WO2023100565A1

WO2023100565A1 - Running form evaluation system, program, and method

Info

Publication number: WO2023100565A1
Application number: PCT/JP2022/040406
Authority: WO
Inventors: 邦彦加地
Original assignee: リオモインク
Priority date: 2021-11-30
Filing date: 2022-10-28
Publication date: 2023-06-08
Also published as: JP2023080950A

Abstract

[Problem] To perform more accurate movement analysis for running sports by minimizing the effects of a measurement error and impact noise from ground contact with the left and right legs of a subject during running, without requiring construction of cost-incurring large devices or expensive systems. [Solution] The present invention comprises: a pair of body motion sensors 40a and 40b that are respectively mounted on the thighs of both legs of a wearer and are capable of detecting three-dimensional displacement or rotation of each site; a ground-contact state detection unit 117h that detects non-ground-contact states of the wearer's legs on the basis of the detection results from the pair of body motion sensors 40a and 40b, and extracts the detection results of the detected non-ground-contact states, respectively, as pieces of ground-contact state data D1-D6 indicating ground-contact-side states; an indicator calculation unit 117g that calculates an indicator for evaluating a running form on the basis of the pieces of ground-contact state data D1-D6 extracted by the ground-contact state detection unit 117h; and an output interface 111 that displays or outputs the indicator calculated by the indicator calculation unit 117g.

Description

Running form evaluation system, program and method

The present invention relates to a system, program, and method for easily and inexpensively estimating the ground reaction force during running and evaluating a running form that enables efficient and fast running.

In general, in running competitions, it is important to press the ground in a short period of time in order to run efficiently and fast, and how to move each part of the body, how to contact the ground, how to exert each muscle output, etc. It will happen. For this reason, in the motion analysis of running competitions, the running form is conventionally evaluated by measuring the final output with a floor reaction force meter.

However, the floor reaction force meter is very expensive and needs to be embedded in a predetermined position on the track. Since it is required to be grounded, there is a problem that it is very difficult to accurately measure changes in stress at arbitrary distances.

In order to solve this problem, we built a floor reaction force meter into a device called a treadmill that moves a belt-conveyor-like step by the power of a motor, and continuously measures the floor reaction force at any installation position. Systems have been developed to allow this. However, even with a device using such a treadmill, the cost of the equipment is still high, and because the floor moves by power, the running form of the subject is different from that of a normal ground, which requires the user to kick the ground by themselves. Many studies have reported that the data are different and accurate data cannot be obtained.

On the other hand, conventionally, as disclosed in Patent Document 1, there has been an attempt to replace the measurement of the floor reaction force meter by having the athlete wear an acceleration sensor on the body and measure the movement of the body. ing. In the motion analysis system disclosed in Patent Document 1, a motion analysis device that incorporates an inertial measurement unit (IMU) in the subject's torso (for example, the right waist, the left waist, or the center of the waist). to capture the movement of the subject during running (including walking), calculate the speed, position, posture angle (roll angle, pitch angle, yaw angle), etc., analyze the subject's movement, After calculating the running state (various types of running information), change points of the running state are detected, and motion analysis information, which is running information including the changing points of the running state, is generated.

According to the system disclosed in Patent Document 1, it is possible to detect, for example, characteristic points such as landing, stepping, and taking off in the subject's running motion. Specifically, the landing is detected at the timing when the vertical acceleration (detection value of the z-axis of the acceleration sensor) changes from a positive value to a negative value, and after landing, the vertical acceleration peaks in the negative direction. After that, it is possible to detect stepping on when the traveling direction acceleration reaches a peak, and to detect taking off (kicking) when the vertical acceleration changes from a negative value to a positive value. Then, motion analysis information can be obtained based on the motion index representing how close the timing of hip rotation is to the timing of kicking. For example, in a running style in which the leg is flowing, the opposite leg is still behind the body when one leg is landed, so if the hip rotation timing comes after the kick, the running style in which the leg is flowing is used. I can judge.

JP 2018-143537 A

However, in the technique disclosed in Patent Document 1, a single inertial measurement unit is attached to the torso portion of the subject (for example, the right hip, left hip, or central portion of the hip) to measure the changing point of the running state. Since it is a sensor, the impact noise caused by landing from both the left and right legs is very large, and the orientation of the sensor is not stable due to the rotation of the waist, and the analysis results have certain limitations. For example, if the user leans forward while running, the sensor that had been vertical until then will rotate and pick up the acceleration in the front-back direction. I had a problem with it.

Therefore, the present invention solves the above-mentioned problems, and does not require the construction of a costly large-scale device or an expensive system. It is an object of the present invention to provide a running form evaluation system, a program, and a method for achieving more accurate motion analysis in running competitions by minimizing the effects of measurement errors.

In order to solve the above problems, the present invention detects a wearer's body movement and evaluates running form based on changes in body movement. A plurality of pairs of body motion sensors capable of detecting three-dimensional displacement or rotation of each part, and a non-grounded state or a grounded state of the wearer's legs based on the detection results of the pair of body motion sensors. a grounded state detection unit for extracting as grounded state data the detected grounded state or each detection result related to the grounded state determined based on the grounded state; and detecting the running form based on the extracted grounded state data. It is characterized by comprising an index calculation unit that calculates an index to be evaluated, and an output device that displays or outputs the index calculated by the index calculation unit.

Further, the present invention is a method for detecting body movements of a wearer and evaluating running form based on changes in body movements,
(1) A plurality of pairs of body motion sensors capable of detecting three-dimensional displacement or rotation of each part are attached to each of the thighs of both legs of the wearer, and a grounded state detection step of detecting a non-grounded state of the leg of the wearer based on the detection result of the movement sensor, and extracting each detection result related to the non-grounded state as grounded state data indicating the state of the grounded side;
(2) The index calculation unit extracts each detection result related to the non-grounded state detected by the grounded state detection unit in the grounded state detection step as grounded state data, and calculates the running form based on the extracted grounded state data. an index calculation step of calculating an index for evaluating the
(3) An output step in which an output device displays or outputs the index calculated by the index calculation unit.

