JP7230385B2 - Biological information detector - Google Patents

Description

TECHNICAL FIELD The present invention relates to a biological information detection device for detecting biological information such as heartbeat.

Autonomous driving systems currently under development operate under the supervision of the driver below a predetermined index level (for example, NHTSA index level 3 or lower), and the driver is responsible for driving. On the other hand, many academic societies report that automatic driving systems reduce the psychological burden on drivers and reduce their arousal. Therefore, in recent years, development of a system that detects the degree of alertness of the driver and displays a warning or the like according to the result has been advanced.

When measuring the heart rate and breathing rate to detect the driver's arousal level, due to constraints such as "does not interfere with driving" and "must be constantly measured" It is desirable to measure by the method of As non-contact methods, there are generally ultrasonic methods, radio wave methods, and piezoelectric sensor methods. be.

Patent Literature 1 describes a technique of designing a filter circuit using statistical values relating to heartbeats as a method of removing noise components contained in an original signal detected from the human body. This technology uses a filter circuit to remove noise components in the high-frequency and low-frequency ranges from the frequency band near the heartbeat contained in the original signal detected from the human body, and detects biological information contained in the original signal. .

JP 2013-153782 A

However, the technique described in Patent Literature 1 has a problem that it is impossible to narrow down the high frequency region and the low frequency region for removing noise components because the heart rate varies for each individual driver. Therefore, with this technique, it is difficult for the filter circuit to remove noise in the frequency band near the heart rate, and the detection accuracy of biometric information may deteriorate.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a biometric information detection apparatus capable of improving biometric information detection accuracy.

In order to achieve the above object, the invention according to claim 1,
a signal acquisition unit (3) that acquires an original signal containing biological information from a human body;
A biometric information detection device comprising a calculation unit (7) that calculates biometric information from the original signal acquired by the signal acquisition unit,
The calculation part is
While converting the original signal into frequency characteristics,
After setting the delay time to a plurality of times and generating a plurality of delayed signals in which the phase of the original signal is shifted by each of the set delay times, a plurality of noise frequencies are obtained from the original signal and the plurality of delayed signals. Calculate the characteristics,
Biological information is calculated from a plurality of noise-removed frequency characteristics obtained by removing a plurality of noise frequency characteristics from the frequency characteristics of the original signal.

According to this, the biological information detection apparatus converts the original signal into frequency characteristics, and removes the frequency characteristics of noise from the frequency characteristics of the original signal, thereby obtaining frequency characteristics having frequencies corresponding to biological information. If the biometric information is, for example, a heartbeat or a respiration rate, the frequency (Hz) of the biometric information can be multiplied by 60 to calculate the heartbeat rate or respiration rate per minute. In this manner, the biometric information detection apparatus can improve the detection accuracy of biometric information by calculating biometric information based on the difference between the frequency characteristics of the original signal and the frequency characteristics of noise. The frequency characteristic of the noise may be calculated from the original signal using the periodicity of biological information, as will be described later, or may be stored in memory in advance according to the vehicle speed or the like.

It should be noted that the reference numerals in parentheses attached to each component etc. indicate an example of the correspondence relationship between the component etc. and specific components etc. described in the embodiments described later.

BRIEF DESCRIPTION OF THE DRAWINGS It is a block diagram which shows the structure of the biometric information detection apparatus which concerns on 1st Embodiment. It is an explanatory view for explaining a function of a living body information detecting device concerning a 1st embodiment. 4 is a flow chart showing an algorithm for biometric information detection according to the first embodiment; 4 is a graph for explaining an original signal, a delayed signal, and a difference signal between them; 5 is a graph for explaining frequency characteristics of noise; 4 is a graph for explaining frequency characteristics of an original signal; 7 is a graph for explaining frequency characteristics after noise removal; It is a flow chart which shows the algorithm of living body information detection of a 2nd embodiment. 7 is a graph for explaining frequency characteristics after noise removal; It is a flow chart which shows an algorithm of living body information detection of a 3rd embodiment. FIG. 12 is a flow chart showing an algorithm for biometric information detection according to the fourth embodiment; FIG. 2 is a flow chart showing a general biometric information detection algorithm. It is a graph which shows the original signal output from a biosensor. 4 is a graph showing a signal obtained by passing an original signal through a filter circuit; 4 is a graph showing frequency characteristics of a signal obtained by passing an original signal through a filter circuit; FIG. 16 is an explanatory diagram for explaining a method of calculating a heart rate from the frequency characteristics of FIG. 15;

