06-12 |
Family
ID=39492209
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2007/073742 WO2008069337A1 (ja) | 2006-12-04 | 2007-12-04 | 覚醒度判定装置及び覚醒度判定方法 |
Country Status (6)
Country | Link |
---|---|
US (1) | US8311621B2 (ja) |
EP (1) | EP2092889B1 (ja) |
JP (1) | JP4697185B2 (ja) |
KR (1) | KR101384439B1 (ja) |
CN (1) | CN101547643B (ja) |
WO (1) | WO2008069337A1 (ja) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010012100A (ja) * | 2008-07-04 | 2010-01-21 | Toyota Motor Corp | 眠気検出装置 |
JP2010131061A (ja) * | 2008-12-02 | 2010-06-17 | Toyota Motor Corp | 眠気検出装置 |
JP2010134533A (ja) * | 2008-12-02 | 2010-06-17 | Toyota Motor Corp | 眠気検出装置 |
WO2010140241A1 (ja) * | 2009-06-04 | 2010-12-09 | 富士通株式会社 | 覚醒度判定装置、覚醒度判定方法および覚醒度判定プログラム |
US8140149B2 (en) | 2008-07-04 | 2012-03-20 | Toyota Jidosha Kabushiki Kaisha | Drowsiness detector |
EP2441387A4 (en) * | 2009-06-08 | 2014-12-31 | Univ Nagoya City | DEVICE FOR ASSESSING SOMNOLENCE |
US10473762B2 (en) * | 2016-08-15 | 2019-11-12 | Microsoft Technology Licensing, Llc | Wireless radio module |
Families Citing this family (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9129505B2 (en) | 1995-06-07 | 2015-09-08 | American Vehicular Sciences Llc | Driver fatigue monitoring system and method |
ES2374396T3 (es) * | 2006-09-07 | 2012-02-16 | Telozo Gmbh | Método para deducir y evaluar informaciones cardiovasculares a partir de curvas de la corriente cardíaca, particularmente para aplicaciones telemédicas. |
JP2009039167A (ja) * | 2007-08-06 | 2009-02-26 | Toyota Motor Corp | 眠気判定装置 |
JP5387367B2 (ja) * | 2008-12-01 | 2014-01-15 | 富士通株式会社 | 覚醒度判定装置および覚醒度判定方法 |
JP5299915B2 (ja) * | 2009-07-22 | 2013-09-25 | 株式会社最新松本技研 | 覚醒度合検出装置 |
WO2011046178A1 (ja) * | 2009-10-14 | 2011-04-21 | 株式会社デルタツーリング | 生体状態推定装置、生体状態推定システム及びコンピュータプログラム |
KR101034886B1 (ko) * | 2009-12-30 | 2011-05-17 | 인제대학교 산학협력단 | 짧은 심장박동 주기의 변화를 이용한 졸음상태 판정 시스템 및 방법 |
JP5585648B2 (ja) * | 2010-03-23 | 2014-09-10 | アイシン精機株式会社 | 覚醒度判定装置、覚醒度判定方法及びプログラム |
JP5696501B2 (ja) * | 2011-01-27 | 2015-04-08 | 富士通株式会社 | 覚醒時データ生成装置、覚醒時データ生成方法、覚醒時データ生成プログラム及び覚醒度判定装置 |
TWI434233B (zh) * | 2011-05-17 | 2014-04-11 | Ind Tech Res Inst | 睡意預警方法 |
US8766819B2 (en) * | 2011-06-17 | 2014-07-01 | The Boeing Company | Crew allertness monitoring of biowaves |
TWI478691B (zh) * | 2012-01-06 | 2015-04-01 | Wistron Corp | 睡意偵測方法及其裝置 |
US9815384B2 (en) | 2012-07-09 | 2017-11-14 | Ts Tech Co., Ltd. | Wakefulness-maintenance apparatus |
JP6209396B2 (ja) * | 2013-04-17 | 2017-10-04 | 株式会社デルタツーリング | 運転支援装置及びコンピュータプログラム |
JP6312193B2 (ja) | 2013-10-21 | 2018-04-18 | テイ・エス テック株式会社 | 覚醒装置及びシート |
CN104484796B (zh) | 2014-12-18 | 2018-03-27 | 天津三星通信技术研究有限公司 | 便携式终端及其日程管理方法 |
JP2016182241A (ja) * | 2015-03-26 | 2016-10-20 | パイオニア株式会社 | 眠気算出装置 |
US10328852B2 (en) * | 2015-05-12 | 2019-06-25 | University Of North Dakota | Systems and methods to provide feedback to pilot/operator by utilizing integration of navigation and physiological monitoring |