The system and method according to the present invention described above can be realized by executing the program of the present invention written in a predetermined language on a computer.
That is, the present invention is a program for detecting body movement of the wearer and evaluating running form based on changes in body movement, comprising:
a plurality of pairs of body motion sensors mounted on each of the thighs of both legs of the wearer and capable of detecting three-dimensional displacement or rotation of each part;
a grounded state detection unit that detects a non-grounded state of the wearer's legs based on detection results from the pair of body motion sensors;
Each detection result related to the non-grounded state detected by the grounded state detection unit is extracted as grounded state data indicating the state of the grounded side, and an index for evaluating the running form is calculated based on the extracted grounded state data. an index calculation unit that
It is characterized by functioning as an output device that displays or outputs the index calculated by the index calculation unit.

Such a program of the present invention is installed in an IC chip or memory device of a mobile terminal device, a smart phone, a wearable terminal, a tablet PC or other information processing terminal, a general-purpose computer such as a personal computer or a server computer, and executed on the CPU. Thus, a system having the functions described above can be constructed to carry out the method according to the present invention.

As described above, in this invention, the positions of the sensors that were conventionally attached to the trunk such as the waist and chest are set to both thighs, and measurements are performed with the opposite leg to the leg that is on the ground. In other words, the sensor on the left leg measures the ground reaction force when the right foot is in contact with the ground. Thus, according to the present invention, by using the data of the opposite leg, the landing impact is not received. From the grounded leg, sensors are attached to the tip of the ankle, knee, grounded hip joint, and reverse hip joint. The movement of the left hip joint) can be measured smoothly and accurately when the right ground is touched. For example, since the rotation of the waist on the horizontal plane does not greatly affect the acceleration in the vertical direction of the thigh of the sensor attached to the thigh, the desired acceleration can be detected because the acceleration is almost vertical.

As a result, according to the present invention, it is possible to minimize the effects of impact noise and measurement errors caused by landing from both the left and right legs during running without the need to construct a costly large-scale device or an expensive system. , more accurate motion analysis of running competitions can be realized.

It is an explanatory view showing a usage mode of the running form evaluation system according to the embodiment. FIG. 4 is an explanatory diagram showing an overview of parameters detected by sensors used in the embodiment; 3 is a block diagram showing the internal configuration of each device according to the embodiment; FIG. It is a sequence diagram showing a running form evaluation method according to the embodiment. FIG. 5 is a flowchart showing motion analysis processing according to the embodiment; FIG. 4 is an explanatory diagram showing a part of body motion reproduction data (R_Thing-AccZ and L_Thing-AccZ) acquired in the embodiment; FIG. 4 is an explanatory diagram showing part of body motion reproduction data acquired in the embodiment; FIG. 4 is an explanatory diagram showing part of body motion reproduction data acquired in the embodiment;

A first embodiment of the present invention will be described below with reference to the drawings. In this embodiment, the information terminal device 100 is used to apply the present invention to motion analysis of a running competition, thereby providing a running form evaluation system that enables coaching for training in a running competition. It should be noted that the embodiments shown below are examples of devices and the like for embodying the technical idea of the present invention, and the technical idea of the present invention is based on the material, shape, structure, and It does not specify the arrangement etc. to the following. The technical idea of this invention can be modified in various ways within the scope of claims.

(Configuration of running form evaluation system)
FIG. 1 shows a usage mode of a running form evaluation system using an information terminal device 100 according to this embodiment, and FIG. 2 shows an outline of parameters detected by the sensor according to this embodiment. FIG. 3 is a block diagram showing the internal configuration of each device. As shown in FIGS. 1 to 3, the running form evaluation system according to the present embodiment includes an information terminal device 100 used by the wearer 1 and information terminal devices 100 worn on both thighs of the wearer 1. body motion sensors 40 (40a and 40b) wirelessly connected to each other.

Then, in this system, using these body motion sensors 40, motion analysis in a running competition is performed, and an index for evaluating running form is provided based on the amount of deviation from the reference value. This index is obtained by referring to index data. In this embodiment, the index data is referred to based on the reference value or threshold value acquired from the body motion data, and the index for evaluating the stabilizing ability is obtained. This reference value can be selected from an average value, maximum value, minimum value, or any representative value within a predetermined period based on the setting operation by the wearer.

(Configuration of each device)
The specific internal configuration of each device constituting this system will be described below.
(1) Body Motion Sensors

Body motion sensors

40a and 40b are a pair of sensors that are attached to both left and right thighs of the wearer 1 and detect three-dimensional displacement or rotation of each thigh. In this embodiment, the

body motion sensors

40a and 40b are attached to the front surfaces of the left and right thighs of the wearer. These

motion sensors

40a and 40b are equipped with a 3-axis accelerometer that measures the acceleration of an object, a 3-axis gyroscope that detects the angular velocity of an object, and a 3-axis magnetic sensor that measures the magnitude and direction of a magnetic field. Axis movement can be detected.

These body motion sensors 40 (

body motion sensors

40a and 40b) each have a wireless communication unit, as shown in FIG. This wireless communication unit has an internal antenna, and the function of executing a data communication protocol for short-range wireless communication such as BTLE (Bluetooth (registered trademark) Low Energy, Bluetooth (registered trademark) 4.0) enables information terminal equipment 100 and communication processing are possible. In this embodiment, the wireless communication unit of each body movement sensor 40 employs BTLE as a protocol for low power consumption communication, but ANT, ANT+, etc., can also be employed. Ordinary Bluetooth (registered trademark) can also be adopted.