(General biometric information detection algorithm)
A biological information detection device is configured to process a biological signal (hereinafter referred to as an original signal) output from a biological sensor provided on the human body and detect biological information such as heartbeat or respiratory rate. . The biometric information detection device is composed of a well-known microcomputer having a CPU, ROM, RAM, I/O, and the like.

Before describing biometric information detection devices according to a plurality of embodiments of the present invention, first, the biometric information detection algorithm performed by a general biometric information detection device will be described with reference to the flow chart of FIG. 12 and FIGS. Description will be made with reference to the graph. Here, heartbeat will be described as an example of biological information. In addition, in the following description, the biological information detection device may be simply referred to as a "detection device".

In a general biometric information detection algorithm, first, in step S1 of the flowchart in FIG. 12, the detection device acquires the original signal output from the biosensor. As shown in FIG. 13, this original signal is a time waveform containing heartbeats and noise.

Next, in step S2, the detector passes the original signal through a predetermined filter circuit. This filter circuit is a so-called frequency filter that removes noise components in high-frequency and low-frequency regions from the frequency band near the frequency of the biometric information and passes only the frequency band near the frequency of the biometric information. As a result, as shown in FIG. 14, the original signal is narrowed down to a frequency band near the frequency of the biological information.

Subsequently, in step S3, the detection device performs processing such as Fourier transform on the time waveform that has passed through the filter circuit to convert it into frequency characteristics. Thereby, as shown in FIG. 15, it is converted into a frequency characteristic. Note that the frequency characteristic indicates the intensity for each of a plurality of frequencies included in the time waveform.

Next, in step S4, the detection device extracts the frequency at which the intensity (dB) is the largest among the frequency characteristics. In FIG. 16, that portion is indicated by P. As shown in FIG.
Subsequently, in step S5, the detection device multiplies the frequency (Hz) at the point marked with P by 60 to calculate the heart rate, that is, the number of beats per minute.

In the general biological information detection algorithm described above, if the temporal waveform that has passed through the filter circuit in the process of step S2 contains almost no noise, the frequency extracted in the process of step S4 corresponds to the heart rate. It will be

However, in a general biometric information detection algorithm, the time waveform that has passed through the filter circuit in the process of step S2 may contain noise in a frequency band near the heartbeat. Therefore, in the frequency characteristic after the processing in step S3, a frequency different from the frequency corresponding to the heart rate may be extracted. In that case, the frequency extracted in the process of step S4 may differ from the frequency corresponding to the heart rate, and the accuracy of the heart rate calculated in step S5 may deteriorate.

The general algorithm for detection of biometric information has been described above. On the other hand, biometric information detection devices according to a plurality of embodiments of the present invention described below are capable of improving biometric information detection accuracy.

(First embodiment)
Next, a first embodiment of the invention will be described with reference to FIGS. 1 to 7. FIG. A biological information detection device 1 according to the first embodiment is mounted, for example, in an automatic driving vehicle, and is used to detect biological information such as the driver's heartbeat.

As shown in FIG. 1, the biometric information detection device 1 is configured to detect biometric information by acquiring an original signal output from a biosensor 2 and processing the original signal. In the present embodiment, the biometric information detecting device 1 detects a heartbeat as an example of biometric information. Note that the heartbeat is a periodic signal that varies within a predetermined frequency range.

The biosensor 2 is a sensor that is provided in contact or non-contact with the human body and that outputs an original signal including biometric information such as heartbeat. Examples of such a biosensor 2 include an ultrasonic sensor, a radio wave sensor, a piezoelectric sensor, and the like.

