Running analysis system and running analysis method

Abstract

To provide technique for facilitating the running analysis based on running-related information.SOLUTION: In a running analysis system 100, an information acquisition unit 70 acquires position information and operation information of a user, who is a runner, measured for each elapsed point during running by a predetermined measurement device 20. A feedback acquisition unit 86 acquires feedback information input by the user regarding the running. An analysis processing unit 87 extracts an essential point on the basis of a predetermined extraction condition from the feedback information. A mode determination unit 90 preferentially extracts, as an output target, from classified measurement values, an operation analysis index matching the essential point in a measurement section in which a change corresponding to a predetermined change condition is detected, and the measurement value thereof, thereby determining an output mode of the measurement value.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present disclosure relates to a drive analysis system, and more particularly to a system for analyzing the results of a drive. [Background technology]

[0002] In recent years, the number of people running has increased due to people's growing health consciousness. In particular, the widespread use of smartphones and watches with built-in GPS (Global Positioning System) modules has made it easy for anyone to record a running log (hereinafter referred to as a "running log"). The use of such running logs motivates people to continue running and make it a habit, which has boosted the popularity of running.

[0003] As the number of runners increases, the number of participants in marathon races and the number of races are also increasing. By recording a running log during a race, runners can review their running results after the race is over. As a technology for displaying running data in this way, a running data display method and an exercise support device are known (for example, see Patent Documents 1 and 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7031234 [Patent Document 2] Patent No. 5984002 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the prior art, even if various information on the amount of running and running form, such as the running time and distance in a race, was available, it was not necessarily easy to find out which information to pay particular attention to in order to improve one's own level. Therefore, for users who could not fully utilize various information even if they obtained it, it was difficult to be motivated to look back on the race results based on the information.

[0006] The present disclosure has been made in consideration of these circumstances, and has a purpose to provide a technology that facilitates driving analysis based on information related to driving. [Means for solving the problem]

[0007] In order to solve the above problems, a running analysis system according to one aspect of the present disclosure comprises an information acquisition unit that acquires position information and movement information of a user, who is a runner, measured at each progress point in the run by a specified measuring device; a measurement acquisition unit that acquires measurement values ​​of multiple types of motion analysis indicators that indicate the running movement state of the user for each measurement section of a specified unit based on the chronologically continuing position information and movement information; a classification processing unit that classifies the measurement values ​​of the multiple measurement sections for each type of motion analysis indicator using a specified classification method for dividing the characteristics of the measurement values; a feedback acquisition unit that acquires feedback information input by the user regarding the run; an analysis processing unit that extracts key points from the feedback information based on specified extraction conditions; a mode determination unit that determines the output mode of the measurement values ​​by preferentially extracting, from the classified measurement values, motion analysis indicators that match the key points and their measurement values ​​in measurement sections where changes that meet specified change conditions were detected as output targets; and an output unit that can output at least information regarding measurement values ​​that match the analysis results of the feedback information based on the determined output mode.

[0008] Another aspect of the present disclosure is a running analysis method. This method includes the steps of acquiring position information and motion information of a user who is a runner, measured at each progress point in the run by a predetermined measuring device, acquiring measurement values ​​of multiple types of motion analysis indicators that indicate the running motion state of the user for each measurement section of a predetermined unit based on the position information and motion information that continue in time series, classifying the measurement values ​​of the multiple measurement sections for each type of motion analysis indicator using a predetermined classification method for classifying the characteristics, acquiring feedback information input by the user regarding the run, extracting key points from the feedback information based on predetermined extraction conditions, detecting changes that correspond to predetermined change conditions from the classified measurement values, matching the measurement section in which the change was detected with the key points among the classified measurement values, preferentially extracting the motion analysis indicators and their measurement values ​​that match the key points in the measurement section in which the change was detected as output targets, determining the output mode of the measurement values ​​based on the extraction result of the output target, and outputting at least information regarding the classified measurement values ​​based on the determined output mode.

[0009] In addition, any combination of the above components, or mutual substitution of the components or expressions of the present disclosure among methods, devices, programs, temporary or non-temporary recording media having programs recorded thereon, systems, etc., are also valid aspects of the present disclosure. Effect of the Invention

[0010] According to the present disclosure, it is possible to provide a technique for facilitating driving analysis based on information related to driving. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating a configuration of a driving analysis system. [Diagram 2] FIG. 2 is a functional block diagram showing each component of the driving analysis system. [Diagram 3] FIG. 2 is a functional block diagram showing each function of a driving analysis server. [Figure 4]4 is a flowchart showing a process in the driving analysis system. [Diagram 5] FIG. 13 is a diagram illustrating an example of a feedback input screen. [Figure 6] 10 is a diagram illustrating an example of a judgment table created by an analysis processing unit. FIG. [Figure 7] 10 is a diagram illustrating a change list table created by a change detection unit. FIG. [Figure 8] FIG. 11 is a diagram showing an example of classifying index measurement values ​​using a first classification method. [Figure 9] FIG. 13 is a diagram showing a preparation process when classifying index measurement values ​​using the second classification method. [Figure 10] FIG. 13 is a diagram illustrating a process of classifying index measurement values ​​using a second classification method. [Figure 11] FIG. 13 is a bar graph showing the proportion of classifications for each motion analysis index. [Figure 12] FIG. 12 is a diagram showing a state in which the index measurement values ​​divided in the horizontal bar graph of FIG. 11 are evaluated using yet another criterion. [Figure 13] 13 is a diagram showing an example of displaying an output target narrowed down to index measurement values ​​detected as a value of interest. FIG. [Figure 14] FIG. 13 is a diagram showing an example of a motion analysis index having the longest period in which a significantly bad trend continues. [Figure 15] FIG. 13 is a diagram showing an example of a screen displaying race results. [Figure 16] FIG. 13 is a diagram showing an example of a screen displaying a race evaluation. [Figure 17] FIG. 13 is a diagram showing an example of a screen displaying a detailed explanation regarding evaluation points. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] In this embodiment, the changes in each motion analysis index are detected from running data in a marathon race or the like, and compared with a reference value to visualize changes in physical condition, providing reference information for the user to review the race. In addition, points that the user particularly noticed or was concerned about are extracted as key points from comments entered by the user, and changes in the motion analysis indexes corresponding to the key points are detected as important changes and visualized, providing reference information for the user to review the race.