In the present embodiment, basically, the system can be constructed within the range constructed by short-range wireless communication between the information terminal device 100 and the body motion sensor 40, and the server or the like on the communication network is It is not connected during actual measurement, and the system can be operated as a so-called stand-alone off-line.

(2) Information Terminal Device FIG. 3 shows the internal configuration of the information terminal device according to this embodiment. The information terminal device 100 according to the present embodiment is, for example, a small terminal device such as a smartphone, and may be a general rectangular terminal device. Various forms can be adopted, such as a mount type to which it is attached. It should be noted that this information terminal device may be stored in a container such as a bag when only recording body motion data during running.

Specifically, the information terminal device 100 includes a wireless interface 113, a control unit 117, a memory 114, an output interface 111, and an input interface 112, as shown in FIG. More specifically, the information terminal device 100 according to the present embodiment has a function of collecting detection results detected by each body motion sensor 40, and performs mutual communication processing with each body motion sensor 40 via the wireless interface 113. The result of detection by each body motion sensor 40 can be obtained by pressing the button. The memory 114 of the information terminal device 100 functions as a body motion recording unit that records the detection results of the body motion sensor 40 as body motion data. Here, the body motion data is the primary data detected by various sensors, and the body motion reproduction data is the secondary data obtained by recording and analyzing this body motion data, extracting necessary information, and correcting it. be.

Sensor identification information for identifying each body motion sensor 40 is added to the detection result transmitted from each body motion sensor 40, and the identification information is accumulated in the memory 114 of the information terminal device 100. When the control unit 117 acquires the detection result from the wireless interface 113 , it can be determined from which body motion sensor 40 the detection result is acquired. Note that this identification information includes mounting site information that specifies the mounting site of each sensor, and it is possible to calculate body movement reproduction data based on this mounting site information. Further, the body motion data also includes time information when the detection result is obtained from each body motion sensor 40 .

The wireless interface 113 is a module that controls transmission and reception of various types of information via a communication network and short-range wireless communication such as wifi and Bluetooth (registered trademark). Also, it transmits and receives data to and from the server device or the like by 3G communication. Further, the information terminal device 100 has an output interface 111 and an input interface 112 . The input interface 112 is a device for inputting user operations, such as a mouse, keyboard, operation buttons, and touch panel. Also, the output interface 111 includes devices such as a display and speakers that output video and audio. In particular, the output interface 111 includes a display section such as a liquid crystal display, and the display section is superimposed on a touch panel that is an input interface.

The display unit connected to the output interface 111 is an output device that displays or outputs the analysis result of the body motion reproduction data, and displays the display information generated by the display information generation unit 117e through the output interface 111. A palm is displayed on this display unit on a display built in the information terminal device 100 or an external display connected to the outside.

On the other hand, the input interface 112 may be provided with a video acquisition unit. This moving image acquisition unit is a device that captures and records the body movements of the wearer and acquires moving image data. In addition to being able to check forms, etc., it is also used to synchronize body movement data acquired by the sensor, which will be described later, with video shot by the camera. The moving image data acquired here includes video data in which video is recorded, audio data recorded together with the video, and metadata such as shooting time, end time, and time stamp such as elapsed time.

A built-in camera 115 built in the information terminal device 100 and an external camera can be connected to the input interface 112 , and moving image data shot by these shooting means are acquired and stored in the memory 114 . or provided for processing in the control unit 117 . The video data acquired from the external camera includes streaming data that is sequentially acquired in real time at the time of shooting, as well as video data that has been captured by the external camera and stored in a file format and downloaded after shooting. included.

Further, in the present embodiment, the information terminal device 100 has a function of analyzing the wearer's body movement based on the body movement data acquired from each sensor and generating body movement reproduction data. Specifically, as shown in FIG. 3, the information terminal device 100 includes a control unit 117. The control unit 117 is an arithmetic processing unit such as a CPU that performs various calculations necessary for controlling each unit. be. Each function of the information terminal device 100 is virtually constructed on the control unit 117 by executing the running form evaluation program of the present invention in the control unit 117 . More specifically, the control unit 117 causes the body motion data acquisition unit 117a, the body motion calculation unit 117b, the analysis unit 117d, and the display information generation unit 117e to be virtually generated by executing the running form evaluation application. be built.

The body motion data acquisition unit 117a is a module that acquires and records body motion data from the body motion sensors 40 via the wireless interface 113. In this embodiment, the body motion data acquisition unit 117a communicates wirelessly with the body motion sensors 40a and 40b. and obtain body motion data, which is the result of these detections. The body motion data acquisition unit 117a functions as a body motion data recording unit, temporarily accumulates body motion data in the memory 114, and stores each detection result by the body motion sensor 40 in the body motion calculation unit 117b. or sent to

The body motion calculation unit 117b calculates body motion data, which are detection results of the

body motion sensors

40a and 40b accumulated in the memory 114, which is a body motion recording unit. This is a module that calculates the wearer's body motion as body motion reproduction data based on acceleration, angular velocity, angular acceleration, and the like. Here, the body motion data, which are the results of detection by the body motion sensor 40, are values measured by a so-called 9-axis sensor. They are the size, the angular velocity of the object (size, direction, center position), and the magnitude and direction of the magnetic field (azimuth).

The body movements calculated here include the rotation of the left and right thighs centering on the hip joints during running, vertical, horizontal, and forward/backward movement and acceleration, angular velocity of rotation, temporal changes in this angular velocity, and The smoothness of that change is included. More specifically, in this embodiment, the

body motion sensors

40a and 40b are attached to the left and right thighs, and as shown in FIG. It becomes a leg exercise.