The biological information detecting device 1 is composed of a well-known microcomputer having a CPU, ROM, RAM, I/O, etc., and is configured to process signals according to a program stored in the ROM or the like. The biological information detection device 1 includes a signal acquisition unit 3, a delay signal generation unit 4, a noise frequency characteristic calculation unit 5, a biological signal frequency characteristic calculation unit 6, and the like as functional units that process signals.

The signal acquisition unit 3 has a function of acquiring the original signal output from the biosensor 2 . The delay signal generator 4 , noise frequency characteristic calculator 5 , and biological signal frequency characteristic calculator 6 constitute a calculator 7 for calculating biological information from the original signal acquired by the signal acquirer 3 .

The functions of the delay signal generator 4, the noise frequency characteristic calculator 5, and the biological signal frequency characteristic calculator 6, which constitute the calculator 7, will be described with reference to FIGS. 1 and 2. FIG. In addition, in FIG. 2, the original signal acquired by the signal acquisition part 3 is described as a "time waveform."

The delay signal generator 4 has a function of generating a delay signal in which the phase of the original signal is shifted. The delay signal generator 4 preferably sets the delay time to a time obtained by multiplying one heartbeat cycle by a natural number, or a time close thereto.

The noise frequency characteristic calculator 5 has a function of generating a time waveform of noise by subtracting the original signal and the delayed signal. Further, the noise frequency characteristic calculator 5 has a function of calculating the frequency characteristic of noise by performing processing such as Fourier transform on the time waveform of the noise.

The biological signal frequency characteristic calculator 6 has a function of calculating the frequency characteristic of the original signal by performing processing such as Fourier transform on the original signal. The biosignal frequency characteristic calculator 6 has a function of calculating biometric information from the extracted value of the frequency characteristic after noise removal, which is obtained by removing the frequency characteristic of noise from the frequency characteristic of the original signal.

Next, the biometric information detection algorithm performed by the biometric information detection apparatus 1 of the first embodiment will be described with reference to the flowchart of FIG. 3 and the graphs of FIGS. 4 to 7. FIG.

First, the signal acquisition unit 3 acquires the original signal output from the biosensor 2 in step S10 of the flowchart of FIG. This original signal is a time waveform containing heartbeats and noise. FIG. 4A shows the time waveform of the original signal separately for the heartbeat signal and the noise signal included in the original signal for explanation. That is, the thick line H indicates the heartbeat signal contained in the original signal, and the thin line N indicates the noise signal. However, the original signal acquired by the signal acquisition unit 3 is a single time waveform containing a heartbeat signal and a noise signal, as shown in the graph of FIG. is obtained as

Next, in step S20, the delay signal generator 4 generates a delay signal by shifting the phase of the original signal. This original signal is also a time waveform containing a heartbeat signal and a noise signal. FIG. 4B shows a delayed signal that is shifted by one heartbeat period included in the original signal. Thus, the delay time for shifting the phase of the original signal is preferably one cycle of heartbeat, a time obtained by multiplying one cycle of heartbeat by a natural number, or a time close thereto.

Specifically, the delay signal generator 4 sets the plurality of delay times between 0.5 and 1 second. In general, a normal human heart rate is likely to be between 120 beats/minute (ie 2 Hz) and 60 beats/minute (ie 1 Hz). This range is statistically within the standard deviation of 3σ for heart rate. Therefore, by setting a plurality of delay times between 0.5 and 1 second, the delay signal generator 4 can make the phase shift between the delay signal and the original signal equal to or close to the heartbeat cycle. It is possible.