[0013] Hereinafter, the present disclosure will be described based on preferred embodiments with reference to the drawings. In the embodiments and modified examples, the same or equivalent components are given the same reference numerals, and redundant descriptions are omitted as appropriate.

[0014] FIG. 1 shows the configuration of a running analysis system 100. The running analysis system 100 includes a wristwatch type device 12, a waist-worn device 14, an information terminal type device 16, and a running analysis server 60, which can be worn by a user 10 who is a runner in a running exercise. The wristwatch type device 12, the waist-worn device 14, and the information terminal type device 16 are collectively referred to as a measurement device 20. The wristwatch type device 12 is a sports watch or a smart watch that can measure position information, motion information, and the like. The waist-worn device 14 is a motion sensor that can be worn near the waist of the user 10 and measure position information and motion information. The information terminal type device 16 is a portable information terminal such as a smartphone that can measure position information and motion information while the user 10 holds it in a pocket or the like. The user 10 wears one or more measurement devices 20 and runs in a race or the like to obtain position information and motion information. When multiple measuring devices 20 are worn, different devices may be used depending on the information to be acquired, such as acquiring position information with a wristwatch type device 12 and acquiring motion information with a waist-worn device 14.

[0015] The measuring device 20 is not limited to devices such as the wristwatch device 12, the waist-worn device 14, and the information terminal device 16, but may be a device worn on or in the runner's shoes. Alternatively, it may be a belt-type device that can be wrapped around the runner's chest, wrist, waist, or arm to obtain position information and motion information. In addition, various wearable devices such as smart glasses may be used as the measuring device 20. In addition, cameras may be installed near each progress point on the running course to photograph the runner, and skeletal information such as the runner's joint positions and motion information such as pitch and stride that can be calculated from such information may be obtained by image recognition. Instead of installing cameras near each progress point, the runner may be photographed by a drone.

[0016] The driving analysis system 100 may be realized with various hardware and software configurations. For example, the driving analysis system 100 may be configured with only the information terminal 50, or may be configured with a combination of the information terminal 50 and the measurement device 20, or may be configured with a combination of the information terminal 50 and the driving analysis server 60. Alternatively, the driving analysis system 100 may be configured with a combination of the information terminal 50, the measurement device 20, and the driving analysis server 60, or may be configured with a combination of the measurement device 20 and the driving analysis server 60, or may be configured with only the driving analysis server 60.

[0017] For example, on the premise that various general-purpose devices are used as the measuring device 20 to detect driving conditions and record them as a driving log, the driving analysis system 100 may be configured with a combination of the information terminal 50 and the driving analysis server 60, or with just the driving analysis server 60, or may be realized as a stand-alone device that includes all of the software configuration included in the information terminal 50 and driving analysis server 60 in this figure. Therefore, regardless of the aspect of its hardware configuration, it is sufficient for the driving analysis system 100 to include at least the software configuration of the information terminal 50 and driving analysis server 60 in this figure.

[0018] The user 10 runs while wearing at least one of the wristwatch type device 12, the waist-worn device 14, and the information terminal type device 16, or all of them, as the measurement device 20. The measurement device 20 synchronizes information with the running analysis server 60 through communication. However, since the communication means of the wristwatch type device 12 and the waist-worn device 14 among the measurement devices 20 is short-range wireless communication, they do not directly communicate with the running analysis server 60, but synchronize information with the information terminal type device 16 (which also functions as an "information terminal 50" described later in detail), and the information terminal 50 synchronizes information with the running analysis server 60. In this way, the wristwatch type device 12 and the waist-worn device 14 transmit information to the running analysis server 60 through synchronization with the information terminal 50, so it is assumed that the information terminal 50 is held. However, it is not necessarily required to wear the information terminal 50 during the running, and it is sufficient to synchronize with the information terminal 50 after the exercise is performed. As a variant, the waist-worn device 14 may first synchronize information with the wristwatch device 12 via short-range wireless communication, and the wristwatch device 12 may then further synchronize information with the information terminal device 16 (information terminal 50) via short-range wireless communication.

[0019] The user 10 wears the measuring device 20 and runs mainly in running races such as marathons. However, the measurement is not limited to during the race, and may be performed during a long-distance practice run simulating a race, or during a shorter run. When the user 10 starts running, the measuring device 20 operates a button or the like to start measurement and running log recording. During the running exercise, the measuring device 20 measures the elapsed time from the start of recording as the running time using a timer, and records position information for each date and time at a predetermined time interval. The measuring device 20 measures motion information such as pitch (the number of steps per unit time, also called cadence), rotational and translational movement of the pelvis, and impact value using a built-in motion sensor. The measuring device 20 measures the heart rate of the user 10 using a built-in optical heart rate meter.