In this embodiment, the body motion calculation unit 117b calculates the body motion of the wearer based on the body motion data, which are the results of each detection by the body motion sensor 40, and the amount of deviation from the reference value of these body motion sensors 40. Calculated as reproduction data. At this time, the body motion calculator 117b specifically calculates the relative displacement, velocity, acceleration, and rotation ( Angular momentum), body motion reproduction data is calculated based on the trajectory of displacement (body motion) of each part of the body.

Specifically, the body motion calculation unit 117b first uses the body motion data, which are the detection results of the

body motion sensors

40a and 40b, to perform processing to detect feature points in the wearer's running. Characteristic points in the wearer's running are detection values that indicate characteristic behavior detected by the body motion sensor, changes in the detected values, and portions of data corresponding to time (time). changes in acceleration based on the characteristic body movements of the wearer. In addition, the body motion calculation unit 117b performs a process of calculating each value of contact time and impact time based on the timing of the detected feature point. Specifically, based on the timing at which the characteristic points are detected from the calculated data including the characteristic points, the length of time that the value continues and the rate of change within a certain unit time are used to calculate the contact duration and impact duration. Calculate each value.

In this embodiment, the body motion calculation unit 117b includes a reference value setting unit 117c that sets a stability reference value for evaluating reproducibility of body motion based on the body motion reproduction data, a pair of body motion sensors 40a and A grounded state detection unit 117h is provided for detecting the non-grounded state of the wearer's leg based on the detection result by 40b.

The reference value setting unit 117c causes the index calculation unit 117g to set a value selected from the average value, the maximum value, the minimum value, or any representative value within a predetermined period based on the setting operation by the wearer 1 as the reference value. set. In this reference value setting operation, for example, the same action is repeated several times at predetermined time intervals, and the average value, minimum value, maximum value, and the value at the time when the wearer 1 thinks the best is ideal. can be a value. In addition, in this embodiment, a database is provided in which reference values of others (a large number of other users, advanced users, professionals, etc.) and body motion reproduction data are linked and accumulated, and the ideal values of advanced users and professionals are provided. You can also call up any numerical value from the database and set it, and also search for body movement reproduction data similar to the input body movement reproduction data, and call the reference value linked to that body movement reproduction data. It can be set in the reference value setting unit 117c.

At this time, the recalled body movement data of the other person, which is the basis of the reference value, is accumulated in chronological order. , a reference value can be set for running forms with similar characteristics and tendencies. By tracking the progress of changes in the body movement reproduction data of the other person used for this setting, it is possible to simulate and predict the results of correcting the running form using the reference value.

The grounded state detection unit 117h detects the non-grounded state of the wearer's leg based on the detected feature points, and uses each detection result related to the detected non-grounded state as grounded state data indicating the state of the grounded side. This is the module to extract. Specifically, the grounding state detection unit 117h in the present embodiment detects values indicating characteristic behavior based on detection values (body motion data) of an acceleration sensor and an angular velocity sensor constituting body motion sensors, and changes in the detected values, Detect feature points based on time (time), and based on that timing, the contact time and impact time, their rate of change, and the temporal range of the contact state or non-contact state depending on the periodicity is specified, and a flag is set to the data for the time length included in the range specified as the non-grounded state indicating the state of the grounded side. Data in which this flag is set is extracted as ground contact state data D1 to D6, and the analysis unit 117d generates body movement reproduction data including this ground contact state data.

The body motion data acquired by the body motion calculation unit 117b described above is input to the analysis unit 117d, and based on the relative displacement, velocity, acceleration, angular velocity, etc., the moment of each left and right thigh of the wearer 1 is calculated. Body movement reproduction data is generated from relative relative displacement (distance and rotation) and relative rotational motion of the back and main parts. Then, the analysis unit 117d evaluates the running form based on body movement timing, posture collapse, etc., using primary data such as body movement data and contact state data, and secondary data such as body movement reproduction data. .

Specifically, the analysis unit 117d is a module that analyzes each element of the body movement of the wearer 1 for each item based on body movement data, grounding state data, and body movement reproduction data. In the present embodiment, the analysis unit 117d calculates the deviation amount from the reference value setting unit 117c, analyzes the reproducibility of the body movement, and changes the angular velocity of each thigh extracted by the body movement calculation unit 117b. It functions as a characteristic analysis unit that analyzes the characteristics of oscillation amplitude and fluctuation. is synchronized with the recorded video data, and then displayed or output on an output device via the display information generation unit 117e.

As another analysis method by the analysis unit 117d, it is possible to generate three-dimensional data in which the wearer 1 is displayed three-dimensionally, or two-dimensional data projected onto the XY plane may be used. It may generate data. Further, for example, model body motion data is extracted from the memory 114 in which model body motion data is accumulated, and is compared with the wearer's body motion reproduction data to detect deviation from normal body motion. The indicated improvement data may be generated. Furthermore, by registering user information such as sex, height, weight, and age in advance, analysis may be performed based on each user information. Then, the analysis unit 117d transmits analysis results such as the stereoscopic image data and the improvement data to the information terminal device 100. FIG.

The analysis unit 117d includes an index calculation unit 117g and a stability calculation unit 117f as modules related to running form evaluation processing. The stability calculation unit 117f is a module that calculates a stability reference value and evaluates the reproducibility of body movements based on the amount of deviation from the stability reference value. The index calculation unit 117g is a module that calculates an index for evaluating running form by referring to index data based on a reference value or a threshold obtained from body motion data. Specifically, the analysis unit 117d analyzes the contact state data for the number of peaks, the height h of each peak, the length (time) B of the skirt of the peak, and the shape of the peak (the sharpness of the peak, the order of height, integral value (area, etc.) per unit time), and when there are multiple peaks, the heights h1 and h2 of each peak, the length of the tail of the peak (time) and the shape of the peak, the distance between the peaks (time difference) is recognized and evaluated based on comparison with the index.