Subsequently, in step S30, the calculator 7 generates a signal obtained by subtracting the original signal and the delayed signal. FIG. 4(C) shows a time waveform of a signal obtained by subtracting the original signal shown in FIG. 4(A) and the delayed signal shown in FIG. 4(B), that is, noise. If the phase shift between the delayed signal and the original signal is the same as the heartbeat cycle, the time waveform of this noise will be the time waveform of only noise. Note that if the phase shift between the delayed signal and the original signal is different from the heartbeat period, the frequency due to the shift between the heartbeat signals will be added to the frequency characteristics of only the noise. However, when the phase shift between the delayed signal and the original signal is close to the heartbeat cycle, the signal strength of the frequency added due to the shift between the heartbeat signals is small with respect to the frequency characteristics of only noise.

Subsequently, in step S40, the noise frequency characteristic calculator 5 calculates the frequency characteristic of the noise by performing processing such as Fourier transform on the signal obtained by subtracting the original signal and the delayed signal, that is, the time waveform of the noise. FIG. 5 shows the frequency characteristics of noise calculated by performing processing such as Fourier transform on the time waveform of noise shown in FIG. 4(C).

On the other hand, in step S50, the calculation unit 7 calculates the frequency characteristics of the original signal by performing processing such as Fourier transform on the original signal. FIG. 6 shows the frequency characteristics of the original signal calculated by performing processing such as Fourier transform on the time waveform of the original signal shown in FIG. 4(A).

Next, in step S60, the biological signal frequency characteristic calculator 6 calculates the frequency characteristic by removing the frequency characteristic of noise from the frequency characteristic of the original signal. FIG. 7 shows frequency characteristics calculated by removing the noise frequency characteristics shown in FIG. 5 from the frequency characteristics of the original signal shown in FIG. This frequency characteristic is referred to as the frequency characteristic after noise removal.
Then, the biosignal frequency characteristic calculator 6 extracts the frequency at which the strength is the highest among the frequency characteristics after the noise removal. In FIG. 7, that portion is indicated by P. As shown in FIG.

Subsequently, in step S70, the calculation unit 7 multiplies the frequency (Hz) at the point marked with P by 60 to calculate the heart rate, that is, the number of beats per minute.

The biological information detection device 1 of the first embodiment described above has the following effects.
(1) In the first embodiment, the biological information detection device 1 converts the original signal acquired by the signal acquisition unit 3 into frequency characteristics, and removes the frequency characteristics of noise from the frequency characteristics of the original signal. That is, both the time waveform of the original signal and the time waveform of the noise are converted into frequency characteristics, and the difference between the two frequency characteristics is calculated to obtain the frequency characteristics after noise removal having the frequency of biological information. be. For example, if the biological information is a heartbeat, the frequency (Hz) of the biological information can be multiplied by 60 to calculate the heart rate per minute. In this manner, the biometric information detection device 1 can improve the detection accuracy of biometric information by calculating the biometric information based on the difference between the frequency characteristics of the original signal and the frequency characteristics of the noise.

(2) In the first embodiment, the computing unit 7 generates a delayed signal obtained by shifting the phase of the original signal by a predetermined delay time, converts the difference between the delayed signal and the original signal into a frequency characteristic, and converts the noise into noise. frequency characteristics.
According to this, biological information such as heartbeat has a predetermined periodicity. Therefore, by setting the delay time for generating the delayed signal to the period of the biological information multiplied by a natural number or to something close to it, the signal obtained by subtracting the delayed signal from the original signal is the time waveform of the noise. Become. Therefore, by removing the frequency characteristic of the noise obtained by converting the time waveform of the noise into the frequency characteristic from the frequency characteristic of the original signal, it is possible to obtain the frequency characteristic after removing the noise having the frequency of the biological information.

(3) In the first embodiment, the biological information calculated by the calculator 7 is a periodic signal that changes within a predetermined frequency range. In the first embodiment, the biometric information detected by the biometric information detection device 1 is described as heartbeat, but other periodic biometric information includes, for example, respiration rate and pulse rate. In that case, the delay time for generating the delay signal is set within a general cycle range for the biological information. It should be noted that, as a general range of cycles related to biometric information, for example, it is possible to set the range within a standard deviation of 3σ in terms of statistics of cycles related to biometric information.