[0020] After the run and the running log recording are completed, the measuring device 20 transmits information such as the running time, position information, movement information, and heart rate as a running log to the running analysis server 60. The measuring device 20 may calculate information such as the running time, running distance, running speed, pitch, stride, altitude, and incline based on the time information and position information, and include this calculated information in the running log and transmit it to the running analysis server 60. The measuring device 20 may also obtain meteorological information such as temperature, humidity, weather, wind direction, and wind speed corresponding to the time during the run from a specified server, and include this meteorological information in the running log and transmit it to the running analysis server 60.

[0021] The driving analysis server 60 is a server computer connected to the Internet and transmits and receives data to and from the information terminals 50 of multiple users 10. The driving analysis server 60 acquires information such as time information, position information, motion information, and heart rate as driving log data of the user 10 received from the information terminal 50 along with the identification information and attribute information of the user 10, and accumulates the information together with the measured values ​​and evaluation values ​​of various types of driving analysis indexes calculated from each piece of information. The driving analysis server 60 may acquire meteorological information such as temperature, humidity, weather, wind direction, and wind speed corresponding to the time during driving from a specified server, and accumulate the meteorological information by including it in the driving log data. In response to a request from the information terminal 50, the driving analysis server 60 transmits the accumulated driving log data, measured values, evaluation values, and the like to the information terminal 50.

[0022] FIG. 2 is a functional block diagram showing each configuration of the driving analysis system 100. FIG. 2 illustrates functional blocks of the measuring device 20 and the information terminal 50, which are realized by the cooperation of various hardware configurations and software configurations. Therefore, it is understood by those skilled in the art that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof. The measuring device 20 is configured by a combination of hardware such as a microprocessor, a display device, a memory, a communication module, a positioning module, a motion sensor, and an optical heart rate monitor. The information terminal 50 is configured by a combination of hardware such as a microprocessor, a touch panel, a memory, a communication module, a positioning module, and a motion sensor. The functions of the measuring device 20 and the information terminal 50 will be described below.

[0023] The measuring device 20 is, for example, a waist-worn device 14. The measuring device 20 has a communication unit 21, a time measurement unit 22, a position measurement unit 24, and a motion detection unit 26. The time measurement unit 22 measures the running start time, i.e., the running time from the measurement start time, by counting a timer. The position measurement unit 24 measures the current position based on position information received by the GPS module from a satellite positioning system. The motion detection unit 26 detects the pitch of the user 10, the rotational and translational motion of the pelvis, impact values, etc., by a motion sensor.

[0024] The measurement device 20 may be a wristwatch type device 12 or an information terminal type device 16. When the wristwatch type device 12 is used as the measurement device 20, the motion detection unit 26 of the wristwatch type device 12 detects the pitch of the user 10 by a motion sensor and detects the heart rate by an optical heart rate meter. When the information terminal type device 16 is used as the measurement device 20, the motion detection unit 26 of the information terminal type device 16 detects the pitch of the user 10 by a motion sensor. The information terminal 50 may also serve as the information terminal type device 16 as the measurement device 20. In this case, for example, a single mobile terminal such as a smartphone may have all the functions of both the measurement device 20 and the information terminal 50. When the information terminal 50 is not used as the measurement device 20, the information terminal 50 is not limited to a smartphone, but may be a tablet terminal or a personal computer owned by the user 10.

[0025] The information terminal 50 has an information acquisition unit 30, a measurement acquisition unit 40, an input / output unit 51, and a communication unit 52. The information acquisition unit 30 receives the position information and the motion information of the user 10, who is a runner in a race, measured at each passing point in the run by the measuring device 20 worn by the user 10, via the communication unit 52. Here, the "passing point" refers to a passing point in time or distance in the run of a race, etc., and the position information and the motion information are recorded in the measuring device 20 at each passing point in time or distance. The position information includes information such as the date and time of measurement, the position coordinates, and the altitude. The motion information includes information indicating the pitch, the rotation and translation of the pelvis, or the impact value. The information acquisition unit 30 may synchronize information with the measuring device 20 during the run and acquire the position information and the motion information as information on the running state during the run by the measuring device 20, or may acquire the position information and the motion information of the entire course from the measuring device 20 collectively after the run ends.

[0026] The measurement acquisition unit 40 acquires measurement values ​​of multiple types of motion analysis indexes indicating the user's running motion state for each measurement section of a predetermined unit (hereinafter, the measurement values ​​of the motion analysis indexes indicating the running motion state are referred to as "indicator measurement values") based on the information acquired by the information acquisition unit 30, and based on the position information and motion information continuing in chronological order. The "measurement section of a predetermined unit" refers to a measurement interval in units of a predetermined elapsed time such as every second or every minute, or a measurement interval in units of a predetermined elapsed distance such as every 100 m or every km.

