WO2014181602A1 - Walking posture meter and program - Google Patents

Abstract

A walking posture meter (100) that evaluates the walking posture of a measurement subject and comprises: an acceleration sensor (112) that is attached above the median line of the waist of the measurement subject; and an arithmetic calculation unit (110) that uses a temporal change waveform for up-down axis acceleration output by the acceleration sensor and a temporal change waveform for front-rear axis acceleration, and calculates an amount corresponding to the posture inclination in the front-rear direction during walking by the measurement subject.

Description

Walking posture meter and program

The present invention relates to a walking posture meter, and more particularly to a walking posture meter that quantitatively evaluates whether or not a human walking posture is a correct posture.

The present invention also relates to a program for causing a computer to execute a method for quantitatively evaluating whether or not a human walking posture is a correct posture.

Conventionally, as an apparatus for evaluating the posture of a human during walking, for example, there is an apparatus disclosed in Patent Document 1 (Japanese Patent Application Laid-Open No. 2011-078728). This device detects the direction of gravitational acceleration based on the output from an acceleration sensor attached to the human hip when the human takes a predetermined posture, so that the tilt angle of the hip with respect to the ground when the posture is taken To estimate the pelvic tilt.

Further, for example, the device disclosed in Patent Document 2 (Japanese Patent Application Laid-Open No. 2011-251013) calculates a movement amount from an output of an acceleration sensor attached to a human waist and acquires a walking locus based on the movement amount. To do.

JP 2011-078728 A JP 2011-251013 A

As a point to evaluate the correctness (or beauty) of a human walking posture, walk in a neutral posture with your back straight to the sky and neither forward leaning (forward leaning) nor waist (back leaning) in the direction of travel. There is a viewpoint of whether or not.

However, conventionally, there has been no means for easily and quantitatively evaluating whether the posture of the walking human is the above-described forward leaning posture, the backward leaning posture, or the neutral posture that is neither of them. .

In view of the above, according to one aspect of the present invention, there is provided a walking posture meter that can easily evaluate whether or not a human walking posture is tilted forward or backward with respect to the traveling direction.

Further, according to another aspect of the present invention, there is provided a program for causing a computer to execute a method for easily evaluating whether or not a human walking posture is tilted forward / backward with respect to the traveling direction.

In order to solve the above problems, a walking posture meter according to one aspect of the present invention is a walking posture meter that evaluates the walking posture of the measurement subject, and an acceleration sensor that is mounted on the midline of the measurement subject's waist; Using the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration output from the acceleration sensor, an amount corresponding to the forward / backward inclination of the posture of the person being measured is calculated. And an arithmetic unit.

In this specification, the “inclination in the front-rear direction” of the posture while walking means the inclination along the traveling direction of the upper body of the human being walking (the upper body from the pelvis). The inclination may be for an arbitrary posture (leg phase) that appears while the gait goes around. For example, the inclination may be about the inclination of the posture of the person to be measured in the vicinity of the time when the heel of the front leg contacts the road surface. Further, for example, the inclination may be about the inclination of the posture of the measurement subject in the vicinity of the time point where the rear leg, which is a free leg, coincides with the front leg, which is a standing leg, along the traveling direction.

In the walking posture meter according to one aspect of the present invention, the acceleration sensor is mounted on the midline of the waist. The arithmetic unit uses the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration output from the acceleration sensor to respond to the degree of inclination of the posture of the subject during walking in the longitudinal direction. The amount to be calculated is calculated. Therefore, according to the amount, it is possible to evaluate the degree of inclination in the front-rear direction of the posture of the measurement subject during walking. In addition, this walking posture meter can perform the above-described evaluation based on the output of the acceleration sensor mounted on the midline of the subject's waist, so that it does not depend on large-scale equipment such as motion capture. It is possible to evaluate easily.

In the walking posture meter according to the embodiment, the calculation unit derives a characteristic parameter indicating characteristics of the temporal change of the vertical axis acceleration and the temporal change of the longitudinal axis acceleration, and the inclination in the front-rear direction based on the characteristic parameter. An amount corresponding to is calculated.

The calculation unit derives a feature parameter indicating the temporal change characteristics of the vertical axis acceleration and the longitudinal axis acceleration, and calculates an amount corresponding to the tilt in the front-rear direction based on the feature parameter. An amount corresponding to the degree of inclination of the posture in the front-rear direction can be quickly calculated.

In the walking posture meter according to one embodiment, the calculation unit derives a plurality of types of feature parameters, and an amount corresponding to the inclination in the front-rear direction based on an amount obtained by weighted addition of the derived types of feature parameters. Is calculated.

The calculation unit can evaluate the degree of inclination in the front-rear direction of the posture of the measurement subject while walking by weight-adding the plurality of derived characteristic parameters. Therefore, the calculation required for the evaluation is extremely simple, and quick evaluation is possible.

In the walking posture meter according to one embodiment, the calculation unit estimates the step length of the subject based on a temporal change waveform of the vertical axis acceleration, and performs the weighted addition based on the estimated step length. The weighting coefficient to be used is changed.

The calculation unit estimates the measured person's stride based on the temporal change waveform of the vertical axis acceleration, and changes the weighting coefficient used for the weighted addition based on the estimated stride. Appropriate evaluation is possible regardless of the characteristics of the direction (the width of the stride).

In the walking posture meter according to an embodiment, the calculation unit is configured to calculate the walking cycle of the subject based on at least one of the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration. Alternatively, a reference period corresponding to one step in the walking cycle is obtained, and a weighting coefficient used for the weighted addition is changed based on the obtained walking cycle or reference period.

The calculation unit obtains a reference period corresponding to one step of the walking cycle of the measured person or the walking cycle based on at least one of the temporal change waveform of the vertical axis acceleration and the longitudinal axis acceleration. Since the weighting coefficient used for the weighted addition is changed based on the walking cycle or the reference period, appropriate evaluation can be performed regardless of the characteristics of the person to be measured (walking cycle length).

In the walking posture meter according to an embodiment, the amount corresponding to the inclination in the front-rear direction is an angle formed between the pelvis of the measurement subject and the horizontal direction during walking.

The calculation unit calculates the tilt angle (angle formed with the horizontal) of the person being measured while walking, and evaluates the tilt of the posture of the person being measured in the front-rear direction using the calculated tilt angle. Therefore, this walking posture meter can perform evaluation with extremely high accuracy.

A program according to another aspect of the present invention is a program for causing a computer to execute a method for evaluating a walking posture of a measured person, the method including an acceleration mounted on a midline of the waist of the measured person. Using the step of obtaining the sensor output and the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration output by the acceleration sensor, Calculating an amount corresponding to the inclination of the direction.