(Second embodiment)
A second embodiment will be described. 2nd Embodiment changes a part of algorithm of detection of biometric information with respect to 1st Embodiment, Since it is the same as that of 1st Embodiment about others, it differs from 1st Embodiment. Only about

The biometric information detection algorithm performed by the biometric information detection device 1 of the second embodiment will be described with reference to the flowchart of FIG. It should be noted that in the flowcharts described in the drawings referred to in the multiple embodiments described below, the steps that are substantially the same as those in the flowchart of FIG. attached.

First, in step S10 of the flowchart of FIG. 8, the signal acquisition unit 3 acquires the original signal output from the biosensor 2, as in the first embodiment. This original signal is a time waveform containing heartbeats and noise.

Next, in step S21, the delay signal generator 4 generates a plurality of delay signals by shifting the phase of the original signal. In the second embodiment, the delay signal generator 4 sets a plurality of delay times for shifting the phase of the original signal, and generates a plurality of delay signals. At that time, the delay signal generator 4 sets the plurality of delay times between 0.5 and 1 second, as in the first embodiment. This makes it possible to set the phase shift between the delayed signal and the original signal to the heartbeat cycle or close to it.

Subsequently, in step S31, the computing unit 7 generates a signal obtained by subtracting the original signal and the delayed signal, that is, a temporal waveform of noise. Since the delay signal generator 4 generates a plurality of delay signals in step S21 described above, the calculator 7 generates a plurality of time waveforms of noise in step S31.

Subsequently, in step S41, the noise frequency characteristic calculator 5 calculates a plurality of noise frequency characteristics by performing processing such as Fourier transform on each of the plurality of time waveforms of noise.

On the other hand, in step S50, the calculation unit 7 calculates the frequency characteristics of the original signal by performing processing such as Fourier transform on the original signal as in the first embodiment.

Then, in step S61, the biological signal frequency characteristic calculator 6 removes the plurality of noise frequency characteristics calculated in step S41 from the frequency characteristics of the original signal, and calculates the plurality of frequency characteristics after noise removal. Then, in each of the plurality of frequency characteristics after noise removal, frequencies related to heartbeats are extracted. A value extracted from the frequency characteristic after noise removal in this process is referred to as an extracted value. The extracted value may be the peak frequency with the largest strength (dB) in the frequency characteristics after noise removal, or may be the frequency band slightly shifted from the peak.

Next, in step S65, the calculation unit 7 calculates a representative value from among a plurality of extraction values extracted for each of the plurality of frequency characteristics after noise removal. In the second embodiment, as an example of a method of calculating the representative value, the ratio of the "intensity of the extracted value" to the "average value of the intensity of each frequency excluding the extracted value" among the frequency characteristics after noise removal is the highest. Calculate from the largest.

In FIG. 9, in the frequency characteristics after predetermined noise removal, the strength of the extracted value is indicated by B1, and the average value of the strength of each frequency excluding the extracted value is indicated by B2. The calculation unit 7 calculates the values of B1/B2 for each of the frequency characteristics after noise removal, and extracts the one with the largest B1/B2 value. Then, the extracted value of the frequency characteristic after noise removal is set as a representative value.

Subsequently, in step S70, the calculation unit 7 calculates the heart rate by multiplying the frequency of the extraction value, which is the representative value, by 60. FIG.

The biological information detection device 1 of the second embodiment described above has the following effects.
(1) In the second embodiment, the computing unit 7 generates a plurality of delayed signals by setting the delay time for shifting the phase of the original signal to a plurality of times, and then calculates the frequency characteristics of the plurality of noises. do. Then, the calculation unit 7 calculates biometric information from representative values of the extraction values of the plurality of noise-removed frequency characteristics obtained by removing the frequency characteristics of a plurality of noises from the frequency characteristics of the original signal.

According to this, by setting the delay time to a plurality of times, it is possible to calculate the biometric information even when the periodicity of the biometric information changes. Therefore, the biometric information detecting apparatus 1 can improve robustness against changes in the periodicity of biometric information for each individual, or changes in the periodicity of biometric information due to changes in physical conditions of individuals.

