JP4797646B2 - Arousal level estimation device and arousal level estimation method - Google Patents

Description

  The present invention relates to a wakefulness estimation apparatus and a wakefulness estimation method for estimating a user's wakefulness.

2. Description of the Related Art Conventionally, there is known a biological state determination device that detects the heart rate, the number of body movements, and the respiration rate of a human body, and determines the arousal level based on the detected heart rate, the number of body movements, and the respiration rate. (For example, refer to Patent Document 1).
JP 2005-95408 A

  By the way, a person's physiological state changes with individual differences, the physical condition of the day, etc. Therefore, even in the heart rate, the number of body movements, and the respiration rate detected by the conventional living body state determination device, there is a possibility that a detection error may occur due to the individual difference of each individual to be detected and the detection timing. Therefore, there is a possibility that the accuracy of the determination of the arousal level is lowered. On the other hand, the accuracy of the determination of the arousal level can be improved by increasing the biological information to be detected, but the configuration may be complicated.

  The present invention has been made to solve such problems, and it is a primary object of the present invention to provide a wakefulness estimation apparatus that can estimate a user's wakefulness with high accuracy with a simple configuration.

In order to achieve the above object, one embodiment of the present invention provides:
A respiration waveform detecting means for detecting a respiration waveform of the user;
Frequency power detection means for detecting a power value for each frequency of the respiratory waveform detected by the respiratory waveform detection means;
A reference frequency setting means for setting a reference frequency based on the respiratory waveform when the user's arousal level is high;
Of the power values detected by the frequency power detection means, a power total calculation means for calculating the sum of the power values having a frequency higher than the reference frequency set by the reference frequency setting means;
An arousal level estimation device comprising: arousal level estimation unit that estimates the arousal level of the user based on a total value of the power values calculated by the total power calculation unit.

  According to this aspect, the user's arousal level can be estimated with high accuracy with a simple configuration.

Further, in this aspect, further comprising normalization means for normalizing the power value for each frequency detected by the frequency power detection means,
The power total calculation unit may calculate a total of the power values having a frequency larger than the reference frequency set by the reference frequency setting unit among the power values normalized by the normalization unit. . Thereby, the variation in the respiratory waveform due to individual differences, physical condition differences, and the like of the user can be suppressed.

  Furthermore, in this one aspect, the arousal level estimation means may estimate that the awakening level of the user has increased as the total value of the power values calculated by the total power calculation means increases.

  In this aspect, the arousal level estimation means may estimate the level of the user's arousal level by comparing the total value of the power values calculated by the power total calculation means with a predetermined threshold value. . Thereby, the user can easily recognize the arousal level.

  In this one aspect, the power value reaches a peak when the total value of the power values calculated by the power total calculation unit is equal to or greater than a predetermined value for the reference frequency set by the reference frequency setting unit. Reference frequency update means for updating may be further provided depending on the frequency of the time. Thereby, a more accurate reference frequency can be set, and the arousal level can be estimated with higher accuracy.

In this aspect, the average value calculation unit is further provided that calculates the total value of the power values a plurality of times in a predetermined cycle, and calculates a moving average value of the calculated total values of the power values.
The arousal level estimation unit may estimate the arousal level of the user based on a moving average value of the total value of the power values calculated by the average calculation unit. Thereby, it becomes possible to suppress variation in the total value of the power values, and it is possible to estimate the arousal level with higher accuracy.

In order to achieve the above object, one embodiment of the present invention provides:
A respiration waveform detection step for detecting a respiration waveform of the user;
A frequency power detection step of detecting a power value for each frequency of the respiratory waveform detected by the respiratory waveform detection step;
A reference frequency setting step for setting a reference frequency based on the respiration waveform when the user's arousal level is high;
Of the power values detected by the frequency power detection step, a power total calculation step for calculating the sum of the power values having a frequency higher than the reference frequency set by the reference frequency setting step;
It may be a wakefulness level estimation method including a wakefulness level estimation step of estimating the wakefulness level of the user based on the total value of the power values calculated in the total power calculation step.

  According to the present invention, it is possible to estimate a user's arousal level with high accuracy with a simple configuration.

  Hereinafter, the best mode for carrying out the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a schematic block diagram showing a system configuration of a wakefulness estimation apparatus according to an embodiment of the present invention. The arousal level estimation apparatus 1 according to the present embodiment is mainly configured by a computer main body 2 that performs various arithmetic processes described later. The computer main body 2 executes various processes in accordance with the control and calculation programs, and controls the CPU 1 (Central Processing Unit), the ROM (Read Only Memory) for storing the CPU execution programs, and the calculation results. And a readable / writable RAM (Random Access Memory), a timer, a counter, an input / output interface (I / O), and the like.