Particularly, in the present embodiment, the index calculator 117g calculates an index for evaluating running form based on the ground contact state data D1 to D6 extracted by the ground contact state detector 117h, as shown in FIG. Further, in the present embodiment, the index calculation unit 117g sets a value selected from the average value, maximum value, minimum value, or arbitrary representative value within a predetermined period as the reference value based on the setting operation by the wearer. It has functionality.

The stability calculation unit 117f analyzes the motion of the wearer using the body movement reproduction data and the reference value, and calculates the calculated change points of the running form in order to evaluate the reproducibility and sustainability of the wearer's running form. is detected. In particular, in the present embodiment, the stability calculation unit 117f analyzes the past running forms of the wearer and others using the body movement reproduction data, and performs a process of generating overall analysis information as an analysis result.

At this time, the stability calculation unit 117f searches for the date selected by the wearer and past travel records with similar feature points, and calculates an average value for some or all of the various exercise information in the run. Final value selection processing at the end of running, determination of whether these values are better (or worse) than the reference value and whether the improvement rate is higher (or lower) than the reference value processing, etc. In addition, the stability calculation unit 117f calculates (or selects) an average value (or a final value) for each date of running for a predetermined item or an item selected by the wearer, and calculates (or selects) an index in chronological order. to generate

The index calculation unit 117g evaluates the running results of the run on the date selected by the wearer, and generates information on evaluation results, methods for improving running style, methods for shortening times, and indices for coaching such as training guidance. The index calculation unit 117g uses various types of exercise information stored in the memory 114 to compare and analyze the results of the wearer's past multiple runs, or to calculate the results of the wearer's past runs in other ways. It is possible to analyze by comparing with the wearer's running result, and include comparative analysis information, which is the information of the analysis result, in the index. Specifically, the index calculation unit 117g generates comparative analysis information similar to the detailed analysis information for each run on a plurality of dates selected by the wearer, or generates comparative analysis information similar to the detailed analysis information for each run on the dates selected by the wearer. Information of comparative analysis similar to detailed analysis information is generated for each of the past runs of other wearers.

In this embodiment, the analysis unit 117d has a synchronization processing function, and this synchronization processing function matches the time axis for displaying the body movement data with the time axis for displaying the image captured by the camera. A calibration process may be performed in order to perform the synchronization process. Specifically, the synchronization processing function according to the present embodiment extracts a predetermined characteristic action (calibration action) by the wearer 1 that causes the body motion sensor 40 to react from video or audio, and Synchronous processing is performed by extracting a characteristic reaction by a characteristic action and matching the timing of the extracted characteristic action with the timing of the characteristic reaction. This characteristic action includes, for example, a vibration that occurs when an acceleration force suddenly acts from a stationary state, such as when starting to run, and a short-term action such as hitting or shaking the body motion sensor 40 itself a predetermined number or times. Actions such as applying vibration to the body motion sensor 40 a predetermined number of times during a period of time can be mentioned.

As a process for extracting such a characteristic action, for example, by detecting that a figure of a predetermined shape and color is displaced with a constant amplitude and rhythm within a predetermined length of time by image recognition processing, the wearer 1 can say " By recognizing characteristic actions such as "start running" and "jumping on the spot", the camera reads out the shooting time of the frame in the video that has the maximum displacement during that time, and calibrates that time information. detected as a motion signal. Further, when audio data is included in the moving image data, the sound of the wearer 1 tapping may be extracted, and the extracted time information may be detected as the calibration signal.

Furthermore, when the recorded data is of a size that spans a long period of time, it is possible to assist the user's operation and extract the above-described characteristic actions by narrowing down the time span specified by the user's operation. On the other hand, the characteristic reaction of the body motion sensor 40 is detected by scanning the detection values of various sensors provided in the body motion sensor 40, such as an acceleration sensor, and detecting a response with a certain amplitude or more within a predetermined time period. is repeated a predetermined number of times. It should be noted that the detection of the characteristic behavior and the characteristic reaction may be performed by performing one of them first, and referring to the time stamp of the detected one, narrowing down the scanning range of the other to execute the detection process. good.

Then, in order to match the timing of the extracted characteristic behavior and the timing of the characteristic reaction, the time stamp (time information) of the time when the characteristic behavior was performed and the time when the characteristic reaction was detected Synchronization processing is performed by aligning the reproduction start times of the two so that the time stamps of the two are matched. At that time, if there is a difference in the interval between repeated actions or reactions, the playback time of the video or the length of the timeline of the body motion sensor is extended, and the playback start time of the video data and the body motion reproduction data Synchronize by matching the end time with

The display information generation unit 117e is a module that generates display information displayed on the output interface 111, and generates display information for displaying or outputting the body movement reproduction data analyzed by the analysis unit 117d in association with a moving image. In the present embodiment, this display information is displayed on the screen as moving images captured by the built-in camera 115 or an external camera, etc., and can be compared with the body movement reproduction data analyzed by the analysis unit 117d and the timeline. Display in sync. This display information includes display data as well as acoustic signals and other output control signals.

In addition, the display screen includes a GUI (Graphical User Interface) for touch operation, and an operation on the touch panel on which this GUI is displayed is input to the input interface 112 and can switch the display by the display information generation unit 117e. can. For example, it is possible to display a moving image of the wearer 1 captured by the built-in camera 115 or an external camera on the screen, or to individually display each movement parameter included in the body movement reproduction data on the timeline, and switch the display mode. As a result, it is possible to display a moving image of the wearer 1 photographed from the front taken by the built-in camera or the like of the information terminal device 100, or to display each motion parameter included in the body motion reproduction data superimposed on the timeline. . As another method for switching the display mode, for example, various methods can be adopted such as superimposing a timeline on a moving image and displaying it on a full screen.