[0027] The measurement acquisition unit 40 calculates index measurement values ​​for multiple types of motion analysis indexes, such as lap pace, lap time, running time, running speed, pitch, stride, stride height ratio, trunk backward inclination, vertical movement, vertical movement height ratio, waist sinking, pelvis left and right inclination, pelvis lift amount, pelvis rotation amount, pelvis rotation timing, left and right impact, kicking off time, ground contact time, ground contact time rate, landing impact, kicking off acceleration, deceleration amount, stiffness, and stiffness to body weight ratio. The motion analysis indexes may include heart rate and its measurement value. The measurement acquisition unit 40 records the index measurement values ​​that continue in chronological order throughout the entire process from the start to the end of the run as time series data.

[0028] The input / output unit 51 transmits the position information and operation information acquired by the measuring device 20 and the index measurement values ​​acquired by the measurement acquisition unit 40 as a driving log together with attribute information of the user 10 to the driving analysis server 60 via the communication unit 52. The input / output unit 51 displays the position information and operation information acquired by the measuring device 20 and the index measurement values ​​acquired by the measurement acquisition unit 40 on a screen, and also displays information received from the driving analysis server 60 on the screen. The input / output unit 51 accepts operational input by the user 10. In terms of hardware, the input / output unit 51 may be configured as a touch panel.

[0029] In this embodiment, an example has been described in which the measurement value acquisition unit 40 acquires a plurality of types of index measurement values, but by having the functions of the measurement value acquisition unit 40 in the driving analysis server 60, it is possible not to calculate the index measurement values ​​in the information terminal 50. In the following description of the driving analysis server 60, an example will be described in which the driving analysis server 60 has a measurement value acquisition unit 80 that corresponds to the functions of the measurement value acquisition unit 40.

[0030] FIG. 3 is a functional block diagram showing each function of the driving analysis server 60. The driving analysis server 60 is illustrated with functional blocks realized by the cooperation of various hardware configurations and software configurations. Therefore, it is understood by those skilled in the art that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof. The driving analysis server 60 is composed of a combination of hardware such as a microprocessor, a memory, a display, and a communication module. The driving analysis server 60 includes a communication unit 62, a log acquisition unit 64, an information acquisition unit 70, a measurement value acquisition unit 80, a classification processing unit 85, a mode determination unit 90, a data storage unit 66, a feedback acquisition unit 86, an analysis processing unit 87, and an output unit 99.

[0031] The log acquisition unit 64 acquires the running log via the communication unit 62 and stores it in the data storage unit 66. The information acquisition unit 70 has a function equivalent to the information acquisition unit 30 of the measuring device 20. In other words, the information acquisition unit 70 acquires, from the running log stored in the data storage unit 66, position information and motion information of the user 10, who is a runner in a race, measured at each progress point during the run by the measuring device 20 worn by the user 10.

[0032] The measurement value acquiring unit 80 is a function equivalent to the measurement value acquiring unit 40 of the information terminal 50. That is, the measurement value acquiring unit 80 acquires index measurement values ​​of a plurality of types of motion analysis indexes that indicate the running motion state of the user for each measurement section of a predetermined unit based on the information acquired by the information acquiring unit 70, that is, on the basis of the position information and the motion information that continue in time series. The measurement value acquiring unit 80 stores the index measurement values ​​that continue in time series over the entire journey from the start to the end of the run in the data storage unit 66 as time series data.

[0033] Depending on the course of a race, etc., the course may be hilly, and external environmental factors such as the season, weather, temperature, humidity, time of day, and number of participants may have a significant effect on the index measurement values. Therefore, the measurement acquisition unit 80 may acquire information on the course conditions and environmental conditions, and normalize the data based on such information to a situation that is not influenced by these course conditions and environmental conditions. In this case, the data to be compared, which will be described later, may also be similarly normalized to achieve a comparison under equal conditions and improve the accuracy of the analysis.

[0034] The classification processing unit 85 classifies the index measurement values ​​of the multiple measurement sections by type of motion analysis index using a predetermined classification method for classifying the index measurement values ​​by characteristics. There are two classification methods. The first classification method is a method of classifying the index measurement values ​​based on the range obtained from the average value and standard deviation of the index measurement values ​​for each type of motion analysis index. The second classification method is a method of classifying the index measurement values ​​by cluster analysis for each type of motion analysis index. These classification methods make it possible to classify each type of motion analysis index based on the similarity between the index measurement values.

[0035] In the first classification method, the classification processor 85 classifies the index measurement values ​​using the average value and standard deviation. The classification processor 85 classifies the time series data of the index measurement values ​​according to whether or not they are within the range of the average value ± the standard deviation. This makes it possible to distinguish between normal and abnormal index measurement values, or between index measurement values ​​that indicate a stable running form and index measurement values ​​that indicate an unstable running form.

[0036] In the second classification method, the classification processing unit 85 classifies the index measurement values ​​by cluster analysis. As a preparatory process for the cluster analysis, the classification processing unit 85 calculates the average value and standard deviation of the data for each predetermined unit in the time series data of the index measurement values ​​from the start point to the goal point. That is, after setting a window of a predetermined length from the start point in the time series data of the index measurement values, the window is slid over the entire range to the goal point, and a window process is executed to calculate the average value and standard deviation of the index measurement values ​​in the window in sequence. The average value and standard deviation values ​​thus calculated are clustered into a predetermined number of groups by cluster analysis, and the index measurement values ​​are classified into a plurality of groups and labeled with group labels. Details of the cluster analysis will be described later.

[0037] The first classification method or the second classification method may be used depending on the index measurement value, or the classification method may be set in advance depending on the index measurement value. Alternatively, the classification processing unit 85 may execute classification processing using both the first classification method and the second classification method for each index measurement value, and determine which method to adopt depending on the execution result. For example, if the execution result using one classification method does not allow classification into an appropriate number, the other classification method may be adopted.