  A respiration waveform detection means 3 such as a pressure sensor for detecting a respiration waveform of the user is connected to the computer main body 2. A pressure sensor detects a user's respiration waveform (Drawing 2 (a)), for example by detecting pressure of air which flows out or flows in from a user's nostril. In addition, the respiratory waveform of the user may be detected using a temperature sensor that detects a temperature change of the user's nose, or a pressure sensor such as a piezoelectric element disposed on the seat belt or the seat may be used. That is, any detection method can be applied as long as the user's respiratory waveform can be detected.

  The computer main body 2 includes a frequency power detection unit 2a, a normalization unit 2b, a reference frequency setting unit 2c, a power total calculation unit 2d, an average calculation unit 2e, a wakefulness estimation unit 2f, and a reference frequency update unit 2g described later. It has. Note that the frequency power detection means 2a, normalization means 2b, reference frequency setting means 2c, power total calculation means 2d, wakefulness estimation means 2f, and reference frequency update means 2g described later are stored in the ROM, for example, by the CPU. Realized by executed software. The software may be stored in an arbitrary readable / writable storage medium and read appropriately by the CPU.

The frequency power detection means 2a acquires data D1 of an arbitrary section N (seconds) from the time-series data of the respiratory waveform detected by the respiratory waveform detection means 3. Further, the frequency power detection means 2a performs an FFT analysis (Fast Fourier Transform) on the data D1 and, as shown in FIG. 2B, a power value P _yy (f) (power) for each frequency. Spectrum).

The normalizing unit 2b normalizes the power value P _yy (f) for each frequency detected by the frequency power detection unit 2a to obtain a normalized power value P _yy (f) ′ for each frequency. (FIG. 2 (c)). For example, the normalizing unit 2b normalizes so that the total value of the power values P _yy (f) for each frequency becomes 1.0 (for example, the integrated value of power between frequencies 0 to 1 Hz is 1.0). I do. By this normalization, for example, it is possible to suppress variations in data due to changes in the user's physical condition due to the detection date and time, individual differences among users, and the like, so that it is possible to improve the estimation accuracy of the arousal level described later.

  The reference frequency setting means 2c sets the reference frequency f1 based on the respiratory waveform when the user's arousal level is high. Hereinafter, a method for setting the reference frequency f1 will be described. FIG. 3 is a flowchart illustrating an example of a flow of a method for setting the reference frequency f1.

  The reference frequency setting unit 2c acquires the respiration waveform detected by the respiration waveform detection unit 3 when the user is awake (S100). Note that when the user is highly awake, for example, the user is highly awake and the user is surely awake from sleepiness, and the respiratory waveform at the time of high awakening is experimentally obtained and stored.

Next, the reference frequency setting unit 2c performs an FFT analysis on the acquired respiratory waveform to obtain a power value P _yy 1 (f) for each frequency (S110).

Thereafter, the reference frequency setting unit 2c obtains the maximum frequency f _max that maximizes the power value P _yy 1 (f) among the obtained power values P _yy 1 (f) for each frequency (S120).

Further, the reference frequency setting unit 2c repeats the processes from (S100) to (S120) N times (for example, 3 times) to obtain N maximum frequencies _fmax . The reference frequency setting means 2c calculates an average value f _{max_m} of these N maximum frequencies f _{max and} sets the average value f _{max_m} as the reference frequency f1 (for example, f1 = 1.0) (S130).

The power sum calculation means 2d is a sum of power values having a frequency larger than the reference frequency f1 set by the reference frequency setting means 2c among the power values P _yy (f) ′ normalized by the normalization means 2b (see FIG. 2 (d) is an oblique line portion, and an integral value S1 (t) of a power value having a frequency higher than the reference frequency f1 is calculated. The total value S1 (t) of the calculated power values changes as shown in FIG.

The power total calculation unit 2d simply calculates the total value S1 (t) of the power values having a frequency higher than the reference frequency f1 among the power values P _yy (f) for each frequency obtained by the frequency power detection unit 2a. May be calculated automatically.

The average calculating means 2e is a moving average value S1 (t) of the total value S1 (t) of the calculated power values a plurality of times at a predetermined period (for example, N1 second interval) in a predetermined time T1 (for example, 5 minutes). _{_M} is calculated as shown in FIG. Thereby, since the variation of total value S1 (t) of a power value can be suppressed, the estimation accuracy of the awakening degree mentioned later can be improved.