The memory 114 is a storage device for recording various data, and includes identification information for identifying each information terminal device 100, attachment site information of each body motion sensor 40, and relative position of the body motion sensor 40 attached to each site. Relationships, user information described above, model body movement data, and the like are accumulated. The memory 114 functions as a storage unit that stores index data, and the index data is the stability period calculated by the stability calculation unit 117f, the deviation amount after the stability period, and the index for evaluating the stabilization ability. This is table data that holds correlations.

(Running form evaluation method)
By operating the running form evaluation system having the above configuration, the running form evaluation method according to the present embodiment can be implemented. FIG. 4 shows the recording operation of the running form evaluation system, and FIG. 5 shows the processing during operation analysis. Note that the processing procedure described below is merely an example, and each processing may be changed as much as possible. Further, in the processing procedures described below, steps can be omitted, replaced, or added as appropriate according to the embodiment.
(1) Recording Operation First, the wearer 1 wears a pair of

body motion sensors

40a and 40b on both left and right thighs. Then, the running form evaluation application, which is the program of the present invention, is activated on the information terminal device 100 side, and a measurement start operation is input to the application to acquire detection results from the body motion sensors 40 . At this time, if necessary, the external camera is operated to start photographing (S201). Upon receiving this measurement start operation, the control unit 117 of the information terminal device 100 performs connection processing with each body movement sensor 40 (S101). After the connection process, each body motion sensor 40 starts detecting the motion of the wearer 1 (S102). Specifically, a body motion sensor 40 attached to the wearer's thigh detects three-dimensional displacement, rotation, or acceleration of each part.

Then, each acquired detection result is transmitted to the wireless interface 113 of the information terminal device 100 by weak radio waves via the wireless communication unit of each body motion sensor 40 (S103). When the wireless interface 113 of the information terminal device 100 starts acquiring each detection result (S202), recording of the detection results by the

body motion sensors

40a and 40b as body motion data is started in the memory 114, which is a body motion recording unit. Then, the detection signals continuously transmitted from each body movement sensor 40 are sequentially recorded (S203). At this time, the wearer may perform a calibration operation before starting the competition if necessary. Specifically, for example, in front of the camera, the wearer 1 jumps, hits or shakes the body of the wearer 1 or the body motion sensor 40 itself by a predetermined number or a number of times, etc., to the body motion sensor 40 in a short time. Perform an operation that applies vibration a predetermined number of times.

Next, running is started, and during running, the detection values of the body motion sensor 40 are continuously acquired, and the recording process is continuously executed. recorded as body movement data. At this time, moving image data captured by a built-in camera built in the information terminal device 100 or an external camera connected to the outside may be acquired, and the moving image data is accumulated in the memory 114 or processed by the control unit 117. served to During this time, the data detected by the body motion sensor 40 is analyzed in real time, and the ground state data D1 to D6 are extracted (S204) and displayed on the display section of the information terminal device 100. FIG. As one of these analyses, a running form evaluation process is performed based on the recorded body movement data (S205).

After that, for example, by detecting a stationary state for a certain period of time or more, or detecting the end operation of the wearer, the end of the competition is detected and the measurement is ended ("Y" in S206). The recording process is stopped (S207). After that, the communication with each sensor is disconnected (S104).

(2) Running Form Evaluation Processing The running form evaluation processing in step S205 will be described in detail. As shown in FIG. 5, body movement data is collected by the body movement data acquisition unit 117a, and video data is collected as necessary (S301). At this time, the body motion data, which are the detected values obtained from the sensors, are input to the display information generation unit 117e as primary data, can be directly output processed (S306), and are also input to the body motion calculation unit 117b. Necessary information is extracted, analyzed and corrected by the analysis unit 117d, and input to the display information generation unit 117d as secondary data, which is body motion reproduction data, and output processing is performed.
With respect to the primary data input to the body motion calculator 117b, the non-grounded state of the wearer's leg is detected by the grounded state detector 117h, and each detection result related to the detected non-grounded state is used to determine the state of the grounded side. (S302). The grounding state detection unit 117h in the present embodiment detects, for example, detection values (body motion data) of an acceleration sensor and an angular velocity sensor that constitute body motion sensors, and detects values indicating characteristic behaviors, changes thereof, and time (time). ), and based on the timing, the contact time and impact time, their rate of change, and the temporal range of the contact state or non-contact state depending on the periodicity are specified. At the same time, a flag is set in the data for the length of time included in the range specified as the non-grounded state. Data in which this flag is set is extracted as ground state data D1 to D6 indicating the state of the ground side.

Then, the body motion calculation unit 117b and the analysis unit 117d perform evaluation and analysis based on the detection results of the body motion sensor 40 and the ground contact state data D1 to D6 stored in the memory 114 and the primary data. and body motion reproduction data, which is secondary data calculated based on the relative positional relationship of each body motion sensor 40 . First, a running motion analysis step is performed based on the body motion data recorded in the memory 114 (S303). More specifically, the body motion calculation unit 117b first uses the body motion data, which are the detection results of the

body motion sensors

40a and 40b, to detect feature points in the wearer's running.
Characteristic points in the wearer's running are detection values that indicate characteristic behavior detected by the body motion sensor, changes in the detected values, and portions of data corresponding to time (time). changes in acceleration based on the characteristic body movements of the wearer. In addition, the body motion calculation unit 117b performs a process of calculating each value of contact time and impact time based on the timing of the detected feature point. Specifically, based on the timing at which the characteristic points are detected from the calculated data including the characteristic points, the length of time that the value continues and the rate of change within a certain unit time are used to calculate the contact duration and impact duration. Calculate each value.