(2) In the second embodiment, the computing unit 7 extracts a representative value for calculating biological information from each of the extracted values of the frequency characteristics after noise removal. The representative value is the extraction value of the frequency characteristics after noise removal that has the largest ratio of the "intensity of the extraction value" to the "average value of the intensity of each frequency excluding the extraction value" among the multiple frequency characteristics after noise removal. It is said that

According to this, the representative value for calculating the biometric information is the most reliable value among the extracted values of the respective frequency characteristics after noise removal. Therefore, with this method, it is possible to improve the detection accuracy of biometric information even when there are variations in the extracted values of the frequency characteristics after noise removal.

(3) In the second embodiment, the calculator 7 sets a plurality of delay times between 0.5 and 1 second.
As a result, it is possible to set the delay time to a plurality of times within the range of a general human heartbeat cycle, thereby making the phase shift between the delayed signal and the original signal equal to or close to the heartbeat cycle. be. Therefore, the signal strength of the frequency added due to the deviation between the heartbeat signals is small with respect to the frequency characteristic of only noise. Therefore, the biometric information detecting device 1 can obtain the extraction value of the frequency characteristic after removing the noise as the value related to the heartbeat, so that the detection accuracy can be improved.

(Third embodiment)
A third embodiment will be described. 3rd Embodiment changes a part of algorithm of the biometric information detection demonstrated in 2nd Embodiment, Since it is the same as that of 2nd Embodiment about others, only a part different from 2nd Embodiment explain.

The biometric information detection algorithm of the third embodiment is shown in the flow chart of FIG. Also in the third embodiment, in step S66, the calculation unit 7 calculates a representative value from among a plurality of extraction values extracted from each of the plurality of frequency characteristics after noise removal. However, in the third embodiment, as an example of a method of calculating a representative value, the average value of a plurality of extraction values extracted from each of a plurality of frequency characteristics after noise removal is used as a representative value for calculating biometric information. do.

As a result, in the third embodiment, by using the average value of the extracted values of each of the plurality of frequency characteristics as the representative value, even when there is variation in the extracted values of each of the plurality of frequency characteristics, the detection accuracy of biometric information can be improved. can be enhanced.

(Fourth embodiment)
A fourth embodiment will be described. The fourth embodiment is obtained by partially changing the biometric information detection algorithm described in the first to third embodiments, etc., and is otherwise the same as the first to third embodiments. Only parts different from the third embodiment will be described.

The biometric information detection algorithm of the fourth embodiment is shown in the flow chart of FIG. In the fourth embodiment, the original signal is passed through a predetermined filter circuit in step S15. This filter circuit is a band-pass filter that attenuates noise components in high-frequency and low-frequency regions more than the frequency band near the frequency of the biological information and passes only the frequency band near the frequency of the biological information. As a result, the original signal is narrowed down to a frequency band near the frequency of the biological information.

In step S20, the delay signal generator 4 generates a delay signal by shifting the phase of the original signal passed through the filter circuit by a predetermined delay time. In step S30, the noise frequency characteristic calculator 5 calculates the frequency characteristic of noise by performing processing such as Fourier transform on the signal obtained by subtracting the original signal passed through the filter circuit and the delayed signal. In step S50, the calculation unit 7 calculates the frequency characteristics of the original signal by performing processing such as Fourier transform on the signal passed through the filter circuit.

In the fourth embodiment described above, by narrowing down the frequency band of the original signal to a frequency band near the frequency of the biometric information using the filter circuit, the biometric information detection accuracy can be improved.

(Fifth embodiment)
A fifth embodiment will be described. In the above-described first embodiment and the like, the biological information detecting device 1 detects heartbeats as an example of biological information. On the other hand, in the fifth embodiment, the biological information detecting device 1 detects the respiratory rate as an example of biological information. Note that the respiratory rate is also a periodic signal that varies within a predetermined frequency range.