The arousal level estimation unit 2f estimates the user's arousal level based on the moving average value S1 (t) _{_M} calculated by the average calculation unit 2e. Note that the arousal level estimation means 2f may estimate the user's arousal level based on the total power value S1 (t) calculated by the power total calculation means 2d.

The awakening level estimation unit 2f indicates that the awakening level of the user increases as the moving average value S1 (t) _{_M} (or the total value S1 (t) of power values) calculated by the average value calculation unit 2e increases. presume. On the other hand, wakefulness estimation means 2f, the higher the moving average value S1 (t) _ _M calculated by the average calculation unit 2e is reduced, arousal level of the user is estimated to have decreased.

FIGS. 4 (a) to (c) is, for any three subjects, the relationship between the degree of awakening and the moving average value S1 (t) _ _M, the experimental results in the case of an experiment. In addition, the high arousal level, the middle arousal level, and the low arousal level in FIGS. 4A to 4C are arousal levels estimated with high accuracy by combining other awakening level estimation methods. 4 as shown in (a) to (c), in any of the subjects, the more degree of awakening decreases, the moving average value S1 (t) _ _M is Tokaru be reduced. Also from this experimental result, it can be confirmed that the above-described estimation reliability by the arousal level estimation means 2f is high.

  Further, the arousal level estimation unit 2f may compare the power value total value S1 (t) calculated by the power total calculation unit 2d with a predetermined threshold value to estimate the level of the user's arousal level. Good.

For example, wakefulness estimation means 2f, based on the moving average value S1 (t) _ _M, the level of awareness of the user may be estimated by 1-4 4 stage level. In this case, as shown in FIG. 2 (f), the arousal level estimation means 2f sets the level of the arousal level to 1 when the moving average value S1 (t) _{_M} is C or less, and the moving average value S1 (t) _{_M} is When it is greater than C and less than or equal to B, the arousal level is 2, and when the moving average value S1 (t) _{_M} is greater than B and less than or equal to A, the arousal level is 3, and the moving average value S1 (t) _{_M} is When larger than A, the level of arousal level is estimated as 4. The user can easily recognize the arousal level, assuming that the awakening level is higher as the level of the awakening level is higher. In addition, the value of moving average value S1 (t) _{_M used} as a level of arousal level and a threshold, and the number of levels are set to arbitrary values obtained experimentally.

  Next, the processing flow of the arousal level estimation apparatus 1 according to the present embodiment will be described. FIG. 5 is a flowchart illustrating an example of a processing flow of the arousal level estimation apparatus 1 according to the present embodiment.

  First, the respiratory waveform detection means 3 detects a user's respiratory waveform (S200).

Next, the frequency power detection means 2a performs FFT analysis on the data D1 acquired by the respiratory waveform detection means 3, and obtains a power value P _yy (f) for each frequency (S210).

Thereafter, the normalizing means 2b normalizes the power value P _yy (f) for each frequency detected by the frequency power detecting means 2a (S220).

  The reference frequency setting unit 2c sets the reference frequency f1 based on the respiration waveform when the user's arousal level is high (S230).

Further, the power total calculation means 2d is a sum of power values having a frequency larger than the reference frequency f1 set by the reference frequency setting means 2c among the power values P _yy (f) ′ normalized by the normalization means 2b. The value S1 (t) is calculated (S240).

The average calculating means 2e calculates a moving average value S1 (t) _{_M} of the total value S1 (t) of the calculated power values a plurality of times in the predetermined period N1 at the predetermined time T1 (S250).

Next, the arousal level estimation unit 2f estimates the arousal level of the user based on the moving average value S1 (t) _{_M} calculated by the average calculation unit 2e (S260).

Above, the wakefulness estimating apparatus 1 according to this embodiment, on the basis of the respiratory waveform of the user, the calculated moving averages S1 (t) _ _M of the total value of the power value, the moving average value S1, the calculated ( t) Based on _M , the user's arousal level is estimated. Thereby, a user's arousal level can be estimated with high accuracy with a simple configuration.

  As mentioned above, although the best mode for carrying out the present invention has been described using one embodiment, the present invention is not limited to such one embodiment, and within the scope not departing from the gist of the present invention, Various modifications and substitutions can be made to the above-described embodiment.

  For example, in the above-described embodiment, the reference frequency update unit 2g for updating the reference frequency f1 set by the reference frequency setting unit 2c may be provided.

When it is determined that the value of the moving average value S1 (t) _{_M} (or the total value S1 (t) of the power values) calculated by the average value calculation means 2e is equal to or greater than a predetermined value and the user's arousal level is high The reference frequency update means 2g updates the reference frequency f1 that has already been set with the frequency f2 when the power value P _yy (f) ′ reaches a peak. Thereby, a more accurate reference frequency can be set.