CN106264449B (zh) * | 2015-06-29 | 2022-01-28 | 松下知识产权经营株式会社 | 人状态推定方法和人状态推定系统 |
CN113509158A (zh) * | 2015-07-22 | 2021-10-19 | 松下电器(美国)知识产权公司 | 清醒度预测方法、清醒度预测装置以及控制对象设备 |
JP6803679B2 (ja) | 2016-04-14 | 2020-12-23 | Joyson Safety Systems Japan株式会社 | バックル及び車載システム |
JP6641642B2 (ja) * | 2016-08-03 | 2020-02-05 | オムロン株式会社 | 生命徴候検出装置 |
JP6791707B2 (ja) * | 2016-09-30 | 2020-11-25 | 東芝情報システム株式会社 | 眠気推定装置及び眠気推定プログラム |
JP7058289B2 (ja) * | 2017-02-10 | 2022-04-21 | ガルデルマ ホールディング ソシエテ アノニム | 痒みの監視及び計測用のシステム及び方法 |
KR102045569B1 (ko) * | 2017-06-08 | 2019-12-02 | 고려대학교 산학협력단 | 조종사 상태의 통합 감시 제어 장치 및 이를 이용한 조종사의 임무 수행 능력 유도 방법 |
JP6937259B2 (ja) * | 2018-03-19 | 2021-09-22 | フォルシアクラリオン・エレクトロニクス株式会社 | 報知装置および報知方法 |
CN108280242A (zh) * | 2018-03-27 | 2018-07-13 | 新日(无锡)发展有限公司 | 一种车内音响智能控制方法及装置 |
WO2019188398A1 (ja) * | 2018-03-30 | 2019-10-03 | ソニーセミコンダクタソリューションズ株式会社 | 情報処理装置、移動装置、および方法、並びにプログラム |
JP7099037B2 (ja) * | 2018-05-07 | 2022-07-12 | オムロン株式会社 | データ処理装置、モニタリングシステム、覚醒システム、データ処理方法、及びデータ処理プログラム |
JP2019069207A (ja) * | 2018-12-26 | 2019-05-09 | パイオニア株式会社 | 眠気算出装置 |
WO2020195168A1 (ja) * | 2019-03-28 | 2020-10-01 | パナソニックIpマネジメント株式会社 | 漫然状態判定装置および漫然状態判定方法 |
TWI749323B (zh) * | 2019-04-30 | 2021-12-11 | 先進光電科技股份有限公司 | 行動載具輔助系統 |
CN110135764A (zh) * | 2019-05-29 | 2019-08-16 | 中国人民解放军海军特色医学中心 | 一种船舶操纵人员作业能力测评系统及方法 |
JP2019205894A (ja) * | 2019-08-08 | 2019-12-05 | パイオニア株式会社 | 眠気算出装置 |
JP2020171734A (ja) * | 2020-07-01 | 2020-10-22 | パイオニア株式会社 | 眠気算出装置 |
CN112128932A (zh) * | 2020-08-26 | 2020-12-25 | 青岛海尔空调器有限总公司 | 空调器及其控制方法 |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01131648A (ja) * | 1987-11-18 | 1989-05-24 | Nippon Denso Co Ltd | 覚醒度判定装置 |
JPH04348759A (ja) * | 1991-05-28 | 1992-12-03 | Matsushita Electric Works Ltd | 覚醒度モニター装置 |
JPH06270711A (ja) | 1993-03-17 | 1994-09-27 | Nissan Motor Co Ltd | 覚醒状態検知装置 |
JPH07231880A (ja) * | 1994-02-24 | 1995-09-05 | Sanyo Electric Co Ltd | ストレス評価方法及び装置 |
US6070098A (en) | 1997-01-11 | 2000-05-30 | Circadian Technologies, Inc. | Method of and apparatus for evaluation and mitigation of microsleep events |
JP2001198113A (ja) | 2000-01-18 | 2001-07-24 | Toyota Central Res & Dev Lab Inc | 疲労度演算装置 |
US20040243013A1 (en) | 2003-05-27 | 2004-12-02 | Taiji Kawachi | Sleepiness level detection device |
US20050148894A1 (en) | 2004-01-06 | 2005-07-07 | Misczynski Dale J. | Method and system for contactless evaluation of fatigue of an operator |
JP2006130046A (ja) | 2004-11-05 | 2006-05-25 | Daikin Ind Ltd | 覚醒度判定装置 |
JP2006158733A (ja) * | 2004-12-08 | 2006-06-22 | Toyota Motor Corp | 覚醒度判定装置及び覚醒度判定方法 |
JP2006247055A (ja) | 2005-03-09 | 2006-09-21 | Toyota Motor Corp | 覚醒度判定装置 |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
LU75977A1 (ja) * | 1976-10-12 | 1978-05-16 | ||
JPS5750097A (en) * | 1980-09-08 | 1982-03-24 | Nissan Motor | Automotive warning device |