[0038] The feedback acquisition unit 86 acquires feedback information input by the user 10 regarding the run. The feedback information includes comments in free text format input by the user 10 regarding the run. The feedback information may include review information in a simple button selection format or item selection format. The review information is input in the form of selecting, for example, multiple types of facial expression marks, multiple types of body part marks, multiple types of distance marks, and multiple types of time marks.

[0039] The analysis processing unit 87 extracts key points from the feedback information based on a predetermined extraction condition. The extraction condition for extracting key points from the feedback information varies depending on the format of the feedback information. For example, when the feedback information includes a comment of one or more sentences, key points are extracted from the text of each comment by language analysis based on a predetermined extraction condition. The extraction condition is determined for each of at least four items of attention that the user 10 subjectively recognizes regarding the running state. The four items of attention are, for example, information on the running form, information on the progress point, information on the mental state, and information on comparison with other runs. The key points of the items of attention are extracted from the comments by language analysis with each of the four items of attention as an analytical viewpoint.

[0040] For example, a plurality of predefined keywords are stored as a word list to extract content corresponding to each attention item from the comment text. A word list is predefined and stored for each of the four attention items. The analysis processing unit 87 extracts the main points for each attention item by extracting at least one of the plurality of keywords from the comment text included in the feedback information.

[0041] When the feedback information includes review information in a simple button selection format or item selection format, the analysis processing unit 87 extracts the matters that the user 10 paid attention to regarding the driving as main points according to the type of mark in the review information. The analysis processing unit 87 creates a judgment table that tabulates the main points extracted for each of the matters of attention from the feedback information.

[0042] The driving analysis system 100 may further include a language model server 160 in which a language model for language analysis of comments included in the feedback information is constructed, or the language model may be constructed inside the driving analysis server 60. Alternatively, the language model server 160 may be a cloud-type large-scale language model server constructed outside the system via an API (Application Programming Interface). Since learning to construct a language model from scratch requires huge computational costs, a unique language model may be efficiently constructed by using a pre-trained external large-scale language model, or by fine-tuning such an external language model. In addition, in such fine tuning, the provider of the driving analysis server 60 may re-train using a data set of past comments accumulated in its own domain, or may re-train using a data set of huge comments provided by a third party.

[0043] The analysis processing unit 87 may perform language analysis of the comments included in the feedback information using a language model of the language model server 160. As a language analysis using a language model, for example, a method of emotion analysis that reads the emotion of the user 10 from the sentence of the comment may be used to analyze the type and degree of emotion of the user 10 for each sentence. This can improve the accuracy of the comment analysis.

[0044] The mode determining unit 90 determines the output mode of the classified index measurement value based on a comparison with a predetermined comparison target. Here, the "output mode" refers to the mode of information to be output to the information terminal 50, including the type of information to be displayed on the screen of the information terminal 50, the display content, the display format, the data content to be transmitted to the information terminal 50, and the like. The mode determining unit 90 determines the type of information to be output, the comparison target of the information, the output format of the information, and the like, based on an instruction from the user 10 via an input to the input / output unit 51 of the measurement device 20. The mode determining unit 90 includes an instruction acquiring unit 91, an output target determining unit 92, a comparison target setting unit 93, a comparison processing unit 94, a change condition setting unit 95, a change detection unit 96, an evaluation determining unit 97, and an output data generating unit 98.

[0045] In this embodiment, there is a first output target determination method in which the analysis processing unit 87 outputs index measurement values ​​that match key points extracted from feedback information with a predetermined matter of interest as the analysis viewpoint, and a second output target determination method in which the output target is an index measurement value that corresponds to the analysis viewpoint selected by the user 10. The specification may include both the first output target method and the second output target method, or may include only one of them.

[0046] The instruction acquisition unit 91 acquires an instruction from the user 10 via an input to the input / output unit 51 of the measurement device 20 as a specification having the second output target method. The instruction acquisition unit 91 may set the index measurement value corresponding to the selected analysis viewpoint as an output target or a comparison target by having the user 10 select an analysis viewpoint that the user 10 is interested in. For example, seven types of analysis viewpoints are prepared as multiple types of analysis viewpoints that the user 10 can select: "grounding with less strain," "stable posture," "linkage of the whole body around the pelvis," "smooth center of gravity movement," "powerfulness of movement," "symmetry," and "speed of the race." Each of these is an analysis viewpoint that individually affects one or more index measurement values.

[0047] The output target determination unit 92 sets the index measurement value corresponding to the analysis viewpoint selected by the user 10 as the output target. For example, the index measurement value corresponding to the analysis viewpoint of "ground contact with less burden" is the vertical movement height ratio, the landing impact, and the kicking acceleration. The index measurement value corresponding to the analysis viewpoint of "stable posture" is the left and right tilt of the pelvis and the backward tilt of the trunk. The index measurement value corresponding to the analysis viewpoint of "whole body linkage with the pelvis as the axis" is the kicking time, the rotation timing of the pelvis, and the sinking of the waist. The index measurement value corresponding to the analysis viewpoint of "smooth center of gravity movement" is the deceleration amount and the left and right impact. The index measurement value corresponding to the analysis viewpoint of "strength of movement" is the stride height ratio, the amount of pelvic rotation, and the amount of pelvic lift. All the index measurement values ​​correspond to the analysis viewpoint of "left and right symmetry". The index measurement value corresponding to the analysis viewpoint of "speed of the race" is the running speed.

