JP2017207326A - Body weight estimation system and body weight estimation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve downsizing and weight reduction to a level at which a body weight estimation system can be carried in a pocket, etc., and to enable a body weight to be measured easily anytime without the need for a subject to step on a scale.SOLUTION: The present invention acquires, from time-series acceleration data measured by an acceleration sensor, a time Δt from when a subject starts going down a step to when the subject finishes going down, an impact acceleration Gat the time when the subject has finished going down the step, and a time Tfrom the moment when the subject has finished going down the step to when the acceleration converges, as information at time when the subject went down the step, and calculated an estimated value m' of the body weight of the subject on the basis of the acquired information at time when the subject went down the step.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a weight estimation system and a weight estimation method using an acceleration sensor.

  Conventionally, when estimating the weight of a subject, a weight scale such as a mechanical health meter or a digital health meter has been used (see, for example, Patent Document 1).

JP 2010-0778504 A

  However, in the conventional weight scale, since the subject measures the weight by riding on the weight scale, there is a limit to downsizing. Moreover, since it is necessary to support the body weight of the subject, it must be made sturdy and the weight is also increased.

  In addition, conventional scales are often placed in a bathroom at home. Therefore, when you want to measure your body weight, when you are away from the bathroom, you need to go to the bathroom, and you may be discouraged from measuring your body weight. Thus, the conventional weight scale sometimes misses the opportunity for weight measurement.

  The present invention has been made to solve such a problem, and the object of the present invention is to reduce the size and weight to such an extent that it can be carried in a pocket or the like, and the subject can take a weight scale. It is another object of the present invention to provide a weight estimation system and a weight estimation method that can easily measure the weight at any time.

  In order to achieve such an object, the present invention provides an acceleration sensor (1) to be worn on the subject's body, and weight estimation for estimating the subject's weight from time-series acceleration data measured by the acceleration sensor (1). A weight estimation system comprising means (2), wherein the weight estimation means (2) stores acceleration data in time series when the subject measures a step measured by the acceleration sensor (1). From the time-series acceleration data stored in the storage means (21) and the acceleration data storage means (21), as information when the subject descends the step, the time from when the subject begins to descend to the end of the step, Information acquisition means (22) for acquiring the impact acceleration when the subject finishes the step and the time from when the subject finishes the step until the acceleration converges, and information acquisition Characterized in that it comprises the estimated weight calculation means for calculating (23) an estimate of the subject's body weight, based on the information when the subject obtained by the step (22) has got off the step.

In the present invention, when the subject wants to know his / her weight, he / she steps down. Then, the weight estimating means (2) stores time-series acceleration data when the subject measured by the acceleration sensor (1) descends the step, and the subject determines the step from the stored time-series acceleration data. As information at the time of getting off, the time from when the subject starts to step down to the end of the step (Δt), the impact acceleration (G ₀ ) when the subject finishes getting off the step, and the moment when the subject finishes getting off the step A time (T ₀ ) until the acceleration converges is acquired, and an estimated value (m ′) of the subject's weight is calculated based on the acquired information. For example, the following equation (1) is used to calculate the estimated weight (m ′) of the subject. However, in the formula (1), A is a constant, and g is gravitational acceleration.

  The present invention also includes an acceleration sensor (1) mounted on the subject's body, and weight estimation means (3) for estimating the subject's weight from time-series acceleration data measured by the acceleration sensor (1). A weight estimation system, wherein the weight estimation means (3) includes acceleration data storage means (31) for storing time-series acceleration data when the subject measured by the acceleration sensor (1) goes down the stairs; From the time-series acceleration data stored in the acceleration data storage means (31), as the information for each step when the subject steps down the stairs, the time from when the subject starts to step down to the end of the stairs, and the subject Information acquisition means (32) for acquiring the impact acceleration when the subject has finished going down one stairs and the time from when the subject finished going down one stairs until the acceleration converges, Subjects acquired by the acquisition means (32) is characterized in that it comprises the estimated weight calculation unit, based on 1 stairs every information when down the stairs to calculate the estimated value of the subject's body weight (33).