Also in the fifth embodiment, the delay signal generator 4 of the biological information detection device 1 has a function of generating a delay signal in which the phase of the original signal is shifted. As described above, the delay signal generator 4 preferably sets the delay time for shifting the phase of the original signal to the time obtained by multiplying one respiratory cycle by a natural number, or a time close thereto.

In the fifth embodiment, the delay signal generator 4 sets the plurality of delay times between 5 and 3 seconds. In general, the normal breathing rate of a person is likely to be 12 to 20 breaths/minute. This range is statistically within a standard deviation of 3σ for respiratory rate. Therefore, by setting a plurality of delay times between 5 and 3 seconds, the delayed signal generation unit 4 can set the phase shift between the delayed signal and the original signal to be the respiratory cycle or close to it. be. Therefore, the signal strength of the frequency added due to the difference between the respiratory signals is small with respect to the frequency characteristic of only noise. Therefore, the biological information detection device 1 can obtain the extraction value of the frequency characteristic after noise removal as the value related to respiration, so that the detection accuracy can be improved.

(Sixth embodiment)
A sixth embodiment will be described. In the sixth embodiment, the delay signal generation unit 4 included in the biological information detection device 1 generates delay signals based on delay times stored in advance in a memory.

When the driver to be measured by the biological information detection device 1 is the same person, the heartbeat, respiration rate, and the like are within a certain range. Therefore, the calculation unit 7 sets the delay time based on the delay time set in the past and stored in the memory. Thereby, the biological information detection device 1 can reduce the amount of calculation of the calculation unit 7 and shorten the processing time.

(Seventh embodiment)
A seventh embodiment will be described. In the first to sixth embodiments and the like described above, the biometric information detection device 1 is described as calculating the frequency characteristics of noise from the original signal acquired by the signal acquisition unit 3 using the periodicity of the biometric information. On the other hand, in the seventh embodiment, the biometric information detecting apparatus 1 detects biometric information using statistical values of frequency characteristics of noise stored in advance in a memory.

When the biological information detection device 1 is mounted on a vehicle or the like, the frequency characteristics of noise may vary depending on the vehicle speed or the like. Therefore, in the seventh embodiment, the computing unit 7 of the biological information detection device 1 uses the statistical values of the noise frequency characteristics calculated in the past and stored in the memory as the frequency characteristics of the noise for calculating the biological information. used as Thereby, the biological information detection device 1 can reduce the amount of calculation of the calculation unit 7 and shorten the processing time.

(Eighth embodiment)
An eighth embodiment will be described. In the above-described second embodiment and the like, the biological information detecting device 1 calculates the frequency characteristics of a plurality of noises by setting the delay time for shifting the phase of the original signal to a plurality of times. Then, the computing unit 7 calculates the biometric information from the representative value of each of the extracted values of the frequency characteristics after removing the noise obtained by removing the frequency characteristics of the noise from the frequency characteristics of the original signal.

However, if there is a large variation in extracted values in frequency characteristics after noise removal, biometric information calculated from representative values may be less reliable.

Therefore, in the eighth embodiment, the following processing is added to step S65 of the flowchart of FIG. 8 referred to in the description of the second embodiment. That is, if the ratio of the "intensity of the extracted value" to the "average value of the intensity of each frequency excluding the extracted value" in each of the plurality of frequency characteristics after noise removal is lower than a predetermined threshold, the calculation unit 7 determines the representative The biometric information calculated from the value is determined to be an abnormal value.

Specifically, referring again to FIG. 9 referred to in the description of the second embodiment, the calculation unit 7 calculates values of B1/B2 in a plurality of frequency characteristics after noise removal. The one with the largest B2 value is extracted. At that time, if the values of B1/B2 calculated in the frequency characteristics after noise removal are all lower than a predetermined threshold value, the calculation unit 7 determines that the biological information calculated from the representative value is an abnormal value. do. It should be noted that the predetermined threshold value is appropriately set through experiments or the like so that the reliability of the biological information output from the biological information detection device 1 can be obtained. Thereby, the biometric information detecting device 1 can prevent the biometric information having low reliability from being output.