[0048] The output target determination unit 92 may output all index measurement values ​​when there is no particular instruction from the user 10. The output target determination unit 92 may output all measurement sections for one index measurement value, or may narrow down the output targets to a specific measurement section based on the instruction from the user 10 when the instruction acquisition unit 91 acquires an instruction to select a measurement section that the user 10 is concerned about. The output target determination unit 92 may narrow down the output targets to a measurement section in which a particular abnormality is observed or a measurement section showing a particularly excellent result, as detected by the change detection unit 96 described later.

[0049] The output data generation unit 98 generates output data so as to include the index measurement values ​​that have been selected as output targets by the output target determination unit 92. The output data defines at least the content to be displayed on the screen by the input / output unit 51 of the measurement device 20.

[0050] The mode determining unit 90 may output the index measurement value in different output modes depending on what value is to be compared. The comparison target setting unit 93 sets some reference value as the comparison target. The comparison target setting unit 93 may set, for example, the average values ​​for each group label classified by the classification processing unit 85 to be compared. The comparison target setting unit 93 may set, for example, the average value in the time series data of the index measurement value or the average value in a range limited to a predetermined measurement section as the comparison target, or may set the same type of index measurement value in another measurement section as the comparison target. When the index measurement value is that of one of the left and right legs, the comparison target setting unit 93 may set the index measurement value of the other leg as the comparison target. The comparison target setting unit 93 may set the same type of index measurement value at another date and time or by another user as the comparison target.

[0051] The comparison processing unit 94 compares the index measurement value to be output with the comparison target set by the comparison target setting unit 93. For example, when the average value for each group label classified by the classification processing unit 85 is used as the comparison target, the comparison processing unit 94 compares the average values ​​for each group label. The evaluation determination unit 97 evaluates each group of index measurement values ​​classified by the classification processing unit 85 based on the comparison result by the comparison processing unit 94. The evaluation determination unit 97 determines the evaluation of the index measurement value for the index measurement value set as the output target by the output target determination unit 92 based on the detection result by the change detection unit 96 in addition to the comparison result by the comparison processing unit 94.

[0052] The change detection unit 96 detects a change that meets a predetermined change condition in the time series data of the index measurement value to be output. The change condition setting unit 95 sets the change condition to be detected by the change detection unit 96. The change condition setting unit 95 may set a specific change condition as a standard detection condition as the change condition, or may set a condition specified by the user 10 as the change condition. The change condition may be, for example, a condition that is set so that an index measurement value is detected as a noteworthy value when the difference between a predetermined reference value and the index measurement value exceeds a predetermined range. For example, an index measurement value that is significantly better or significantly worse than the reference value may be detected as a noteworthy value. The reference value to be compared may be a comparison target set by the comparison target setting unit 93.

[0053] The change detection unit 96 creates a change list table based on data related to the index measurement values, measurement sections, and magnitude of change in which changes were detected. The evaluation determination unit 97 compares the change list table with the judgment table to calculate an overall evaluation value for each motion analysis index. More specifically, the evaluation determination unit 97 calculates a calculated value as an overall evaluation value by combining a normalized value of the magnitude of change in the index measurement values ​​in the measurement sections in which changes were detected by the change detection unit 96 among the classified index measurement values ​​and a normalized value of the degree of agreement between the key points extracted from the feedback information and the motion analysis indexes and the changes in those measurements. The evaluation determination unit 97 prioritizes the motion analysis evaluation values ​​with the highest overall evaluation value as the targets to be presented to the user 10 as evaluation points.

[0054] When an analytical perspective of interest is selected by the user 10 as the second output target method, the evaluation determination unit 97 determines an evaluation of the index measurement value for the index measurement value selected as the output target by the output target determination unit 92, based on the comparison results by the comparison processing unit 94 and the detection results by the change detection unit 96.

[0055] The output data generation unit 98 generates output data in such a way that the measurement section of the index measurement value in which a change was detected by the change detection unit 96 is distinguished from other index measurement values, i.e., measurement sections in which no change was detected. The output data generation unit 98 also generates output data in such a way that the motion analysis index of the index measurement value in which a change was detected by the change detection unit 96 is distinguished from other motion analysis indexes, i.e., index measurement values ​​of motion analysis indexes in which no similar change was detected. When an evaluation is determined by the evaluation determination unit 97, the output data generation unit 98 generates output data to include the evaluation result.

[0056] The output unit 99 outputs information related to the index measurement value based on the determined output mode. For example, the output unit 99 outputs output data including information related to the index measurement value that matches the analysis result of the feedback information to the information terminal 50 via the communication unit 62.