In this invention, when the subject wants to know his / her weight, he goes down the stairs. Then, the weight estimating means (3) stores time-series acceleration data when the subject measured by the acceleration sensor (1) descends the stairs, and the subject takes the stairs from the stored time-series acceleration data. As information for each staircase when getting off, the time from when the subject starts to go down one stairs to the end (Δt), the impact acceleration when the subject finishes going down one stairs (G ₀ ), and the subject The time (T ₀ ) until the acceleration converges from the moment when the user finishes going down one staircase is acquired, and the estimated value (m ′) of the subject's weight is calculated based on the acquired information. For example, the following equation (2) is used to calculate the estimated weight (m ′) of the subject. In calculating the estimated value (m ′) of the subject's weight, for example, the following equation (2) is used, and an estimated value m _i ′ of the subject's weight for each step is calculated from the equation (2). The average value of the calculated estimated weights m _i ′ for each step is taken as the estimated weight m ′ of the subject. In equation (2), A is a constant, and g is gravitational acceleration.

  In the present invention, since the weight of the subject is estimated from time-series acceleration data measured by the acceleration sensor, the weight estimation system can be reduced in size and weight to the extent that it can be carried in a pocket or the like. In addition, the subject only needs to jump or walk and can easily measure his / her weight at any time. In addition, when an acceleration sensor is mounted on a mobile terminal such as a smartphone, it is possible to measure the weight of the subject using this mobile terminal.

  In the above description, as an example, constituent elements on the drawing corresponding to the constituent elements of the invention are indicated by reference numerals with parentheses.

  As described above, according to the present invention, it can be reduced in size and weight to the extent that it can be carried in a pocket or the like, and it is possible to easily measure the weight at any time without the subject getting on the scale. Become. In addition, when an acceleration sensor is mounted on a mobile terminal such as a smartphone, it is possible to measure the weight of the subject using this mobile terminal.

FIG. 1 is a diagram showing a main part of the weight estimation system according to Embodiment 1 of the present invention. FIG. 2 is a flowchart for explaining the operation of the weight estimation system according to the first embodiment. FIG. 3 is a diagram illustrating time-series acceleration data when the subject measured by the acceleration sensor steps down the step. FIG. 4 shows the weight estimation of the first embodiment when the actual body weight m of the subject is 58 kg (Example 1), 65 kg (Example 2), 68 kg (Example 3), 71 kg (Example 4), and 80 kg (Example 5). It is a figure which illustrates the estimated value (measured weight) m 'of the test subject's weight estimated by the system. FIG. 5 is a diagram showing a main part of the weight estimation system according to Embodiment 2 of the present invention. FIG. 6 is a flowchart for explaining the operation of the weight estimation system according to the second embodiment. FIG. 7 shows the weight estimation of the second embodiment when the actual weight m of the subject is 58 kg (Example 1), 65 kg (Example 2), 68 kg (Example 3), 71 kg (Example 4), and 80 kg (Example 5). It is a figure which illustrates the estimated value (measured weight) m 'of the test subject's weight estimated by the system. FIG. 8 is a diagram showing a main part of the weight estimation system according to Embodiment 3 of the present invention. FIG. 9 is a diagram exemplifying a result of measuring the body weight of a subject with a weight (2 kg, 4 kg, 6 kg, 8 kg, 10 kg) attached to a subject having an actual weight m of 71 kg in the weight estimation system of the third embodiment. is there. FIG. 10 is a diagram showing a main part of the weight estimation system according to Embodiment 4 of the present invention. FIG. 11 is a diagram exemplifying a result of measuring the body weight of a subject with a weight (2 kg, 4 kg, 6 kg, 8 kg, 10 kg) attached to a subject having an actual weight m of 71 kg in the weight estimation system of the fourth embodiment. is there.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[Embodiment 1: Example of estimating weight by descending step]
FIG. 1 shows a main part of weight estimation system 100 (100A) according to Embodiment 1 of the present invention. The weight estimation system 100A includes an acceleration sensor 1 attached to the body of the subject and a weight estimation unit 2 that estimates the weight of the subject from time-series acceleration data α measured by the acceleration sensor 1.