By executing the program, the computer first acquires the output of the acceleration sensor mounted on the midline of the subject's waist. Then, using the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration output by the acceleration sensor, it corresponds to the degree of the forward / backward inclination of the posture of the measurement subject during walking. Calculate the amount. Therefore, the inclination of the posture of the measurement subject during walking can be evaluated according to the amount. In addition, this program enables the above-mentioned evaluation based on the output of the acceleration sensor mounted on the midline of the subject's waist, so it is easy to use regardless of large equipment such as motion capture. It is possible to evaluate.

As is clear from the above, according to the walking posture meter according to one aspect of the present invention, it is possible to easily evaluate the inclination in the front-rear direction of the posture of the measurement subject during walking.

In addition, by causing a computer to execute a program according to another aspect of the present invention, it is possible to easily evaluate the tilt in the front-rear direction of the posture of the person being measured while walking.

It is a figure which shows the system configuration | structure of the walking posture meter of one Embodiment of this invention. It is a figure which shows the block configuration of the active mass meter which makes the system of the said walking posture meter. It is a figure which shows the block configuration of the smart phone which makes the system of the said walking posture meter. FIG. 4A is a diagram showing a mode in which the activity meter is attached to the measurement subject. FIG. 4B is a diagram for explaining the X-axis (front-rear axis), the Y-axis (left-right axis), and the Z-axis (up-down axis). FIG. 5A is a diagram for explaining the relationship between the forward inclination (backward inclination) of the walking posture and the angle between the pelvis and the horizontal. 5 (B), 5 (C), and 5 (D) are diagrams showing a posture of a human being walking (timing when a heel of a front leg is grounded). FIG. 5B is a schematic diagram when a human walking backward is viewed from the side. FIG. 5C is a schematic view of a human walking in a good posture that is neither backward nor forward tilted when viewed from the side. FIG. 5D is a schematic diagram when a human walking forward is viewed from the side. Relationship between an example of vertical axis acceleration (time domain) observed by an acceleration sensor attached to the waist when a human walks, a reference period corresponding to one step in the walking cycle, and the gait of one walking step FIG. It is a figure (vertical axis acceleration time change waveform figure) explaining the characteristic parameter which shows the characteristic of the time change of the vertical axis acceleration which the said acceleration sensor outputs. It is a figure (vertical axis speed time change waveform figure) explaining the characteristic parameter which shows the characteristic of the time change of the vertical axis acceleration which the said acceleration sensor outputs. It is a figure (vertical axis orbit time change waveform figure) explaining the characteristic parameter which shows the characteristic of the time change of the vertical axis acceleration which the said acceleration sensor outputs. It is a figure (front-rear axis acceleration time change waveform figure) explaining the characteristic parameter which shows the characteristic of the time change of the longitudinal axis acceleration which the said acceleration sensor outputs. It is a figure (front-rear axis speed time change waveform figure) explaining the characteristic parameter which shows the characteristic of the time change of the longitudinal acceleration which the said acceleration sensor outputs. It is a figure which shows the operation | movement flow of the control part of the said active mass meter. It is a figure which shows the flow of the pelvic inclination angle estimation process by the control part of the said active mass meter. It is a figure which shows the example of the reference table for changing the weighting coefficient set used according to the magnitude | size of a step and the length of the said reference period.

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

FIG. 1 shows a system configuration of a walking posture meter (the whole is denoted by reference numeral 1) according to an embodiment of the present invention. This walking posture meter 1 includes an activity meter 100 and a smartphone 200. In this example, the active mass meter 100 and the smartphone 200 can communicate with each other by BLE (Bluetooth low energy; specified in Bluetooth® Core Specification Ver. 4.0).

As shown in FIG. 2, the activity meter 100 includes a casing 100M, a control unit 110, an oscillation unit 111, an acceleration sensor 112, a memory 120, an operation unit 130, and a display mounted on the casing 100M. Unit 140, BLE communication unit 180, power supply unit 190, and reset unit 199.

The casing 100M is formed in a size that fits in the palm of a human hand so that the activity meter 100 can be easily carried.

The oscillating unit 111 includes a crystal resonator and generates a clock signal that serves as a reference for the operation timing of the activity meter 100. The oscillation unit 111 may be a module chip having a function as a clock generator.

The acceleration sensor 112 detects the three-axis (three directions) acceleration received by the casing 100M and outputs it to the control unit 110. The acceleration sensor 112 may be a module chip of a triaxial acceleration sensor.

The memory 120 includes a ROM (Read Only Memory) and a RAM (Random Access Memory). The ROM stores program data for controlling the activity meter 100. The RAM stores setting data for setting various functions of the activity meter 100, acceleration measurement results, calculation result data, and the like.

The control unit 110 includes a CPU (Central Processing Unit) that operates based on the clock signal. The control unit 110 receives a signal from the acceleration sensor 112 according to a program for controlling the activity meter 100 stored in the memory 120. Based on the detection signal, each part of the activity meter 100 (including the memory 120, the display part 140, and the BLE communication part 180) is controlled. The control unit 110 includes at least a signal processing system that can process time-series data of vertical axis acceleration and longitudinal axis acceleration.

The operation unit 130 includes a button switch in this example, and accepts appropriate operation inputs such as a power on / off switching operation and a display content switching operation.

In this example, the display unit 140 includes a display screen made up of an LCD (liquid crystal display element) or an organic EL (electroluminescence) display, and displays predetermined information on the display screen according to a signal received from the control unit 110. The display unit 140 may be an LED (light emitting diode) that displays on / off of a power source, an operation state, and the like by turning on, turning off, flashing, and the like.

The power supply unit 190 is composed of a button battery in this example, and supplies power to each part of the activity meter 100.

The BLE communication unit 180 communicates with the smartphone 200 in real time. For example, information representing a measurement result is transmitted to the smartphone 200. In addition, an operation instruction is received from the smartphone 200. The BLE communication unit 180 may be a module chip having a BLE function.

The reset unit 199 includes a switch, and resets and initializes the operation of the control unit 110 and the stored content of the memory 120.

As illustrated in FIG. 3, the smartphone 200 includes a main body 200M, a control unit 210, a memory 220, an operation unit 230, a display unit 240, a BLE communication unit 280, and network communication that are mounted on the main body 200M. Part 290. This smartphone 200 is obtained by installing application software (computer program) so that a commercially available smartphone can instruct the activity meter 100.

The control unit 210 includes a CPU and its auxiliary circuit, controls each unit of the smartphone 200, and executes processing according to programs and data stored in the memory 220. That is, data input from the operation unit 230 and the communication units 280 and 290 is processed, and the processed data is stored in the memory 220, displayed on the display unit 240, or output from the communication units 280 and 290. Or

The memory 220 includes a RAM used as a work area necessary for executing the program by the control unit 210 and a ROM for storing a basic program to be executed by the control unit 210. Further, as a storage medium of an auxiliary storage device for assisting the storage area of the memory 220, a semiconductor memory (memory card, SSD (Solid State Drive)) or the like may be used.

The operation unit 230 includes a touch panel provided on the display unit 240 in this example. A keyboard or other hardware operation device may be included.