(Other embodiments)
The present invention is not limited to the above-described embodiments, and can be appropriately modified within the scope of the claims. Moreover, the above-described embodiments are not unrelated to each other, and can be appropriately combined unless the combination is clearly impossible. Further, in each of the above-described embodiments, it goes without saying that the elements constituting the embodiment are not necessarily essential, unless it is explicitly stated that they are essential, or they are clearly considered essential in principle. stomach. In addition, in each of the above-described embodiments, when numerical values such as the number, numerical value, amount, range, etc. of the constituent elements of the embodiment are mentioned, when it is explicitly stated that they are particularly essential, and when they are clearly limited to a specific number in principle It is not limited to that specific number, except when In addition, in each of the above embodiments, when referring to the shape, positional relationship, etc. of the constituent elements, the shape, It is not limited to the positional relationship or the like.

For example, in each of the embodiments described above, the biological information detection device 1 has been described as being mounted on a vehicle, but it is not limited to this. The biological information detection device 1 may be used for medical purposes, for example.

1 biological information detection device 3 signal acquisition unit 7 calculation unit

Claims (9)

a signal acquisition unit (3) that acquires an original signal containing biological information from a human body;
A biometric information detection device comprising a calculation unit (7) that calculates biometric information from the original signal acquired by the signal acquisition unit,
The calculation unit is
While converting the original signal into frequency characteristics,
After setting the delay time to a plurality of times and generating a plurality of delayed signals in which the phase of the original signal is shifted by each of the set delay times, a plurality of noise frequencies are obtained from the original signal and the plurality of delayed signals. Calculate the characteristics,
A biometric information detection apparatus that calculates biometric information from a plurality of frequency characteristics after noise removal obtained by removing frequency characteristics of a plurality of noises from the frequency characteristics of an original signal. wherein the calculating unit uses an average value of each of a plurality of noise-removed frequency characteristics obtained by removing a plurality of noise frequency characteristics from the frequency characteristics of the original signal as a representative value for calculating the biological information. Item 1. The biological information detection device according to item 1 . The computing unit extracts a representative value for calculating biological information from each of the extracted values of a plurality of noise-removed frequency characteristics obtained by removing the frequency characteristics of a plurality of noises from the frequency characteristics of the original signal,
The representative value is the extracted value of the frequency characteristic after noise removal, which has the highest ratio of the "intensity of the extracted value" to the "average value of the intensity of each frequency excluding the extracted value" among the multiple frequency characteristics after noise removal. The biological information detection device according to claim 1 , wherein If the ratio of the "intensity of the extracted value" to the "average value of the intensity of each frequency excluding the extracted value" in multiple frequency characteristics after noise removal is lower than a predetermined threshold, the calculated biological information is an abnormal value. 4. The biological information detecting device according to any one of claims 1 to 3 , which determines that there is. The biological information calculated by the computing unit is a periodic signal that changes within a predetermined frequency range,
5. The biometric information detecting device according to claim 1 , wherein said delay time is set within a general cycle range for said biometric information. 6. The biometric information detecting device according to claim 5 , wherein when the biometric information calculated by said calculation unit is a heartbeat, said delay time is set between 0.5 and 1 second. 6. The biological information detecting device according to claim 5 , wherein the delay time is set between 3 and 5 seconds when the biological information calculated by the computing unit is a respiratory rate. 8. The biological information according to any one of claims 1 to 7 , wherein said calculator sets said delay time for generating a delayed signal by shifting the phase of said original signal based on a time set in the past. detection device. The calculation unit is
converting a signal passed through a filter circuit that passes only a frequency band near the frequency of biological information into a frequency characteristic to obtain the frequency characteristic of the original signal;
generating a delayed signal by shifting the phase of the original signal passed through the filter circuit by a predetermined delay time;
9. The biometric information detection device according to claim 1 , wherein a signal obtained by subtracting a signal obtained by passing an original signal through said filter circuit and a delayed signal is converted into a frequency characteristic and used as a frequency characteristic of noise.

Images