  The weight estimation unit 2 is realized by hardware including a processor and a memory, and a program that realizes various functions in cooperation with these hardware, and includes an acceleration data storage unit 21, a data processing unit 22, an estimated weight And a calculation unit 23.

  In the weight estimation system 100A according to the first embodiment, when the subject wants to know his / her weight, the subject steps down the step. The acceleration data α at this time, that is, the time-series acceleration data α measured by the acceleration sensor 1 when the subject goes down the step is stored in the acceleration data storage unit 21 (steps S101 and S102 shown in FIG. 2). .

FIG. 3 exemplifies time-series acceleration data (time-series acceleration data from when the subject steps down the step and stops) when the subject steps down the step, measured by the acceleration sensor 1. Time-series acceleration data α when the subject steps down the step is stored in the acceleration data storage unit 21. The data processing unit 22 uses the time-series acceleration data α stored in the acceleration data storage unit 21 as information when the subject steps down the step Δt from when the subject starts to step down until the end of the step, The impact acceleration G ₀ when the subject finishes stepping down and the time T ₀ until the acceleration converges from the moment when the subject finishes stepping down are acquired (step S103) and sent to the estimated weight calculation unit 23.

The estimated weight calculation unit 23 is information when the subject sent from the data processing unit 22 steps down, that is, a time Δt from when the subject starts to step down until it finishes, and when the subject finishes stepping down. Substituting the impact acceleration G ₀ and the time T ₀ until the acceleration converges from the moment when the subject finishes stepping into the following equation (3), the estimated weight (measured weight) m ′ of the subject is Obtained (step S104). In Equation (3), A is a constant (a constant including the coefficient of restitution of the ground and the footwear of the subject), and g is the gravitational acceleration.

  FIG. 4 shows the weight estimation system of the first embodiment when the actual weight m of the subject is 58 kg (Example 1), 65 kg (Example 2), 68 kg (Example 3), 71 kg (Example 4), and 80 kg (Example 5). The measured weight (estimated value of the subject's weight) m ′ obtained at 100A is exemplified. In this example, a MEMS (Micro Electro Mechanical Systems) acceleration sensor is used as the acceleration sensor 1.

  According to the weight estimation system 100A of the first embodiment, since the weight of the subject is estimated from the time-series acceleration data α measured by the acceleration sensor 1, the weight estimation system can be downsized to the extent that it can be carried in a pocket or the like. The weight can be reduced. In addition, the subject only needs to go down the stairs and can easily measure his / her weight at any time. In addition, when an acceleration sensor is mounted on a mobile terminal such as a smartphone, it is possible to measure the weight of the subject using this mobile terminal.

[Embodiment 2: Example of estimating body weight by going down stairs]
FIG. 5 shows a main part of weight estimation system 100 (100B) according to Embodiment 2 of the present invention. The weight estimation system 100B includes an acceleration sensor 1 attached to the subject's body, and a weight estimation unit 3 that estimates the subject's weight from time-series acceleration data α measured by the acceleration sensor 1.

  The weight estimation unit 3 is realized by hardware including a processor and a memory, and a program that realizes various functions in cooperation with these hardware, and includes an acceleration data storage unit 31, a data processing unit 32, an estimated weight And a calculation unit 33. The estimated weight calculation unit 33 includes an estimated weight calculation unit 33-1 for each step and an average value calculation unit 33-2.

  In the weight estimation system 100B of the second embodiment, when the subject wants to know his / her weight, he / she goes down the stairs (goes down 10 steps or more). In this case, time series acceleration data α similar to that shown in FIG. 3 is obtained each time the subject steps down one step. The time-series acceleration data α when the subject goes down the stairs is stored in the acceleration data storage unit 31 (steps S201 and S202 shown in FIG. 6).

The data processing unit 32 uses the time-series acceleration data α stored in the acceleration data storage unit 31 as information for each step when the subject steps down the stairs until the subject starts to step down and ends. The time Δt, the impact acceleration G ₀ when the subject finishes stepping down, and the time T ₀ until the acceleration converges from the moment when the subject finishes stepping down are acquired (step S203), and the estimated weight calculation unit Send to 33.