The display unit 240 includes a display screen (for example, an LCD or an organic EL display). The display unit 240 is controlled by the control unit 210 to display a predetermined image on the display screen.

BLE communication unit 280 communicates with activity meter 100 in real time. For example, an operation instruction is transmitted to the activity meter 100. In addition, information representing a measurement result is received from the activity meter 100.

The network communication unit 290 transmits information from the control unit 210 to other devices via the network 900, and receives information transmitted from other devices via the network 900 and passes the information to the control unit 210. be able to.

For example, as shown in FIG. 4 (A), when this walking posture meter 1 is used by a person 90 to be measured as a user, for example, the activity meter 100 is attached by a wearing clip 100C (shown in FIG. 1). It is attached to the back side of the waist on 90 midline 91.

In this example, as shown in FIG. 4B, for the measurement subject 90, the longitudinal direction is the X axis, the horizontal direction is the Y axis, and the vertical direction is the Z axis. The acceleration sensor 112 of the activity meter 100 includes an X-axis (front-rear axis) acceleration, a Y-axis (left-right axis) acceleration, a Z-axis (which is measured by the casing 100M as the subject 90 walks forward. The acceleration (vertical axis) is output.

When measuring with this walking posture meter 1, the person under measurement 90 turns on the power of the activity meter 100 and the smartphone 200. At the same time, the application software of the smartphone 200 is activated, and the activity meter 100 is instructed to start measurement via the operation unit 230 and the BLE communication unit 280.

In this state, the person to be measured 90 walks straight forward by a predetermined number of steps, in this example, 10 steps. Then, the person under measurement 90 instructs the activity meter 100 to output the calculation and the calculation result via the operation unit 230 and the BLE communication unit 280 of the smartphone 200.

Then, the control part 110 of the active mass meter 100 works as a calculating part, and performs the calculation mentioned later. Then, information representing the calculation result is transmitted to the smartphone 200 via the BLE communication unit 180.

FIG. 12 shows an operation flow by the control unit 110 of the activity meter 100. When the power is turned on, the control unit 110 of the activity meter 100 waits for an instruction to start measurement from the smartphone 200 as shown in step S1. When the measurement start instruction is received from the smartphone 200 (YES in step S1), the control unit 110 acquires the output of the triaxial acceleration by the acceleration sensor 112 as shown in step S2. Acquisition of the output of the acceleration sensor 112 is performed only for a predetermined period (for example, 14 seconds) as a period including acceleration time-series data for 10 steps in this example. The acquired acceleration time series data is temporarily stored in the memory 120. Next, as shown in step S <b> 3, the control unit 110 waits for a measurement start instruction from the smartphone 200. When the calculation instruction is received from the smartphone 200 (YES in step S3), as shown in step S4, the control unit 110 calculates an amount corresponding to the inclination angle of the pelvis. And as shown to step S5, the control part 110 works as an evaluation part, and uses the calculation result (pelvic inclination angle estimation result) to measure the degree of inclination in the front-rear direction of the posture of the person being measured while walking. To evaluate. Then, as shown in step S6, the evaluation result is output (transmitted) to the smartphone 200. Note that the control unit 110 may execute the process of step S4 as soon as acceleration time-series data for at least one step is obtained. In that case, the step of determination shown as step S3 may be omitted.

FIG. 5 (A) is a diagram for explaining the definition of the pelvic inclination angle. This figure is a schematic view of the human waist when viewed from the side, and the pelvis PV is represented by a triangle. In this specification, the pelvic inclination angle θ is a line from the superior anterior iliac spine ASIS (ASIS: Anterior Superior Iliac Spine) to the superior posterior iliac spine PSIS (PSIS: Posterior Superior Iliac Spine) with the clockwise rotation direction being positive. The angle between the minute and the horizontal H. In the following description, a pelvic inclination angle θ of less than +3 degrees is associated with the walking posture “backward inclination”, and a pelvic inclination angle θ of +3 degrees or more and less than +12 degrees is associated with the walking posture “middle”, and +12 degrees. The above pelvic tilt angle θ is associated with the walking posture “forward tilt”. The boundary value is an example, and other boundary values may be used. In the above example, the posture is divided into three stages of “forward leaning”, “middle”, and “backward leaning”, but other stages may be used.

FIGS. 5B to 5D are schematic diagrams showing examples of human walking postures “backward tilt”, “middle”, and “forward tilt”, respectively. As can be seen from FIG. 5B, a human who has a tendency to walk in the walking posture “backward tilting” walks in a posture that turns upside down. (In addition, some people who are “backward tilted” walk with a very stooped posture.) FIG. 5C is an example of the walking posture “middle”. In this way, a human walking in the walking posture “middle” has a straight back and is able to walk correctly (beautifully). Referring to FIG. 5D, a human who has a tendency to walk in a walking posture “forward leaning” walks in a leaning posture with forward leaning. (In addition, some people who are “forward leaning” walk with their stomachs protruding.)

FIG. 6 illustrates the acceleration sensor 112 of the activity meter 100 attached to the waist during a human gait and a reference period (T7 (= Step T) in the figure) corresponding to one step in the walking cycle. It is a figure which shows the relationship with the typical example of the time-change waveform of the vertical-axis acceleration (Z-axis direction acceleration which makes vertical upper direction positive) output.

In the vicinity of the timing when the heel of the extended forefoot (right foot in the figure) comes into contact with the moving surface (the heel contact timing), the vertical axis acceleration changes from negative to positive through the zero cross point.

Thereafter, the vertical axis acceleration includes three peaks (maximum points) (P ₁ (time t = T ₁ ), P ₂ (time t = T ₃ ), P ₃ (time t = T ₅ )) and a valley between them. (Minimum point) (V ₁ (time t = T ₂ ), V ₂ (time t = T ₄ )) appears. In the gait, the timing at which the stance (right foot in the figure) and the free leg (left leg in the figure) substantially coincide with each other in the traveling direction (mid-stance stance timing) corresponds to the vicinity of the timing at which the third peak P3 appears.

When the mid-stance timing of the gait in the gait is exceeded, the vertical axis acceleration passes through the zero cross point again, changes from positive to negative, passes through the minimum point (V ₃ (time t = T ₆ )), and eventually time t = T. ₇ again passes the zero cross point (time t = T ₇ ) and turns from negative to positive. The zero-cross point at time t = T ₇ is the next one-step heel contact timing (with the left foot as the front foot in the figure).

Thus, in the vertical axis acceleration, a waveform as illustrated and described appears during one step of human walking. In this specification, a period (Step T) from the timing when the forefoot heel contacts the ground (heel contact timing) to the next heel contact timing is defined as a reference period. Note that the period from the left foot heel contact timing to the right foot heel contact timing is referred to as the left foot reference period, and from the right foot heel contact timing to the left foot heel contact timing only when there is a particular need to distinguish in the following description. By referring to the period as the right foot reference period, the reference periods of one step with the left foot and one step with the right foot are distinguished.

