JP3098843B2 - In-vehicle heart rate detector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technology for detecting a heart rate of a passenger of a moving body.

[0002]

2. Description of the Related Art Conventionally, as a technique for detecting a passenger's heartbeat, a wristwatch-type heartbeat detection that can be worn by a band on a part to be measured as disclosed in Japanese Utility Model Laid-Open No. 2-213325, for example, has been proposed. There was equipment.

[0003]

However, in the above-mentioned conventional method, when detecting a heartbeat, the main body must be fixed to the arm with a band, which is very troublesome for the user.

When a heartbeat is detected in a non-contact manner in a mobile body such as an automobile, information which does not exist when the heartbeat is detected in a bed, such as knocking or unevenness of a road surface, is generated. However, there is a problem that information different from the heartbeat is detected as noise. At this time, if knocking or irregularities on the road surface are regularly maintained at a period close to the heartbeat, this is erroneously calculated as a heart rate.

An object of the present invention is to solve the above-mentioned problem, and a first object of the present invention is to improve heartbeat detection accuracy by removing noise components other than heartbeat information.

A second object is to calculate an accurate heart rate even when the noise component cannot be completely removed.

[0007]

According to the present invention, there is provided an on-vehicle heartbeat detecting device according to the present invention, comprising: a biological information detecting means provided on a seat in a vehicle; and a position where the biological information detecting means is provided. And a subtraction unit that subtracts an output signal of the vibration sensing unit from an output signal of the biological information detection unit, and a subtraction unit that outputs an output signal from the subtraction unit. Heart rate information detecting means for detecting heart rate information, wherein the heart rate information detecting means calculates an auto-correlation coefficient for an input signal; and a maximum point of the auto-correlation coefficient is the auto-correlation coefficient. Maximal points that satisfy the condition greater than the threshold function value set for
1 / integer of the cycle corresponding to the element indicating the maximum value of the cases
It has basic period calculating means for calculating a basic period for another element of the set of local maximum points present in the vicinity, and calculating means for obtaining a cycle per unit time from the basic period.

[0008]

[0009]

With the above arrangement, the living body information detecting means having the sensor section disposed on the seat detects a pressure change occurring at the contact surface between the passenger and the seat and converts the pressure change into an electric signal. This pressure change includes that caused by the heartbeat of the occupant. The vibration sensing means converts the vibration received by the vehicle body into an electric signal and outputs it. The subtraction means subtracts the output signal of the vibration sensing means from the output signal of the biological information detection means. The heart rate information detecting means detects heart rate information from the fundamental frequency of the electric signal output from the subtracting means. Then, in the heartbeat information detecting means, the autocorrelation coefficient calculating means calculates an autocorrelation coefficient for the input signal. The basic period calculating means sets a threshold function for the autocorrelation coefficient,
The maximum point set that satisfies the condition greater than the threshold function value
In the vicinity of an integer fraction of the period corresponding to the element indicating the maximum value
The basic period of the input signal is calculated for the other elements of the maximum point set that exist . The calculating means calculates the number of cycles per unit time from the obtained basic cycle.

[0010]

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram of a first embodiment of the present invention. The biological information detecting means 1 includes a piezoelectric element 2 as a sensor unit, and a filter circuit 3, an amplifier circuit 4, and a smoothing circuit 5 as a circuit unit. Similarly, the vibration sensing means 6 includes the piezoelectric element 7,
It comprises a filter circuit 8, an amplifier circuit 9, and a smoothing circuit 10. The biological information detecting means 1 and the vibration sensing means 6 are both connected to the subtracting means 11. The subtraction means 11 is connected to the heartbeat information detection means 12.

The piezoelectric element 2 is buried, for example, in a portion of the backrest of the seat near the occupant's heart. The piezoelectric element 7 is disposed, for example, near a mounting bracket of a seat. As the piezoelectric element 2 and the piezoelectric element 7, polyvinylidene fluoride or the like processed into a thin film is used.

Next, the operation of the first embodiment will be described. Piezoelectric element 2 in biological information detecting means 1
Converts various pressure changes applied to the buried portion, such as the occupant's heartbeat, into electrical signals. Next, the filter circuit 3 filters the signal in the frequency band corresponding to the heartbeat. In the case of automobiles, at least 8
Signals with a frequency of 10 Hz or more are cut off. The amplifier circuit 4 amplifies the electric signal that has passed through the filter circuit 3. The smoothing circuit 5 rectifies and integrates the amplified signal.

In the vibration sensing means 6, the piezoelectric element 7 converts vibration other than the occupant's biological information into an electric signal. The filter circuit 8, the amplifier circuit 9, and the smoothing circuit 10 operate similarly to the filter circuit 3, the amplifier circuit 4, and the smoothing circuit 5, respectively.

The subtraction means 11 subtracts the signal output from the vibration sensing means 6 from the signal output from the biological information detection means 1. The body overlaps with the piezoelectric element 2 in addition to the biological information, whereby only the biological information can be extracted. FIG.
Is a graph showing that noise other than biological information is removed by subtraction processing.

The heart rate information detecting means 12 detects the heart rate information contained in the signal output from the subtracting means 11. Since the signal output by the subtraction means 11 is a heartbeat signal in this embodiment, the heart rate per minute can be calculated if the unit time is one minute.

Next, a second embodiment of the present invention will be described. FIG. 3 is a block diagram showing the configuration of the second embodiment. The heart rate information detecting means 12 includes an autocorrelation coefficient calculating means 13, a basic cycle calculating means 14, and a calculating means 15. The autocorrelation coefficient calculation means 13 can be realized by an A / D conversion circuit and a microcomputer. Similarly, the calculation / arithmetic processing in the basic cycle calculating means 14 and the arithmetic means 15 can be incorporated as microcomputer software.