In the estimated weight calculation unit 33, the estimated weight calculation unit 31-1 for each stairs is information for each stairs when the subject sent from the data processing unit 22 steps down the stairs, that is, the subject has gone down the stairs. As the information for each step, the time Δt from when the subject starts to step down to the end, the impact acceleration G ₀ when the subject finishes the step, and the acceleration from the moment when the subject finishes the step, The time T ₀ until convergence is substituted into the following equation (4) to calculate an estimated value m _i ′ of the weight for each step of the subject (step S204). In Equation (4), A is a constant (a constant including the coefficient of restitution of the ground and the footwear of the subject), and g is the gravitational acceleration.

When the average value calculation unit 33-2 collects a predetermined number n (for example, n = 10) of estimated weights m _i ′ of the subject for each step calculated by the estimated weight calculation unit 31-1 for each step. Then, an average value of the estimated weight values m _i ′ for each step is obtained, and the obtained average value is set as an estimated value m ′ (m ′ = m _i ′ / n) of the subject's weight (step S205).

  FIG. 7 shows the weight estimation system 100B according to the second embodiment when the weight of the subject is 58 kg (Example 1), 65 kg (Example 2), 68 kg (Example 3), 71 kg (Example 4), and 80 kg (Example 5). The obtained measured body weight (estimated value of the body weight of the subject) m ′ is exemplified. In this example, a MEMS (Micro Electro Mechanical Systems) acceleration sensor is used as the acceleration sensor 1.

  According to the weight estimation system 100B of the second embodiment, since the subject's weight is estimated from the time-series acceleration data α measured by the acceleration sensor 1, the weight estimation system can be downsized to the extent that it can be carried in a pocket or the like. The weight can be reduced. In addition, the subject only needs to go down the stairs and can easily measure his / her weight at any time. In addition, when an acceleration sensor is mounted on a mobile terminal such as a smartphone, it is possible to measure the weight of the subject using this mobile terminal.

[Embodiment 3: Example of calibrating body weight estimated by descending step based on personal data]
FIG. 8 shows a main part of weight estimation system 100 (100C) according to Embodiment 3 of the present invention. The weight estimation system 100C according to the third embodiment is obtained by adding a calibration function to the weight estimation system 100A (FIG. 1) according to the first embodiment. The weight estimation unit 2 includes an acceleration data storage unit 21 and data processing. In addition to the unit 22 and the estimated weight calculation unit 23, a personal data storage unit 24, a calibration value calculation unit 25, and an estimated weight calibration unit 26 are further provided.

  In the personal data storage unit 24, the time-series acceleration data α when the subject (71 kg) measured by the acceleration sensor 1 descends the step and the actual weight m (71 kg) of the subject at that time are personal data. Is remembered as

  The calibration value calculation unit 25 estimates the weight of the subject (71 kg) from the time-series acceleration data α stored in the personal data storage unit 24 in the same manner as the processing in the data processing unit 22 and the weight estimation calculation unit 23. A value m ′ is obtained, and a calibration value c is calculated as c = m / m ′ from the obtained estimated weight m ′ of the subject and the actual weight m (71 kg) of the subject.

  The estimated weight calibration unit 26 receives the calibration value c calculated by the calibration value calculation unit 25 and calibrates the estimated weight m ′ of the subject calculated by the estimated weight calculation unit 23. In this example, the estimated value m ′ of the subject is calibrated by multiplying the estimated value m ′ of the subject by the calibration value c, and the measured weight m ″ (m ″ = m ′ × c).

  FIG. 9 illustrates the result of measuring the weight of a subject with a weight (2 kg, 4 kg, 6 kg, 8 kg, 10 kg) attached to a subject whose actual weight m is 71 kg in the weight estimation system 100C of the third embodiment. . In this case, the personal data storage unit 24 stores in advance personal data of a subject whose actual weight m is 71 kg (time-series acceleration data α + actual weight m when descending a step).

  As can be seen from this result, by using the weight estimation system 100C of the third embodiment, the estimated weight m ′ of the subject is calibrated based on the personal data stored in the personal data storage unit 24. Even if the subject's body weight increases or decreases, it is possible to measure the body weight with little error in accordance with the increase or decrease of the body weight.