In the temporal change waveform of the vertical axis acceleration with the upper side being positive, there is a period from the appearance timing of the zero cross point at which the acceleration value changes from negative to positive to the appearance timing of the next zero cross point at which the negative value changes from positive to positive. It corresponds to one reference period.

The control unit 110 described above uses the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration over at least one reference period in the process of step S4 in FIG. 12 (pelvic inclination angle estimation process). Then, an amount corresponding to the inclination in the front-rear direction of the posture of the measurement subject while walking is calculated. Here, as an amount corresponding to the degree of forward / backward tilt in the traveling direction, the subject's pelvic inclination angle θ _{1 at} the heel contact timing and the subject's pelvic inclination angle θ at the mid-stance timing in the same reference period ₂ and the average pelvic inclination angle θ (θ = (θ ₁ + θ ₂ ) / 2) is used. The amount corresponding to the degree of inclination in the front-rear direction is not limited to the average pelvic inclination angle θ.

Hereinafter, the processing (pelvic tilt angle estimation processing) in step S4 in FIG. 12 will be described in detail with reference to the graphs in FIGS. 7 to 11, the flowchart in FIG. 13, and the reference table in FIG. Each step of the flowchart of FIG. 13 may be executed by the control unit 110 described above. The execution subject of each step may be the control unit 210 instead of the control unit 110.

Referring to the flowchart of FIG. 13, in step S41, the control unit 110 calculates the vertical axis (Z-axis) acceleration time series data (FIG. 7) and the longitudinal axis (FIG. 7) from the time change waveform of the triaxial acceleration acquired in step S2. X axis) acceleration time series data (FIG. 10) is generated, and the timing of the zero cross point where the generated vertical axis acceleration time series data changes from negative to positive (time t on the lower horizontal axis in FIG. 7 = about 6.2 (up Time t = 0 on the horizontal axis) is detected, and the detected timing (based on the experimental fact that the zero cross point substantially coincides with the heel contact timing) is specified as the heel contact timing. Control unit 110 similarly detects the next zero-crossing point that changes from negative to positive, and specifies the timing as the start of the next reference period (the period of the current reference period). Thereby, the reference period (Step T) is determined.

Next, in step S42, the controller 110 calculates vertical axis velocity time series data (FIG. 8) and vertical axis trajectory time series data (FIG. 9) from the vertical axis acceleration time series data (FIG. 7).

Next, in step S43, the control unit 110 calculates longitudinal axis velocity time series data (FIG. 11) from longitudinal axis acceleration time series data (FIG. 10).

Next, in step S44, the control unit 110 derives a characteristic parameter indicating characteristics of the temporal change in the vertical axis acceleration and the temporal change in the longitudinal axis acceleration. The control unit 110 uses the time-series data of the vertical axis acceleration, the time-series data of the vertical axis velocity, and the time-series data of the vertical axis trajectory to indicate the characteristic of the temporal change of the vertical axis acceleration. Deriving parameters. Similarly, the control unit 110 derives a characteristic parameter indicating a characteristic of the temporal change in the longitudinal axis acceleration using at least one of the time series data of the longitudinal axis acceleration and the time series data of the longitudinal axis velocity. The target to be derived in step S44 is a feature amount that represents the temporal change feature of each axis acceleration. However, depending on the feature amount, it is easier to use the time-series data of each axis velocity and the temporal change of the trajectory. There is a case. For this reason, the control unit 110 derives an amount (feature parameter) representing a characteristic of the temporal change in the acceleration of each axis using the time series data of the velocity of each axis and the time series data of the trajectory.

In the present embodiment, there are 33 types of characteristic parameters (characteristic parameters PRM1 to PRM33, that is, PRMi (i: an integer of 1 to 33)). Hereinafter, each feature parameter will be described in detail. Note that the number of types of feature parameters and the definition of each feature parameter are examples, and different types of features and definitions of feature parameters may be used.

First, with reference to FIG. 7, 12 types of feature parameters from the first feature parameter PRM1 to the twelfth feature parameter PRM12 will be described.

Regarding the first characteristic parameter PRM1 (Z-axis acceleration first maximum point timing ratio):
The characteristic parameter PRM1 is a parameter related to the appearance timing of the maximum point that appears first when the vertical peak acceleration is based on the first peak in the reference period, that is, the zero cross point that changes from negative to positive. For example, PRM1 is a dimensionless quantity T _ZAP1 / StepT obtained by normalizing the timing T _ZAP1 with a reference period StepT.

Regarding the second characteristic parameter PRM2 (Z-axis acceleration second maximum point timing ratio):
The characteristic parameter PRM2 is a parameter related to the appearance timing of the second maximum peak that appears when the vertical peak acceleration is based on the second peak in the reference period, that is, the zero cross point that changes from negative to positive. For example, PRM2 is a dimensionless quantity T _ZAP2 / StepT obtained by normalizing the timing T _ZAP2 with a reference period Step T.

Regarding the third characteristic parameter PRM3 (Z-axis acceleration third maximum point timing ratio):
The characteristic parameter PRM3 is a parameter relating to the appearance timing of the third maximum point that appears when the vertical peak acceleration is based on the third peak in the reference period, that is, the zero cross point that changes from negative to positive. For example, PRM3 is a dimensionless amount T _ZAP3 / StepT obtained by normalizing the timing T _ZAP3 with a reference period StepT.

Regarding the fourth characteristic parameter PRM4 (Z-axis acceleration second minimum point timing ratio):
The characteristic parameter PRM4 exists between the second maximum point that appears second and the maximum point that appears third when the vertical valley acceleration is based on the second valley in the reference period, that is, the zero cross point that changes from negative to positive. This is a parameter related to the appearance timing of the minimum point. For example, PRM4 is a dimensionless amount T _ZAV2 / StepT obtained by normalizing the timing T _ZAV2 with the reference period Step T.

About the fifth characteristic parameter PRM5 (Z-axis acceleration first maximum point amplitude ratio):
The characteristic parameter PRM5 is a parameter related to the size of the maximum point that appears first when the vertical peak acceleration is based on the first peak in the reference period, that is, the zero cross point that changes from negative to positive. For example, PRM5 is a dimensionless quantity ZAP1 / ZAP2 obtained by normalizing the value ZAP1 of the point with the maximum value ZAP2 that appears second.

About the sixth characteristic parameter PRM6 (Z-axis acceleration third maximum point amplitude ratio):
The characteristic parameter PRM6 is a parameter relating to the size of the maximum point that appears third when the vertical axis acceleration is based on the third peak in the reference period, that is, the zero cross point that changes from negative to positive. For example, PRM6 is a dimensionless amount ZAP3 / ZAP2 obtained by normalizing the value ZAP3 of the point with the maximum value ZAP2 that appears second.