US5012226A (en) * | 1990-02-23 | 1991-04-30 | Love Samuel D | Safety alertness monitoring system |
JPH08299443A (ja) * | 1995-05-12 | 1996-11-19 | Seiko Epson Corp | 居眠り防止装置 |
JPH11314534A (ja) * | 1998-05-06 | 1999-11-16 | Nissan Motor Co Ltd | 車両用注意能力低下防止装置 |
KR100646868B1 (ko) * | 2004-12-29 | 2006-11-23 | 삼성전자주식회사 | 피부전도도와 심박 정보를 이용한 홈 제어시스템 및 그 방법 |
JP4867215B2 (ja) | 2005-06-28 | 2012-02-01 | トヨタ自動車株式会社 | 生理・心理状態判定装置、生理・心理状態判定方法、リファレンスデータ生成装置、及びリファレンスデータ生成方法。 |
JP2007195886A (ja) | 2006-01-30 | 2007-08-09 | Toyota Motor Corp | 車両用シート装置 |
JP2007229218A (ja) | 2006-03-01 | 2007-09-13 | Toyota Motor Corp | 覚醒度推定装置及びシステム並びに方法 |
JP4677940B2 (ja) | 2006-03-27 | 2011-04-27 | トヨタ自動車株式会社 | 眠気検出装置 |
-
2007
- 2007-05-28 JP JP2007140975A patent/JP4697185B2/ja not_active Expired - Fee Related
- 2007-12-04 CN CN2007800446557A patent/CN101547643B/zh not_active Expired - Fee Related
- 2007-12-04 US US12/514,692 patent/US8311621B2/en not_active Expired - Fee Related
- 2007-12-04 WO PCT/JP2007/073742 patent/WO2008069337A1/ja active Application Filing
- 2007-12-04 KR KR1020097005588A patent/KR101384439B1/ko active IP Right Grant
- 2007-12-04 EP EP07850316.6A patent/EP2092889B1/en not_active Not-in-force
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01131648A (ja) * | 1987-11-18 | 1989-05-24 | Nippon Denso Co Ltd | 覚醒度判定装置 |
JPH04348759A (ja) * | 1991-05-28 | 1992-12-03 | Matsushita Electric Works Ltd | 覚醒度モニター装置 |
JPH06270711A (ja) | 1993-03-17 | 1994-09-27 | Nissan Motor Co Ltd | 覚醒状態検知装置 |
JPH07231880A (ja) * | 1994-02-24 | 1995-09-05 | Sanyo Electric Co Ltd | ストレス評価方法及び装置 |
US6070098A (en) | 1997-01-11 | 2000-05-30 | Circadian Technologies, Inc. | Method of and apparatus for evaluation and mitigation of microsleep events |
JP2001198113A (ja) | 2000-01-18 | 2001-07-24 | Toyota Central Res & Dev Lab Inc | 疲労度演算装置 |
US20040243013A1 (en) | 2003-05-27 | 2004-12-02 | Taiji Kawachi | Sleepiness level detection device |
US20050148894A1 (en) | 2004-01-06 | 2005-07-07 | Misczynski Dale J. | Method and system for contactless evaluation of fatigue of an operator |
JP2006130046A (ja) | 2004-11-05 | 2006-05-25 | Daikin Ind Ltd | 覚醒度判定装置 |
JP2006158733A (ja) * | 2004-12-08 | 2006-06-22 | Toyota Motor Corp | 覚醒度判定装置及び覚醒度判定方法 |
JP2006247055A (ja) | 2005-03-09 | 2006-09-21 | Toyota Motor Corp | 覚醒度判定装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP2092889A4 |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010012100A (ja) * | 2008-07-04 | 2010-01-21 | Toyota Motor Corp | 眠気検出装置 |
US8140149B2 (en) | 2008-07-04 | 2012-03-20 | Toyota Jidosha Kabushiki Kaisha | Drowsiness detector |
JP2010131061A (ja) * | 2008-12-02 | 2010-06-17 | Toyota Motor Corp | 眠気検出装置 |
JP2010134533A (ja) * | 2008-12-02 | 2010-06-17 | Toyota Motor Corp | 眠気検出装置 |
WO2010140241A1 (ja) * | 2009-06-04 | 2010-12-09 | 富士通株式会社 | 覚醒度判定装置、覚醒度判定方法および覚醒度判定プログラム |
WO2010140273A1 (ja) * | 2009-06-04 | 2010-12-09 | 富士通株式会社 | 覚醒度判定装置、覚醒度判定方法および覚醒度判定プログラム |