[Embodiment 4: Example of calibrating body weight estimated by going down stairs based on personal data]
FIG. 10 shows a main part of weight estimation system 100 (100D) according to Embodiment 4 of the present invention. The weight estimation system 100D according to the fourth embodiment is obtained by adding a calibration function to the weight estimation system 100B (FIG. 5) according to the second embodiment. The weight estimation unit 3 includes an acceleration data storage unit 31 and data processing. In addition to the unit 32 and the estimated weight calculation unit 33, a personal data storage unit 34, a calibration value calculation unit 35, and an estimated weight calibration unit 36 are further provided.

  In the personal data storage unit 34, the time-series acceleration data α when the subject (71 kg) measured by the acceleration sensor 1 goes down the stairs and the actual weight m (71 kg) of the subject at that time are personal data. Is remembered as

  The calibration value calculation unit 35 estimates the weight of the subject (71 kg) from the time-series acceleration data α stored in the personal data storage unit 34 in the same manner as the processing in the data processing unit 32 and the weight estimation calculation unit 33. A value m ′ is obtained, and a calibration value c is calculated as c = m / m ′ from the obtained estimated weight m ′ of the subject and the actual weight m (71 kg) of the subject.

  The estimated weight calibration unit 36 receives the calibration value c calculated by the calibration value calculation unit 35 as input, and calibrates the estimated weight m ′ of the subject calculated by the estimated weight calculation unit 33. In this example, the estimated value m ′ of the subject is calibrated by multiplying the estimated value m ′ of the subject by the calibration value c, and the measured weight m ″ (m ″ = m ′ × c).

  FIG. 11 illustrates the result of measuring the body weight of a subject with a weight (2 kg, 4 kg, 6 kg, 8 kg, 10 kg) attached to a subject having an actual weight m of 71 kg in the weight estimation system 100D of the fourth embodiment. . In this case, the personal data storage unit 34 stores in advance personal data of a subject whose actual weight m is 71 kg (time-series acceleration data α + actual weight m when going down the stairs).

  As can be seen from this result, when the weight estimation system 100D of the fourth embodiment is used, the estimated value m ′ of the subject's weight is calibrated based on the personal data stored in the personal data storage unit 34. Even if the subject's body weight increases or decreases, it is possible to measure the body weight with little error in accordance with the increase or decrease of the body weight.

  Note that the weight estimation system 100 (100A to 100D) and the digital health meter described above may be linked wirelessly, and the weight measured by the digital health meter may be used in the weight estimation system 100. For example, the acceleration data measured by the weight estimation system 100 and the weight measured by the digital health meter are linked, and when similar acceleration data is output next time, the weight is displayed.

  In the above-described weight estimation system 100 (100A to 100D), time-series acceleration data measured by the acceleration sensor 1 may be transmitted and stored on the cloud with a time stamp.

[Extension of the embodiment]
The present invention has been described above with reference to the embodiment. However, the present invention is not limited to the above embodiment. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the technical idea of the present invention.

  DESCRIPTION OF SYMBOLS 1 ... Acceleration sensor, 2, 3 ... Weight estimation part, 21, 31 ... Acceleration data storage part, 22, 32 ... Data processing part, 23, 33 ... Estimated weight calculation part, 24, 34 ... Personal data storage part, 25, 35 ... Calibration value calculation unit, 26, 36 ... Estimated weight calibration unit, 31-1 ... Estimated weight calculation unit for each step, 31-2 ... Average value calculation unit, 100 (100A to 100D) ... Weight estimation system.