  In the above embodiment, the awakening level estimation device 1 may estimate the awakening level of a passenger of a moving body such as a vehicle, a train, a ship, or an aircraft. In this case, after the occupant is seated in the driver's seat of the vehicle, it is estimated that the occupant is in a state of high arousal because the occupant is in some tension at a predetermined time T2 (for example, about 5 minutes).

  Therefore, the reference frequency setting unit 2c may set the reference frequency f1 by performing the processes (S100) to (S130) during the predetermined time T2 after the occupant is seated in the driver's seat. Thereby, the reference frequency setting means 2c can set the reference frequency f1 more reliably based on the respiration waveform when the user's arousal level is high.

  INDUSTRIAL APPLICABILITY The present invention can be used for a wakefulness estimation device that estimates a user's wakefulness.

It is a schematic block diagram which shows the system configuration | structure of the arousal level estimation apparatus which concerns on one Example of this invention. (A) It is a figure which shows an example of a user's respiration waveform. (B) It is a figure which shows an example of the power value (power spectrum) for every frequency. (C) It is a figure which shows an example of the power value for every frequency by which normalization was performed. (D) It is a figure which shows an example of the total value of the power value of a frequency larger than a reference frequency. (E) It is a figure which shows an example of the change of the total value of the power value by a time series. (F) It is a figure which shows an example of the moving average value of the total value of a power value. It is a flowchart which shows an example of the flow of the setting method of a reference frequency. (A) It is an example of the experimental result which shows the relationship between an arousal level and a moving average value. (B) It is an example of the experimental result which shows the relationship between an arousal level and a moving average value. (C) It is an example of the experimental result which shows the relationship between an arousal level and a moving average value. It is a flowchart which shows an example of the processing flow of the arousal level estimation apparatus which concerns on one Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Arousal level estimation apparatus 2 Computer main body 2a Frequency power detection means 2b Normalization means 2c Reference frequency setting means 2d Power total calculation means 2e Average calculation means 2f Arousal level estimation means 2g Reference frequency update means 3 Respiration waveform detection means

Claims (7)

A respiration waveform detecting means for detecting a respiration waveform of the user;
Frequency power detection means for detecting a power value for each frequency of the respiratory waveform detected by the respiratory waveform detection means;
A reference frequency setting means for setting a reference frequency based on the respiratory waveform when the user's arousal level is high;
Of the power values detected by the frequency power detection means, a power total calculation means for calculating the sum of the power values having a frequency higher than the reference frequency set by the reference frequency setting means;
An arousal level estimation device comprising: an arousal level estimation unit that estimates the arousal level of the user based on a total value of the power values calculated by the total power calculation unit. The arousal level estimation apparatus according to claim 1,
Further comprising normalization means for normalizing the power value for each frequency detected by the frequency power detection means;
The power total calculation means calculates a total of the power values having a frequency larger than the reference frequency set by the reference frequency setting means among the power values normalized by the normalization means. A device for estimating arousal level. The arousal level estimation apparatus according to claim 1 or 2,
The wakefulness level estimation unit estimates that the wakefulness level of the user has increased as the total value of the power values calculated by the total power calculation unit increases. The arousal level estimation apparatus according to claim 3,
The arousal level estimation means compares the total value of the power values calculated by the power total calculation means with a predetermined threshold value, and estimates the level of the user's arousal level. Estimating device. The arousal level estimation apparatus according to any one of claims 1 to 4,
The reference frequency set by the reference frequency setting means is a frequency at which the power value peaks when the total value of the power values calculated by the power total calculation means is a predetermined value or more. A wakefulness level estimation device, further comprising reference frequency update means for updating. The wakefulness estimation device according to any one of claims 1 to 5,
An average calculating means for calculating a total value of the power values a plurality of times in a predetermined cycle, and calculating a moving average value of the calculated total values of the power values a plurality of times;
The wakefulness level estimation unit estimates the wakefulness level of the user based on a moving average value of the total value of the power values calculated by the average calculation unit. A respiration waveform detection step for detecting a respiration waveform of the user;
A frequency power detection step of detecting a power value for each frequency of the respiratory waveform detected by the respiratory waveform detection step;
A reference frequency setting step for setting a reference frequency based on the respiration waveform when the user's arousal level is high;
Of the power values detected by the frequency power detection step, a power total calculation step for calculating the sum of the power values having a frequency higher than the reference frequency set by the reference frequency setting step;
A wakefulness level estimation method comprising: a wakefulness level estimation step of estimating the wakefulness level of the user based on the total value of the power values calculated in the total power calculation step.

Images