JP5360206B2 (ja) * | 2009-06-04 | 2013-12-04 | 富士通株式会社 | 覚醒度判定装置、覚醒度判定方法および覚醒度判定プログラム |
US9700255B2 (en) | 2009-06-04 | 2017-07-11 | Fujitsu Limited | Consciousness level determination apparatus and consciousness level determination method |
EP2441387A4 (en) * | 2009-06-08 | 2014-12-31 | Univ Nagoya City | DEVICE FOR ASSESSING SOMNOLENCE |
US8979761B2 (en) | 2009-06-08 | 2015-03-17 | Nagoya City University | Sleepiness assessment apparatus |
US10473762B2 (en) * | 2016-08-15 | 2019-11-12 | Microsoft Technology Licensing, Llc | Wireless radio module |
Also Published As
Publication number | Publication date |
---|---|
KR20090089841A (ko) | 2009-08-24 |
CN101547643A (zh) | 2009-09-30 |
KR101384439B1 (ko) | 2014-04-14 |
US20100049066A1 (en) | 2010-02-25 |
EP2092889A1 (en) | 2009-08-26 |
JP2008161664A (ja) | 2008-07-17 |
US8311621B2 (en) | 2012-11-13 |
JP4697185B2 (ja) | 2011-06-08 |
EP2092889B1 (en) | 2019-02-27 |
EP2092889A4 (en) | 2012-08-01 |
CN101547643B (zh) | 2012-02-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2008069337A1 (ja) | 覚醒度判定装置及び覚醒度判定方法 | |
De Rosario et al. | Controlled inducement and measurement of drowsiness in a driving simulator | |
KR101259229B1 (ko) | 졸림 판정 장치 | |
KR101023512B1 (ko) | 사람의 상태 검출 장치 | |
JP6209396B2 (ja) | 運転支援装置及びコンピュータプログラム | |
JP4867215B2 (ja) | 生理・心理状態判定装置、生理・心理状態判定方法、リファレンスデータ生成装置、及びリファレンスデータ生成方法。 | |
JP5007975B2 (ja) | 居眠り検知装置 | |
JP4910547B2 (ja) | 眠気判定装置及び眠気判定プログラム | |
US7250029B2 (en) | Human condition evaluation system, computer program, and computer-readable record medium | |
JP2008125801A (ja) | バイオフィードバック装置及びバイオフィードバック方法 | |
JP5390851B2 (ja) | 車両用自律神経機能診断装置、車両用自律神経機能診断方法 | |
JP2009022370A (ja) | 人の状態推定装置 | |
US10596963B2 (en) | Fatigue alarm apparatus and method | |
JP2009028239A (ja) | 居眠り運転検出装置 | |
KR100718941B1 (ko) | 졸음 경고장치 및 졸음 경고방법 | |
JP2006149470A (ja) | 快適感評価方法及び快適感評価装置 | |
JP2005312868A (ja) | 瞬きを用いた覚醒度計測法 | |
JP4822796B2 (ja) | 睡眠計及び睡眠状態判定方法 | |
CN112888366A (zh) | 脑电波解析装置、脑电波解析系统以及脑电波解析程序 | |
Dolezalek et al. | Physiological signal monitoring system to analyze driver attentiveness | |
TWI582727B (zh) | 自適性(adaptive)生理反應之參數預測方法及其預測裝置 | |
Sengupta et al. | Biosignal acquisition system for stress monitoring | |
Bernarding et al. | Time-Resolved in-vehicle drowsiness monitoring using multimodal electrophysiological data | |
Peruzzini et al. | Toward the Definition of a Technological Set-up for Drivers’ Health Status Monitoring | |
Das et al. | Experimentation on detection and analysis of drowsiness and fatigue based on permutation entropy and hurst exponent |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200780044655.7 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 07850316 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020097005588 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12514692 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007850316 Country of ref document: EP |