[0057] FIG. 4 is a flowchart showing the process of the running analysis system. The information acquisition unit 70 acquires position information and motion information measured by the measurement device 20 while the user 10 is running in a race or the like (S10), and the measurement acquisition unit 80 acquires the measurement values ​​of various motion analysis indices based on the acquired position information and motion information (S12). The classification processing unit 85 classifies the index measurement values ​​corresponding to various motion analysis indices by a predetermined classification method (S14). As a first output target determination method, if there is feedback information including comments or review information, the analysis processing unit 87 analyzes the feedback information (S15) and creates a judgment table (S16). As a second output target determination method, if there is an input of an analysis perspective by the operation of the user 10, the instruction acquisition unit 91 acquires information of the analysis perspective (S17), and the output target determination unit 92 determines the output target of the index measurement values ​​according to the analysis perspective (S18). The comparison processing unit 94 executes a predetermined comparison process on the index measurement values ​​(S20), and the change detection unit 96 detects a motion analysis index or its index measurement value corresponding to a predetermined change condition (S21). The change detection unit 96 creates a change list table for the index measurement values ​​in which changes have been detected (S22). The evaluation determination unit 97 calculates an overall evaluation value by adding up the magnitude of the detected change and the number of matches with the feedback information (S23). The evaluation determination unit 97 determines the evaluation of the index measurement values ​​(S24). The output data generation unit 98 generates output data for the index measurement values ​​to be output, together with the comparison results, detection results, and evaluations. The output unit 99 outputs the index measurement values ​​based on the output data generated by the output data generation unit 98 (S26). Note that the processes of S12 to S26 may be executed in various orders, and the order of S12 to S26 in the flowchart of this figure is merely an order determined for convenience. In particular, processes such as creating a judgment table (S16), acquiring an analysis perspective (S17), determining an output target (S18), creating a change list table (S22), and calculating an overall evaluation value (S23) may be executed at different times from other processes.

[0058] 5 is a schematic diagram showing an example of a feedback input screen. After a race or the like, the user 10 looks back on the run and inputs his / her impressions on the feedback input screen of an application program or web application for viewing the run data. The feedback screen 200 includes a result display field 201, an emotion selection field 202, an item selection field 204, and a free text input field 206.

[0059] The result display field 201 is an area where the measurement results by the measuring device 20 are displayed as running details such as a race. The measurement results include information such as the measurement date and time, running time, running pace, and running distance. The emotion selection field 202 and item selection field 204 are areas where buttons or items that the user 10 can select as review information are displayed. The free text input field 206 is an area where the user 10 can input free text as a comment.

[0060] In the emotion selection field 202, five facial expression marks 210a-e are displayed in a button selection format. The first facial expression mark 210a indicates a very bad state or a very dissatisfied state. The second facial expression mark 210b indicates a somewhat bad state or a somewhat dissatisfied state. The third facial expression mark 210c indicates an average or moderate state that is neither good nor bad. The fourth facial expression mark 210d indicates a somewhat good state or a somewhat satisfied state. The fifth facial expression mark 210e indicates a very good state or a very satisfied state. The user 10 looks back on his / her running and selects from the facial expression marks 210a-e the mark that best represents his / her thoughts on whether or not the running went according to plan, his / her mental fatigue state, and his / her physical fatigue state.

[0061] In the item selection field 204, five body part marks 211, one weather mark 212, four distance marks 213, and two time marks 214 are displayed in a button selection format. The user 10 selects the body part mark 211 or weather mark 212 indicating a body part that the user is concerned about in association with the type of facial expression mark selected in the emotion selection field 202. The first body part mark 211a indicates "Breathing". The second body part mark 211b indicates "Upper Legs". The third body part mark 211c indicates "Lower Legs". The fourth body part mark 211d indicates "Knees". The fifth body part mark 211e indicates "Feet". The weather mark 212 indicates "Weather".

[0062] The user 10 selects a distance mark 213 indicating a point or section where a change in mental state or physical state occurred in association with the type of facial expression mark 210 selected in the emotion selection field 202. The first distance mark 213a indicates "up to 10km". The second distance mark 213b indicates "up to 20km". The third distance mark 213c indicates "up to 30km". The fourth distance mark 213d indicates "up to Finish". The user 10 selects a time mark 214 indicating a time point or time period where a change in mental state or physical state occurred in association with the type of facial expression mark 210 selected in the emotion selection field 202. The first time mark 214a indicates "first half". The second time mark 214b indicates "second half".

[0063] In the free text input field 206, the user 10 inputs his / her impressions, points he / she noticed, and points he / she noticed in the form of free text after looking back on the running. For example, one or more sentences such as "I aimed for a sub-4 time, and the first 10 km was as expected," "After that, my pace increased to about 10 to 20 seconds faster per kilometer than expected, and I ran out of energy when I passed 30 km," "I also felt that there was a problem with the way I took in the energy in the second half," "In terms of form, I was swinging my arms a little too forcefully, and I felt fatigue around my shoulders," and "I am trying to use an efficient form that uses as little energy as possible, but my right calf was about to cramp, and I feel like I am still only halfway there" are input in the free text input field 206.

[0064] When feedback information is input by the user 10, if the feedback information is input in both button selection format and free text format, both are analyzed. If the feedback information is input in either button selection format or free text format, only one of them is analyzed.

[0065] As a modified example, the feedback screen 200 may include only one or two of the emotion selection field 202, the item selection field 204, and the free text input field 206. In other words, the feedback information may include only comments or only selective review information.

[0066] If the feedback information includes a comment in free text format, the analysis processing unit 87 extracts the sentences included in the comment. If the comment includes multiple sentences, each sentence is analyzed. In one sentence, it is analyzed whether it includes references to each of the four analysis perspectives of "form information," "course information," "mental information," and "presence or absence of comparison." As will be described later, it is also analyzed whether there have been significant changes in the measured values ​​of the motion analysis index from the four analysis perspectives of "form information," "course information," "mental information," and "presence or absence of comparison." By comparing the analysis results of the comment with the analysis results of the measured values ​​of the motion analysis index, it is determined which change in the measured values ​​of the motion analysis index corresponds to the point mentioned in the comment. The evaluation determination unit 97 considers the change in the measured values ​​of the motion analysis index that is determined to have been mentioned in the comment to be a more important change than other changes, and adds weight to that change.