Regarding the seventh characteristic parameter PRM7 (Z-axis acceleration second minimum point amplitude ratio):
The characteristic parameter PRM7 exists between the second maximum and the third maximum appearing when the vertical axis acceleration is based on the second valley in the reference period, that is, the zero cross point that changes from negative to positive. This is a parameter related to the size of the local minimum point. For example, PRM7 is a dimensionless quantity ZAV2 / ZAP2 obtained by normalizing the value ZAV2 of the point with the value ZAP2 of the maximum point that appears second.

Regarding the eighth feature parameter PRM8 (Z-axis acceleration minimum point timing ratio):
The characteristic parameter PRM8 is a parameter related to the appearance timing of the minimum point in the reference period for the vertical axis acceleration. For example, PRM8 is a dimensionless amount T _ZAMN / StepT obtained by normalizing the timing T _ZAMN with a reference period Step T.

Regarding the ninth feature parameter PRM9 (Z-axis acceleration minimum point value):
The characteristic parameter PRM9 is a parameter related to the size of the minimum point in the reference period for the vertical axis acceleration. For example, PRM8 is the value ZAMN (unit: [m / sec ² ]) of the point.

About the tenth characteristic parameter PRM10 (Z-axis acceleration first maximum point minimum point difference):
The characteristic parameter PRM10 relates to the difference between the first peak in the reference period with respect to the vertical axis acceleration, that is, the size of the maximum point that appears first when the zero cross point that changes from negative to positive is used as a reference, and the size of the minimum point in the same reference period. It is a parameter. For example, PRM10 is ZAP1-ZAMN (unit: [m / sec ² ]).

Regarding the eleventh characteristic parameter PRM11 (Z-axis acceleration second maximum point minimum point difference):
The characteristic parameter PRM11 is the second peak in the reference period with respect to the vertical axis acceleration, that is, the size of the maximum point that appears second when the zero cross point that changes from negative to positive is used as the reference, and the size of the minimum point in the same reference period. It is a parameter related to the difference. For example, PRM11 is ZAP2-ZAMN (unit: [m / sec ² ]).

Regarding the twelfth characteristic parameter PRM12 (Z-axis acceleration third maximum point minimum point difference):
The characteristic parameter PRM12 is the third peak in the reference period with respect to the vertical axis acceleration, that is, the size of the maximum point that appears third when the zero cross point that changes from negative to positive is used as the reference, and the size of the minimum point in the same reference period. It is a parameter related to the difference. For example, the PRM 12 is ZAP3-ZAMN (unit: [m / sec ² ]).

Next, the four types of feature parameters from the thirteenth feature parameter PRM13 to the sixteenth feature parameter PRM16 will be described with reference to FIG.

About the thirteenth characteristic parameter PRM13 (Z-axis speed maximum point timing ratio):
The characteristic parameter PRM13 is a parameter relating to the appearance timing of the maximum point in the reference period for the first-order integral (vertical axis velocity) of the vertical axis acceleration. For example, the PRM 13 is a dimensionless amount T _ZVMX / StepT obtained by normalizing the timing T _ZVMX with the reference period StepT.

About the 14th characteristic parameter PRM14 (90% timing ratio before the maximum Z-axis speed):
The characteristic parameter PRM14 is a parameter related to the timing at which 90% of the maximum point is reached for the first time in the reference period with respect to the first-order integral (vertical axis velocity) of the vertical axis acceleration. For example, the PRM 14 is a dimensionless quantity T _ZVMX90 / StepT obtained by normalizing the timing T _ZVMX90 with the reference period StepT.

Regarding the fifteenth characteristic parameter PRM15 (80% timing ratio before the maximum Z-axis speed):
The characteristic parameter PRM15 is a parameter related to the timing at which 80% of the maximum point is reached for the first time in the reference period for the first-order integral (vertical axis velocity) of the vertical axis acceleration. For example, the PRM 15 is a dimensionless amount T _ZVMX80 / StepT obtained by normalizing the timing T _ZVMX80 with the reference period StepT.

Regarding the sixteenth characteristic parameter PRM16 (50% timing ratio before the maximum Z-axis speed):
The characteristic parameter PRM16 is a parameter related to the timing at which 50% of the maximum point is reached for the first time in the reference period with respect to the first-order integral (vertical axis velocity) of the vertical axis acceleration. For example, the PRM 16 is a dimensionless amount T _ZVMX50 / StepT obtained by normalizing the timing _TZVMX50 with the reference period StepT.

Next, with reference to FIG. 9, six types of feature parameters from the 17th feature parameter PRM17 to the 22nd feature parameter PRM22 will be described.

Regarding the seventeenth characteristic parameter PRM17 (Z-axis trajectory maximum point timing ratio):
The characteristic parameter PRM17 is a parameter related to the appearance timing of the maximum point in the reference period for the second-order integral (vertical axis trajectory) of the vertical axis acceleration. For example, the PRM 17 is a dimensionless amount _TZTMX / StepT obtained by normalizing the timing _TZTMX with the reference period StepT.

About the 18th characteristic parameter PRM18 (Z-axis trajectory minimum point timing ratio):
The characteristic parameter PRM18 is a parameter related to the appearance timing of the minimum point in the reference period for the second-order integral (vertical axis trajectory) of the vertical axis acceleration. For example, the PRM 18 is a dimensionless amount T _ZTMN / StepT obtained by normalizing the timing T _ZTMN with a reference period Step T.

About the 19th characteristic parameter PRM19 (Z-axis trajectory neutral point timing ratio):
The characteristic parameter PRM19 is a parameter related to the appearance timing of the zero point (orbit neutral point) in the reference period (not including the period boundary) for the second-order integral (vertical axis trajectory) of the vertical axis acceleration. For example, the PRM 19 is a dimensionless amount T _ZTNTL / Step T obtained by normalizing the timing T _ZTNTL with a reference period Step T.

Regarding the twentieth feature parameter PRM20 (Z-axis trajectory maximum point value):
The characteristic parameter PRM20 is a parameter related to the size of the maximum point in the reference period for the second-order integral (vertical axis trajectory) of the vertical axis acceleration. For example, the PRM 20 is the value ZTMX (unit: [m]) of the point.

Regarding the twenty-first feature parameter PRM21 (Z-axis trajectory minimum point value):
The characteristic parameter PRM21 is a parameter related to the size of the minimum point in the reference period for the second-order integral (vertical axis trajectory) of the vertical axis acceleration. For example, PRM21 is the value ZTMN (unit: [m]) of the point.

Regarding the twenty-second feature parameter PRM22 (Z-axis trajectory maximum point minimum point difference):
The characteristic parameter PRM22 is a parameter related to the difference between the maximum point size in the reference period and the minimum point size in the same reference period for the second-order integral (vertical axis trajectory) of the vertical axis acceleration. For example, the PRM 22 is ZTMX-ZTMN (unit: [m]).