In this running motion analysis step, the reference value set by the user's operation can be used as the stability reference value, and the reproducibility may be evaluated by comparing the amount of deviation from the stability reference value. When the motion is repeated multiple times, the average value of the parameters related to the repeated motion over a predetermined period (or a predetermined number of times) may be calculated as the stability reference value.
Regarding the setting of this reference value, the reference value setting unit 117c sets a stable reference value for evaluating the reproducibility of body movement based on the body movement reproduction data accumulated in the memory 114 according to the operation of the wearer 1. set. Specifically, based on the setting operation by the wearer 1, a value selected from an average value, maximum value, minimum value, or arbitrary representative value within a predetermined period is set as the reference value. For example, in this reference value setting operation, for example, the same action is repeated several times at predetermined time intervals, and in addition to the average value, minimum value, maximum value, and the value that the wearer 1 thought was the best, can be set as the ideal value, or by inputting any numerical value such as the ideal value for advanced users or professionals.

Then, based on the result of the running motion analysis, the index data is referred to, and an index calculating unit 117g calculates an index for evaluating the running form and stabilizing ability (S304). In calculating this index, as shown in FIGS. 6 to 8, predetermined extracted grounding state data D1 to D6 are analyzed. For example, as shown in FIG. height h, the length (time) B of the tail of the peak, and the shape of the peak (the sharpness of the peak, the order of high and low, the integrated value (area, etc.) per unit time, etc.). As shown in FIG. 4B, when there are a plurality of peaks, the heights h1 and h2 of the respective peaks, the lengths (time) of the skirts of the peaks B1 and B2, the shape of the peaks, the distance between the peaks (time difference ).

Then, based on the information about this recognized peak, an index for evaluating running form is calculated. For example, as shown in FIG. 7A, when the number of peaks is one and the length B of the tail is narrow, the evaluation is made such that the timing is early. When the number of peaks is two and the first peak is higher than the second peak, as shown in FIG. Even if the number of peaks is two, when the height of the first and second peaks is about the same and the second peak has a wide and gentle shape, it is evaluated as "ordinary timing".

Furthermore, as shown in FIG. 8(d), even if there are two peaks, even if the first preceding peak is extremely low, the timing is judged to be "ordinary", but FIG. 8(a). As shown in , when the shape of the peak is trapezoidal, it is determined that "the timing is a little late". In FIG. 8(b), when the number of peaks is 1 and the width of the foot is general, but the right side is gently sloping, it is judged that the timing is early. Also when the length B of the skirt is narrow and the shape forms an acute angle, it is also judged that the timing is early. Also, as shown in FIG. 8(d), when the number of peaks is two, the second is trapezoidal, the foot is wide and the right shoulder is gentle, the timing is normal, and as shown in FIG. 8(e). Secondly, even if there are two peaks, if the preceding peak is extremely low and both peaks are wide and gentle, it is determined that the timing is slow and normal.

In addition, in this index calculation step, whether the amount of deviation from the stability reference value is within a predetermined threshold value is monitored as needed, and a stable period during which these body movement parameters maintain a stable state is calculated. good too. For example, the average values are sequentially calculated from the start of the competition, and the average value while the average value is within a predetermined amount of change is used as the stable reference value. The stable reference value is updated at any time, and the deviation amount of the current value from the stable reference value is monitored at any time. After that, evaluation processing and predetermined diagnostic processing are performed based on the calculated index (S305), and the result of the diagnostic processing and the index are displayed together with the animation and body movement reproduction data synchronized so as to be able to be compared with them, together with the information terminal device. 100 display or sound by speakers or the like (S306).

(running form evaluation program)
The running form evaluation system and running form evaluation method according to the present embodiment described above execute the running form evaluation program of the present invention written in a predetermined language, like the running form evaluation application described above, on a computer. It can be realized by That is, the program of the present invention is installed in an IC chip or memory device of a mobile terminal device, a smart phone, a wearable terminal, a mobile PC or other information processing terminal, a general-purpose computer such as a personal computer or a server computer, and executed on a CPU. Thus, a system having the functions described above can be constructed and a running form evaluation method can be implemented.

(action/effect)
According to this embodiment, the positions of the sensors are both thighs, and measurement is performed with the leg opposite to the leg on the ground. In other words, the sensor on the left leg measures the ground reaction force when the right foot is in contact with the ground. Thus, according to the present embodiment, by using the data of the opposite leg, the impact due to landing can be avoided. From the grounded leg, sensors are attached to the tip of the ankle, knee, grounded hip joint, and reverse hip joint. The movement of the left hip joint) can be measured smoothly and accurately when the right ground is touched. For example, since the rotation of the waist on the horizontal plane does not greatly affect the acceleration in the vertical direction of the thigh of the sensor attached to the thigh, the desired acceleration can be detected because the acceleration is almost vertical.

As a result, according to the present embodiment, it is possible to minimize the impact noise and measurement error caused by landing from both the left and right legs when the subject is running without requiring the construction of a costly large-scale device or an expensive system. Therefore, it is possible to realize more accurate motion analysis of running competitions.

In addition, in the running form evaluation program according to the present embodiment, for example, it is possible to distribute it through a communication line, and by recording it on a computer-readable recording medium, a package that operates on a stand-alone computer Can be transferred as an application. Specifically, the recording medium can be a magnetic recording medium such as a flexible disk or a cassette tape, an optical disk such as a CD-ROM or a DVD-ROM, or various recording media such as a RAM card. Then, according to the computer-readable recording medium recording this program, it is possible to easily implement the above-described system and method using a general-purpose computer or a dedicated computer, and save, transport, and Easy to install.