Claims (8)

A weight estimation system comprising: an acceleration sensor attached to the body of a subject; and weight estimation means for estimating the weight of the subject from time-series acceleration data measured by the acceleration sensor,
The weight estimating means includes
Acceleration data storage means for storing time-series acceleration data when the subject measured by the acceleration sensor has stepped down,
From the time-series acceleration data stored in the acceleration data storage means, as information when the subject descends the step, the time from when the subject begins to descend until the end of the step, and the subject finishes descending the step Information acquisition means for acquiring the impact acceleration at the time of the acceleration and the time from the moment when the subject finishes stepping down until the acceleration converges,
A weight estimation system comprising: estimated weight calculation means for calculating an estimated value of the weight of the subject based on information obtained when the subject has stepped down acquired by the information acquisition means. The weight estimation system according to claim 1,
The estimated weight calculating means includes
Δt is the time from when the subject starts to step down to the end of the step, G _{0 is} the impact acceleration when the subject finishes the step, and the time from the moment when the subject finishes the step until the acceleration converges The estimated weight value m ′ of the subject is calculated from the following equation (1) as T ₀ .

However, A is a constant and g is gravitational acceleration. In the weight estimation system according to claim 1 or 2,
Calibration value calculation means for calculating a calibration value from time-series acceleration data when the subject has entered the step as personal data and the weight of the subject;
A weight estimation system comprising: an estimated weight calibration unit that calibrates the estimated weight of the subject calculated by the estimated weight calculation unit using the calibration value calculated by the calibration value calculation unit. . A weight estimation system comprising: an acceleration sensor attached to the body of a subject; and weight estimation means for estimating the weight of the subject from time-series acceleration data measured by the acceleration sensor,
The weight estimating means includes
Acceleration data storage means for storing time-series acceleration data measured by the acceleration sensor when the subject goes down the stairs;
From the time-series acceleration data stored in the acceleration data storage means, as information for each stairs when the subject steps down the stairs, the time from when the subject starts to step down to the end of the stairs, Information acquisition means for acquiring the impact acceleration when the subject has finished going down one stairs and the time from when the subject finished going down one stairs until the acceleration converges;
A weight estimation system comprising: estimated weight calculation means for calculating an estimated value of the weight of the subject on the basis of information for each step when the subject gets down the stairs acquired by the information acquisition means. . In the weight estimation system according to claim 4,
The estimated weight calculating means includes
The time from when the subject starts to get down to one stairs to the end is Δt, the impact acceleration when the subject finishes going down one stairs is G ₀ , and the acceleration from the moment the subject finishes going down one stairs The time until convergence is set as T ₀ , the estimated weight m _i ′ of each step of the subject is calculated from the following equation (2), and the average of the calculated estimated weights m _i ′ of each step is calculated. A weight estimation system, wherein the value is an estimated value m ′ of the subject's weight.

However, A is a constant and g is gravitational acceleration. In the weight estimation system according to claim 4 or 5,
Calibration value calculation means for calculating a calibration value from time-series acceleration data when the subject has entered the stairs and the subject's body weight, which are input in advance as personal data;
A weight estimation system comprising: an estimated weight calibration unit that calibrates the estimated weight of the subject calculated by the estimated weight calculation unit using the calibration value calculated by the calibration value calculation unit. . A weight estimation method for estimating the weight of a subject from time-series acceleration data measured by an acceleration sensor attached to the subject's body,
Acceleration data storage step for storing time-series acceleration data when the subject measured by the acceleration sensor steps down a step;
From the time-series acceleration data stored in the acceleration data storage step, as the information when the subject descends the step, the time from when the subject begins to descend until the end of the step, and the subject finishes descending the step An information acquisition step of acquiring the impact acceleration at the time of the acceleration and the time from the moment when the subject finishes stepping down until the acceleration converges,
A weight estimation method comprising: an estimated weight calculation step of calculating an estimated value of the weight of the subject based on information obtained when the subject has stepped down acquired by the information acquisition step. A weight estimation method for estimating the weight of a subject from time-series acceleration data measured by an acceleration sensor attached to the subject's body,
Acceleration data storage step for storing time-series acceleration data when the subject measured by the acceleration sensor goes down the stairs;
From the time-series acceleration data stored in the acceleration data storage step, as information for each step when the subject steps down the stairs, the time from when the subject starts to step down to the end of the steps, An information acquisition step of acquiring the impact acceleration when the subject has finished going down one stairs and the time from when the subject finished going down one stairs until the acceleration converges;
A weight estimation method comprising: an estimated weight calculation step of calculating an estimated value of the weight of the subject based on information for each step when the subject has gone down the stairs acquired by the information acquisition step. .

Images