As described above, the first feature parameter PRM1 to the twenty-second feature parameter PRM22 are feature parameters indicating the temporal change feature of the vertical axis acceleration. Below, the characteristic parameter which shows the characteristic of the time change of a longitudinal-axis acceleration is demonstrated.

Referring to FIG. 10, the six types of feature parameters from the 23rd feature parameter PRM23 to the 28th feature parameter PRM28 will be described.

About the 23rd characteristic parameter PRM23 (X-axis acceleration first maximum point timing ratio):
The characteristic parameter PRM23 is a parameter related to the appearance timing of the maximum point that appears first when the first peak that appears after the minimum point in the reference period with respect to the longitudinal acceleration, that is, the zero cross point that turns from the minimum point negative to positive, is used as a reference. . For example, the PRM 23 is a dimensionless amount T _XAP1 / StepT obtained by normalizing the timing T _XAP1 with the reference period StepT.

Regarding the 24th characteristic parameter PRM24 (X-axis acceleration second maximum point timing ratio):
The characteristic parameter PRM24 is a parameter relating to the appearance timing of the maximum point appearing second when the peak appears second after the minimum point in the reference period with respect to the longitudinal acceleration, that is, when the zero cross point that turns from the minimum point negative to positive is used as a reference. It is. For example, the PRM 24 is a dimensionless amount T _XAP2 / StepT obtained by normalizing the timing T _XAP2 with the reference period StepT.

Regarding the 25th characteristic parameter PRM25 (X-axis acceleration minimum point timing ratio):
The characteristic parameter PRM25 is a parameter related to the appearance timing of the minimum point that appears in the reference period with respect to the longitudinal acceleration. For example, the PRM 25 is a dimensionless amount T _XAMN / StepT obtained by normalizing the timing T _XAMN with the reference period StepT.

About the 26th characteristic parameter PRM26 (X-axis acceleration first maximum point value):
The characteristic parameter PRM26 is a parameter relating to the size of the maximum point that appears first when the first peak that appears after the minimum point in the reference period with respect to the longitudinal acceleration, that is, the zero cross point that changes from the minimum point negative to positive, is used as a reference. . For example, the PRM 26 is the value XAP1 (unit: [m / sec ² ]) of the point.

Regarding the 27th characteristic parameter PRM27 (X-axis acceleration second maximum point amplitude ratio):
The characteristic parameter PRM27 is a parameter relating to the size of the maximum point that appears second when the peak appears second after the minimum point in the reference period with respect to the longitudinal acceleration, that is, the zero cross point that turns from the minimum point negative to positive. It is. For example, the PRM 27 is a dimensionless amount XAP2 / XAP1 obtained by normalizing the value XAP2 of the point with the value of the first maximum point.

Regarding the twenty-eighth characteristic parameter PRM28 (X-axis acceleration minimum point value):
The characteristic parameter PRM28 is a parameter relating to the size of the minimum point in the reference period for the longitudinal acceleration. For example, the PRM 28 is the value XAMN (unit: [m / sec ² ]) of the point.

Finally, the five types of feature parameters from the 29th feature parameter PRM29 to the 33rd feature parameter PRM33 will be described with reference to FIG.

Regarding the 29th characteristic parameter PRM29 (X-axis speed maximum point timing ratio):
The characteristic parameter PRM29 is a parameter related to the appearance timing of the maximum point in the reference period for the first-order integration (front-rear axis velocity) of the front-rear axis acceleration. For example, the PRM 29 is a dimensionless amount T _ZVMX / StepT obtained by normalizing the timing T _XVMX with the reference period StepT.

About the 30th characteristic parameter PRM30 (90% timing ratio before the maximum X-axis speed):
The characteristic parameter PRM30 is a parameter related to the timing at which 90% of the maximum point is reached for the first time in the reference period with respect to the first order integration (front-rear axis velocity) of the front-rear axis acceleration. For example, the PRM 30 is a dimensionless amount T _XVMX90 / StepT obtained by normalizing the timing T _XVMX90 with the reference period StepT.

About the 31st characteristic parameter PRM31 (80% timing ratio before the maximum X-axis speed):
The characteristic parameter PRM31 is a parameter related to the timing at which 80% of the maximum point is reached for the first time in the reference period with respect to the first order integral (front-rear axis velocity) of the front-rear axis acceleration. For example, the PRM 31 is a dimensionless quantity T _XVMX80 / StepT obtained by normalizing the timing T _XVMX80 with the reference period StepT.

Regarding the thirty-second feature parameter PRM32 (X-axis speed minimum point timing ratio):
The characteristic parameter PRM32 is a parameter related to the appearance timing of the minimum point in the reference period for the first order integral (front-rear axis velocity) of the front-rear axis acceleration. For example, the PRM 32 is a dimensionless amount T _XVMN / StepT obtained by normalizing the timing T _XVMN with the reference period StepT.

About the thirty-third feature parameter PRM33 (50% timing ratio before the X-axis speed minimum point):
The characteristic parameter PRM33 is a parameter related to the timing at which 50% of the minimum point is reached for the first time in the reference period with respect to the first order integration (front-rear axis velocity) of the front-rear axis acceleration. For example, the PRM 33 is a dimensionless amount T _XVMN50 / StepT obtained by normalizing the timing T _XVMN50 with the reference period StepT.

により、上述した平均骨盤傾斜角θ[degree]が得られる。ここで重み付け加算に用いられる重み係数（ｋ_１，ｋ_２，．．．，ｋ_３２，ｋ_３３）および定数項ｋ_０（ｋ_０～ｋ_３３は、ゼロを含む実定数。）は、後述するようにステップＳ４５ａ～Ｓ４５ｃ（図１３）において、被測定者の歩幅および／または基準期間に応じて適切に決定される。 The above are the 33 types of feature parameters derived by the control unit 110 in the present embodiment. These feature parameters PRM1 to PRM33 are weighted and added in step S46 (FIG. 13) as will be described later. In the present embodiment, weighted addition for the characteristic parameters PRM1 to PRM33 represented by the following equations:

Thus, the above-described average pelvic inclination angle θ [degree] is obtained. Here, the weighting factors (k ₁ , k ₂ ,..., K ₃₂ , k ₃₃ ) and constant terms k ₀ (k ₀ to k ₃₃ are real constants including zero) used for weighted addition will be described later. As described above, in steps S45a to S45c (FIG. 13), it is appropriately determined according to the stride and / or the reference period of the measurement subject.

Referring back to FIG. 13, the processing after deriving the characteristic parameters will be described. In step S45a, the control unit 110 uses the vertical axis trajectory time series data (FIG. 9) to evaluate the step width of the measurement subject. For example, the control unit 110 estimates the step length of the person to be measured based on the magnitude of the difference (ZTPP in FIG. 9) between the maximum value and the minimum value of the vertical axis trajectory in one reference period. The estimation may be performed using a conversion table (conversion formula) between the difference ZTPP between the maximum value and the minimum value of the vertical axis trajectory experimentally obtained in advance and the stride.