It should be noted that the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying constituent elements without departing from the gist of the present invention at the implementation stage. Also, various inventions can be formed by appropriate combinations of the plurality of constituent elements disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments.

D1 to D6 Grounding state data 1 Wearer 40 Body motion sensor 100 Information terminal device 111 Output interface 112 Input interface 113 Wireless interface 114 Memory 115 Built-in camera 117 Control unit 117a Body motion data acquisition Part 117b... Body motion calculation part 117c... Reference value setting part 117d... Analysis part 117e... Display information generation part 117f... Stability calculation part 117g... Index calculation part 117h... Ground state detection part

Claims

A system that detects body movements of a wearer and evaluates running form based on changes in body movements,
a plurality of pairs of body motion sensors mounted on each of the thighs of both legs of the wearer and capable of detecting three-dimensional displacement or rotation of each part;
The ungrounded state of the legs of the wearer is detected based on the detection results of the pair of body motion sensors, and each detection result related to the detected ungrounded state is extracted as grounded state data indicating the state of the grounded side. a ground state detector;
an index calculation unit that calculates an index for evaluating the running form based on the contact state data extracted by the contact state detection unit;
A running form evaluation system, comprising: an output device for displaying or outputting the index calculated by the index calculation unit.
a storage unit that stores index data that holds a correlation between an amount of deviation from a reference value for evaluating the running form and an index for evaluating the running form;
2. The running according to claim 1, wherein the index calculation unit refers to the index data based on the ground contact state data, and calculates an index for evaluating running form according to an amount of deviation from a reference value. form evaluation system.
further comprising a stability calculation unit that calculates a stable period during which the stable state of the running form is continued based on the amount of deviation from the reference value;
The storage unit stores, as the index data, a correlation between the stable period calculated by the stability calculating unit, the deviation amount after the stable period, and an index for evaluating the stabilizing ability,
3. The running form evaluation system according to claim 2, wherein the index calculation unit calculates an index for evaluating the running form by referring to the index data according to the calculation result of the stability calculation unit. .
The index calculation unit sets a value selected from an average value, a maximum value, a minimum value, or any representative value within the predetermined period as the reference value based on a setting operation by the wearer. The running form evaluation system according to claim 2 or 3.
A program for detecting body movements of a wearer and evaluating running form based on changes in body movements, comprising:
a plurality of pairs of body motion sensors mounted on each of the thighs of both legs of the wearer and capable of detecting three-dimensional displacement or rotation of each part;
The ungrounded state of the legs of the wearer is detected based on the detection results of the pair of body motion sensors, and each detection result related to the detected ungrounded state is extracted as grounded state data indicating the state of the grounded side. a ground state detector;
an index calculation unit that calculates an index for evaluating the running form based on the contact state data extracted by the contact state detection unit;
A running form evaluation program characterized by functioning as an output device for displaying or outputting the index calculated by the index calculation unit.
the computer further comprising:
Functioning as a storage unit that stores index data that holds correlation between an amount of deviation from a reference value for evaluating the running form and an index for evaluating the running form,
6. The running according to claim 5, wherein the index calculation unit refers to the index data based on the ground contact state data, and calculates an index for evaluating running form according to an amount of deviation from a reference value. form evaluation program.
the computer further comprising:
Functioning as a stability calculation unit for calculating a stable period in which a stable state of running form is continued based on the amount of deviation from the reference value,
The storage unit stores, as the index data, a correlation between the stable period calculated by the stability calculating unit, the deviation amount after the stable period, and an index for evaluating the stabilizing ability,
7. The running form evaluation program according to claim 6, wherein the index calculation unit calculates an index for evaluating the running form by referring to the index data according to the calculation result of the stability calculation unit. .
The index calculation unit sets a value selected from an average value, a maximum value, a minimum value, or any representative value within the predetermined period as the reference value based on a setting operation by the wearer. A running form evaluation program according to claim 6 or 7.
A method for detecting body movement of a wearer and evaluating running form based on changes in body movement, comprising:
A plurality of pairs of body motion sensors capable of detecting three-dimensional displacement or rotation of each part are attached to each of the thighs of both legs of the wearer, and the grounding state detection unit is detected by the pair of body motion sensors. a grounded state detection step of detecting a non-grounded state of the leg of the wearer based on the detection result, and extracting each detection result related to the detected non-grounded state as grounded state data indicating a state of the grounded side;
an index calculation step in which an index calculation unit calculates an index for evaluating the running form based on the ground contact state data extracted in the ground contact state detection step;
and an output step in which an output device displays or outputs the index calculated by the index calculation unit.
further comprising a storage step in which the storage unit stores index data holding a correlation between an amount of deviation from a reference value for evaluating the running form and an index for evaluating the running form;
10. The running according to claim 9, wherein in said index calculation step, said index data is referred to based on said ground contact state data, and an index for evaluating running form is calculated according to the amount of deviation from a reference value. Form evaluation method.
further comprising a stability calculation step in which a stability calculation unit calculates a stable period during which the stable state of the running form is continued based on the amount of deviation from the reference value;
In the storing step, the correlation between the stable period calculated by the stability calculating unit, the deviation amount after the stable period, and an index for evaluating the stabilizing ability is stored as the index data,
11. The running form evaluation method according to claim 10, wherein in the index calculation step, the index data is referred to according to the calculation result of the stability calculation unit to calculate the index for evaluating the running form. .
In the index calculation step, a value selected from an average value, a maximum value, a minimum value, or any representative value within the predetermined period is set as the reference value based on a setting operation by the wearer. The running form evaluation method according to claim 9 or 10.