FIG. 4 is a flowchart showing an algorithm for calculating a fundamental frequency using an autocorrelation coefficient. This implementation
In the example, discrete data is sampled at a sampling rate of 20 Hz.
It is assumed that it is obtained from the A / D conversion circuit 6, and the obtained discrete
Data is represented as data (j). Here, the number of data is N
And j = 0 to N−1. In this embodiment, the peripheral waveform
Is subjected to rectification and integration processing in the smoothing circuit 5.
, The data (j) has a DC component

[0019]

[Outside 1]

Is included. DC component is data
(J) corresponds to the average value. To find the autocorrelation coefficient
A waveform obtained by removing the DC component from the original waveform must be used.
No. Data were but I, first, at step 16 of FIG. 3
An average value of (j) is obtained.

Next, at step 17, an autocorrelation coefficient is obtained.
The autocorrelation function C (i) required for is calculated. Where i
= 0 to N-1. Let Δt be the sampling interval
And C (i) is a value at a certain point in time and an arbitrary time i ·
This is a function representing a correlation with a value at a time point separated by Δt. d
If ata (j) is a signal with a period,
Accordingly, the value of C (i) changes, and is changed by an integer multiple of the fundamental period.
The change which takes the maximum value when the phases are shifted repeats.
In step 17, as described above, data (j)
The calculation is performed using the signal from which the DC component is removed.

Next, at step 18, the autocorrelation coefficient R (i)
Ask for. FIG. 5 is a graph of a calculation example of R (i).
It is. R (i) theoretically has a phase shift of data
It takes a maximum value when it is an integral multiple of the basic period of (j). Figure
As shown in Fig. 5, there are local maxima at a, b, c, d, and e.
And If we find the true fundamental period from these maxima,
That is one cycle of the heartbeat, and the heart rate per unit time is
Can be calculated. Steps 1-9 and Step 24
This is the step of finding the maximum point of. These maxima are
There are also those that happen to be the local maximum due to errors,
In step 21, a threshold function f (i) is set and exceeds
Are treated as maximum points in step 22. Threshold
The value function f (i) is set, for example, as shown in FIG. this
Thus, the points a and b are smaller than f (i), so
Instead, the points c, d, and e are the maximum points. Step 22 and
And 23, the function representing these maximum points is peak
It is replaced by a function (x). That is, the point c,
To d, respectively x i _c corresponding to e, i _d, and i _e
If, peak (x) is the maximum value when x is i _c, i _d, i _{e
R (i c), (i} d), take (i _e), x is I other than the above
If it is, peak (x) becomes 0.

Next, in step 25, peak (j) is maximized.
The value of the argument j which is a value is set as a temporary basic period T. However,
T may still be an integer multiple of the true fundamental period
Therefore, in steps 26 and 27, a cycle of a fraction of T
There is a second maximum point in the vicinity of, ie peak
(J) Search for an argument j having a value of> 0
You. In this embodiment, T = ie in step 25.
As in 5, ie is actually a position twice the true fundamental period T.
Because it is (2T), 2 of T (= i _e) in step 26
Argument j taking a value such that peak (j)> 0 in the vicinity of 1 /
Is searched for, and at step 27 a new i _c
Is assigned to T. Returning to step 26 again, T (=
peak (j)> 0 near half of i _c )
Take but whether there is argument j is searched, less than i _c
Has no such j, so i _c is the true fundamental period T
You. Then, using this T, in step 28, the unit time
The heart rate is calculated.

In the present invention, the value of the filter is set to, for example,
Set to 0.1-0.3Hz to detect respiration
It is also possible.

[0025]

As described above, the on-vehicle heart rate detecting device of the present invention has the following effects.

(1) Since the vibration noise of the vehicle body can be eliminated, the heartbeat can be detected more accurately even in the vehicle.

(2) Autocorrelation coefficient is larger than threshold function
Since the basic period is calculated for the elements corresponding to the basic period of the input signal in the set of maximal points , even if various noises such as automobiles are present, the basic period is calculated. The heart rate can be calculated accurately without being affected by the heart rate.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an in-vehicle heartbeat detection device according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram illustrating the operation of a subtraction unit in the apparatus.

FIG. 3 is a block diagram showing a configuration of a heartbeat information detecting unit according to a second embodiment of the present invention.

FIG. 4 is a flowchart for calculating a heart rate in the apparatus.

FIG. 5 is a graph of an autocorrelation coefficient in the apparatus.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Biological information detection means 6 Seismic sensing means 11 Subtraction means 12 Heart rate information detection means 13 Autocorrelation coefficient calculation means 14 Basic period calculation means 15 Calculation means

Continuation of the front page (56) References JP-A-53-5888 (JP, A) JP-A-2-1566 (JP, A) JP-A-4-3709 (JP, U)

Claims (1)

(57) [Claims] 1. A living body information detecting means provided on a seat in a vehicle, a vibration sensing means provided at a position different from a position where said living body information detecting means is provided, and an output of said living body information detecting means. A subtraction unit that subtracts an output signal of the seismic unit from a signal, and a heartbeat information detection unit that detects heartbeat information of a person at the seat from an output signal from the subtraction unit,
Heart rate information detecting means for calculating an auto-correlation coefficient for the input signal; and a maximum point of the auto-correlation coefficient being larger than a value of a threshold function set for the auto-correlation coefficient. The maximum value of the maximum point set satisfying the condition is
Before being in the vicinity of an integer fraction of the period corresponding to the indicated element
An in-vehicle heart rate detecting device comprising: a basic period calculating unit that calculates a basic period for another element of the maximum point set; and a calculating unit that calculates a cycle per unit time from the basic period.