Next, in step S45b, the control unit 110 obtains a reference period corresponding to one step in the walking cycle of the subject using the vertical axis acceleration time-series data (FIG. 7). The control part 110 should just obtain | require the time interval between the zero cross point which changes from negative to positive, and the zero cross point which changes from negative to positive, ie, reference | standard period StepT, for example.

Next, in step S45c, the control unit 110 determines a weighting coefficient used for calculating the average pelvic inclination angle θ according to the above equation 1. FIG. 14 is a reference table used for determining the weight coefficient set. First, the control unit 110 determines whether the step length of the measurement subject is “small”, “medium”, or “large” based on the step width of the measurement subject evaluated in step S45a. Next, based on the determination of the stride described above, the control unit 110 further determines whether the reference period is “short”, “medium”, or “long”, thereby using the weight coefficient set K _mn to be used. (M: 1, 2, 3, n: 1, 2, 3) is determined. Each of the weighting coefficient sets K _mn is a set including weighting coefficients (k ₁ , k ₂ ,..., K ₃₂ , k ₃₃ ) and a constant term k ₀ used in Equation 1. For example, the weighting coefficient set K ₁₁ are stride relatively small and reference period the average pelvic tilt angle for the Formula 1 the group consisting of a number of the subject having the common feature that relatively short θ can be calculated accurately Thus, it is a set of values of weight coefficients and constant terms obtained experimentally. Similarly, the weight coefficient set K ₃₃ is obtained from a large number of subjects having a common feature that the stride is relatively large and the reference period is relatively long so that the average pelvic inclination angle θ can be accurately calculated by Equation 1. It is a set of values of weighting factors and constant terms obtained experimentally for a group. Each of the other weight coefficient sets is a set of weight coefficient values and constant term values obtained experimentally in the same manner.

Note that the processing from step S45a to S45c is processing performed to select a weighting factor used in the evaluation of Equation 1 when a plurality of weighting factor sets are provided according to the step length and / or the reference period of the measurement subject. . Therefore, when there is one weighting coefficient set used by the control unit 110, the processing from steps S45a to S45c may be skipped.

ここで、ＰＲＭ_ｉ（ｉ：１～３３の整数）は、ステップＳ４４で求めた特徴パラメータであり、ｋ_ｍｎ，０およびｋ_ｍｎ，ｉ（ｉ：１～３３の整数）は、重み係数セットＫ_ｍｎに含まれる定数項および重み係数である。 In step S46, the control unit 110 uses the feature parameters PRM1 to PRM33 calculated in step S44 and the weighting coefficient set K _mn determined in step S45c to

Here, PRM _i (i: an integer from 1 to 33) is the feature parameter obtained in step S44, and k _{mn, 0} and k _{mn, i} (i: an integer from 1 to 33) are weight coefficient sets K _It is a constant term and a weighting factor included in _mn .

Returning to FIG. 12, in step S5, the control unit 110, based on the average pelvic inclination angle θ obtained in step S4, indicates that the posture of the person to be measured is “forward tilt” or “middle”. Or “backward tilt” (see FIG. 5A).

Finally, the control unit 110 outputs (transmits) the evaluation result obtained in step S5 to the smartphone 200 (step S6).

When the smartphone 200 receives the information from the activity meter 100, the smartphone 200 displays the evaluation result together with the total score on the display unit 240. For example, a message such as “Your walking posture has a tendency to lean forward” is displayed on the display unit 240 of the smartphone 200. It should be noted that the display unit 240 may display such that the forward / backward inclination can be intuitively understood using illustrations, animations, and the like. Further, the display unit 240 may display the average pelvic inclination angle θ itself obtained in step S46 or an evaluation numerical value obtained by appropriately processing the value of θ.

Referring to the display content of the display unit 240, the user can know whether the posture during walking is forward leaning or backward leaning.

Thus, as a result of earnest research, the inventor has obtained one or a plurality of feature quantities (feature parameters) that capture the characteristics of the waveform shape of the time-varying acceleration output from the acceleration sensor attached to the waist of the subject. It was experimentally found that the result of the weighted addition of has a good correlation with the inclination angle of the pelvis when the subject walks. According to this walking posture meter 1, it is possible to appropriately evaluate whether or not the posture of the measurement subject during walking is forward / backward. Moreover, since this walking posture meter 1 performs the evaluation based on the output of the acceleration sensor 112, it can be easily evaluated regardless of a large facility such as motion capture.

In the above-described embodiment, 33 types of characteristic parameters PRM1 to PRM33 are calculated, but the present invention is not limited to this. For example, instead of calculating all 33 types of feature parameters PRM1 to PRM33, only a part is calculated, and only the feature parameters are used to evaluate the tendency of the subject to lean forward / backward during walking. May be.

In the above-described embodiment, the weight coefficient set K _mn is set in accordance with the step length and / or the reference period of the measurement subject, but the present invention is not limited to this. Instead of the reference period, a walking cycle (corresponding to a period of two steps including the left foot reference period and the right foot reference period) is used, and the weighting coefficient sets K _mn are respectively set according to the step length and / or the walking cycle of the measurement subject. May be set.

In the above-described embodiment, the activity meter 100 and the smartphone 200 communicate with each other by BLE communication, but the present invention is not limited to this. For example, the activity meter 100 and the smartphone 200 may communicate with each other when the smartphone 200 and the activity meter 100 approach each other by NFC (Near Field Communication).

In the above-described embodiment, the walking posture meter of the present invention is configured as a system including the activity meter 100 and the smartphone 200. However, the present invention is not limited to this.

For example, the walking posture meter of the present invention may be configured with only the smartphone 200. In that case, the smartphone 200 includes an acceleration sensor. Further, the memory 220 of the smartphone 200 has a program for quantitatively evaluating whether or not the human walking posture is a correct posture in the control unit 210, more specifically, the forward / backward tilt of the posture while walking. Install a program to assess the degree. Thereby, the walking posture meter of this invention can be comprised compactly and compactly.

The program can be recorded as application software on a recording medium such as a CD, a DVD, or a flash memory. The application software recorded on this recording medium is installed in a substantial computer device such as a smartphone, personal computer, PDA (Personal Digital Assistance), etc., so that the human walking posture is correct in these computer devices. It is possible to execute a method for quantitatively evaluating whether or not.

1 walking posture meter 100 activity meter 112 acceleration sensor 110 control unit CPU
DESCRIPTION OF SYMBOLS 120 Memory 180 BLE communication part 200 Smartphone 210 Control part 220 Memory 290 Network communication part 230 Operation part 240 Display part 280 BLE communication part

Claims (18)

A walking posture meter that evaluates a walking posture of a measurement subject,
An acceleration sensor mounted on the midline of the waist of the person being measured;
Using the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration output from the acceleration sensor, an amount corresponding to the forward / backward inclination of the posture of the person being measured is calculated. A walking posture meter, comprising: The walking posture meter according to claim 1,
The arithmetic unit derives a characteristic parameter indicating characteristics of the temporal change of the vertical axis acceleration and the temporal change of the longitudinal axis acceleration, and calculates an amount corresponding to the tilt in the front-rear direction based on the characteristic parameter; Walking posture meter characterized by The walking posture meter according to claim 2,
The operation unit derives a plurality of types of feature parameters, and calculates an amount corresponding to the tilt in the front-rear direction based on an amount obtained by weighted addition of the derived types of feature parameters. Walking posture meter. In the walking posture meter according to claim 3,
The computing unit estimates a step length of the person to be measured based on a temporal change waveform of the vertical axis acceleration, and changes a weighting factor used for the weighted addition based on the estimated step length. A characteristic walking posture meter. The walking posture meter according to claim 4,
The calculation unit is configured to calculate one step of the walking cycle of the measurement subject or the walking cycle based on at least one of the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration. And a weighting factor used for the weighted addition is changed based on the obtained walking cycle or reference period. In the walking posture meter according to any one of claims 1 to 5,
An amount corresponding to the tilt in the front-rear direction is an angle formed by the pelvis of the person being measured and a horizontal position during walking. In the walking posture meter according to any one of claims 2 to 5,
In the temporal change waveform of the vertical axis acceleration with the upper side being positive, the arithmetic unit is from the appearance timing of the zero cross point at which the acceleration value changes from negative to positive to the next occurrence timing of the zero cross point at which the negative value changes from positive to positive. Is determined as a reference period corresponding to one step of the person to be measured, and the values and timings of the three local maximum points appearing in the temporal change waveform of the vertical axis acceleration in the reference period, and the three local maximums A walking posture meter, wherein the feature parameter is derived using at least one of a value and an appearance timing of two minimum points between points, and a value and an appearance timing of a minimum point. The walking posture meter according to claim 7,
The calculation unit may include at least one of a ratio of the time from the start point of the reference period to the appearance timing of the three maximum points, the two minimum points, and the minimum point and the length of the reference period. Any one of them is used as the feature parameter. The walking posture meter according to claim 7,
The arithmetic unit uses at least one of absolute values of differences between the values of the three maximum points and the value of the minimum point as the feature parameter. In the walking posture meter according to any one of claims 2 to 5,
The calculation unit obtains a temporal change waveform of the vertical axis velocity from a temporal change waveform of the vertical axis acceleration,
In the temporal change waveform of the vertical axis acceleration with the upper side being positive, the arithmetic unit is from the appearance timing of the zero cross point at which the acceleration value changes from negative to positive to the next occurrence timing of the zero cross point at which the negative value changes from positive to positive. Is determined as a reference period corresponding to one step of the person to be measured, and the appearance timing of the maximum point of the temporal change waveform of the vertical axis velocity in the reference period is 90% of the value of the maximum point. At least one of an appearance timing of a point having a value, an appearance timing of a point having a value of 80% of the value of the maximum point, and an appearance timing of a point having a value of 50% of the value of the maximum point A walking posture meter, wherein the feature parameter is derived using one. The walking posture meter according to claim 10,
The calculation unit is configured to start each point of the reference period, the maximum point, the point having the value of 90%, the point having the value of 80%, and the point having the value of 50%. A walking pose meter using at least one of the ratio of the time of the reference period and the length of the reference period as the feature parameter. In the walking posture meter according to any one of claims 2 to 5,
The calculation unit obtains a temporal change waveform of the vertical axis trajectory from the temporal change waveform of the vertical axis acceleration,
In the temporal change waveform of the vertical axis acceleration with the upper side being positive, the arithmetic unit is from the appearance timing of the zero cross point at which the acceleration value changes from negative to positive to the next occurrence timing of the zero cross point at which the negative value changes from positive to positive. Is determined as a reference period corresponding to one step of the person to be measured, and the maximum point value and appearance timing, minimum point value and appearance timing of the temporal change waveform of the vertical axis trajectory in the reference period And the feature parameter is derived using at least one of an absolute value of a difference between the maximum point and the minimum point. The walking posture meter according to claim 12,
The computing unit calculates at least one of a ratio of a time from the start point of the reference period to the appearance timing of each of the maximum point and the minimum point and the length of the reference period as the characteristic parameter. It is used as a walking posture meter. In the walking posture meter according to any one of claims 2 to 5,
In the temporal change waveform of the vertical axis acceleration with the upper side being positive, the arithmetic unit is from the appearance timing of the zero cross point at which the acceleration value changes from negative to positive to the next occurrence timing of the zero cross point at which the negative value changes from positive to positive. Is determined as a reference period corresponding to one step of the person to be measured, and the minimum point value and appearance timing of the temporal change waveform of the longitudinal acceleration in the reference period, and after the minimum point A walking pose meter, wherein the feature parameter is derived using at least one of the value of the two maximum points that appear and the appearance timing. The walking posture meter according to claim 14,
The calculation unit calculates at least one of a ratio of a time from the start point of the reference period to the appearance timing of each of the minimum point and the two maximum points and the length of the reference period. A walking pose meter that is used as a feature parameter. In the walking posture meter according to any one of claims 2 to 5,
The calculation unit obtains a temporal change waveform of the longitudinal speed from the temporal change waveform of the longitudinal acceleration,
In the temporal change waveform of the vertical axis acceleration with the upper side being positive, the arithmetic unit is from the appearance timing of the zero cross point at which the acceleration value changes from negative to positive to the next occurrence timing of the zero cross point at which the negative value changes from positive to positive. Is determined as a reference period corresponding to one step of the measured person, and the appearance timing of the maximum point of the temporal change waveform of the longitudinal speed in the reference period, a value of 90% of the value of the maximum point An appearance timing of a point having a value of 80% of the value of the maximum point, an appearance timing of a point having a value of 80% of the value of the maximum point, and an appearance timing of a point having a value of 50% of the value of the minimum point A walking posture meter, wherein the characteristic parameter is derived using at least one of them. The walking posture meter according to claim 16,
The calculation unit includes the maximum point from the start point of the reference period, a point having a value of 90% of the value of the maximum point, a point having a value of 80% of the value of the maximum point, the minimum point, and At least one of the ratios of the time until the appearance timing of each point having a value of 50% of the value of the minimum point and the length of the reference period is used as the feature parameter. Walking posture meter. A program for causing a computer to execute a method for evaluating a walking posture of a measurement subject,
The method
Obtaining an output of an acceleration sensor mounted on the midline of the subject's waist;
Using the temporal change waveform of the vertical axis acceleration and the temporal change waveform of the longitudinal axis acceleration output from the acceleration sensor, an amount corresponding to the forward / backward inclination of the posture of the person being measured is calculated. A program comprising the steps of:

Images