[0067] "Form information" is information about the running form that the user 10 subjectively focuses on while running in a race or the like. Contents corresponding to "form information" that may be included in a comment may include, for example, keywords corresponding to "body parts," "motion analysis index," and "more specific parts and directions" in the word list. For example, contents that refer to running form and body parts such as "arm swing," "shoulder area," "efficient form that uses as little energy as possible," "footwork," "landing," "calf," "thigh," and "leg" are applicable. In addition, contents that refer to up and down, left and right, and front and back that more specifically indicate running form and body parts such as "right calf," "pain in left knee," "up and down movement," and "forward leaning posture" are applicable to "form information." Contents that specifically refer to each motion analysis index are also applicable to "form information."

[0068] "Course information" is information about a progress point that the user 10 subjectively paid attention to during a race or other run. Content corresponding to "course information" that may be included in a comment may include keywords corresponding to course conditions, such as "distance," "location," "range," "time," "time zone," "section," "road surface," "undulations," and "weather," in the word list. For example, content that refers to a run location or distance point, such as "the first 10 km," or "after 30 km," or content that refers to a time zone, such as "the beginning," "first half," or "last 10 minutes," is relevant. Content that refers to the type of road surface or undulations, such as "track," "uphill," "downhill," or "up-and-down," is also relevant to "course information." Content that refers to the weather, such as "direct sunlight," "temperature rising," "the road surface was wet," "humid," or "headwind," is also relevant to "course information."

[0069] "Mental information" is information about the mental state that the user 10 subjectively focuses on while running in a race or the like. As content corresponding to "mental information" that may be included in a comment, the word list may include keywords corresponding to types of emotions such as "satisfaction," "dissatisfaction," and "ambition." For example, the word list may include words such as "as expected," "as aimed," "done," "tried hard," "pushed," "happy," and "satisfied" as keywords indicating the emotion of "satisfaction." For example, the word list may include words such as "still have a long way to go," "got to ~," "no good," "fell," "tired," "painful," and "fatigue" as keywords indicating the emotion of "ambition." For example, the word list may include words such as "want to do ~," "still have a long way to go," and "want to be ~" as keywords indicating the emotion of "ambition." In this way, content that expresses the subjective impressions and feelings of the user 10 corresponds to "mental information."

[0070] "Whether or not there is a comparison" is information regarding a comparison with other runs that the user 10 subjectively noted during a run such as a race. Content corresponding to "whether or not there is a comparison" that may be included in a comment may include, for example, keywords corresponding to comparison with "the past," "others," "expectations," and "plans" in a word list. For example, content that refers to a comparison with the user 10's expected plan or past races, such as "a pace about 10 to 20 seconds faster per kilometer than expected" or "a better pace than last year," corresponds to "whether or not there is a comparison." In addition, content that refers to a comparison with other runners, such as "30 seconds behind the leader" or "overtaken by a sub-4 pace runner," also corresponds to "whether or not there is a comparison."

[0071] When the feedback information includes review information, the analysis processing unit 87 identifies the subjectively recognized attentional matter of the user 10 from at least three perspectives of "form information," "course information," and "mental information" according to the type of mark selected as the review information. For example, when any of the facial expression marks 210a-e is selected, the fatigue level or satisfaction level according to the type of the selected facial expression mark is identified as "mental information." When any of the body part marks 211a-e is selected in relation to the content of the "mental information," the analysis processing unit 87 identifies the attention paid to the body part according to the type of the selected body part mark as "form information." When any of the weather mark 212, distance marks 213a-d, and time marks 214a, b is selected in relation to the content of the "mental information" or "form information," the analysis processing unit 87 identifies the content according to the type of the selected mark as "course information."

[0072] FIG. 6 is a schematic diagram showing an example of a judgment table created by the analysis processing unit. The analysis processing unit 87 creates a judgment table 220 based on the extracted or identified matters of interest, namely, "form information," "course information," "mental information," and "presence or absence of comparison." In the illustrated judgment table 220, the first row 221 shows the matters of interest extracted from the first sentence. The second row 222 shows the matters of interest extracted from the second sentence. The third row 223 shows the matters of interest extracted from the third sentence. The fourth row 224 shows the matters of interest extracted from the fourth sentence. The fifth row 225 shows the matters of interest extracted from the fifth sentence. The sixth row 226 shows the matters of interest identified based on the mark. The first column 231 shows form information. The second column 232 shows course information. The third column 233 shows mental information. The fourth column 234 shows the presence or absence of comparison.

[0073] Key points are extracted from each of the first to fifth sentences in the comments of the feedback information. The key points "The first 10km" and "As expected" are extracted from the first sentence, "Aiming for a sub-4 time, the first 10km was as expected." Based on these key points, "~10km" is displayed in the "Progress point" column in the second column 232 of the first row 221, "Satisfied" is displayed in the "Satisfaction" column in the third column 233, and "Yes" is displayed in the fourth column 234.

[0074] From the second sentence, "After that, my pace increased to about 10 to 20 seconds per kilometer faster than expected, and I ran out of energy when I passed 30 km," the key points that "I increased my pace more than expected," "ran out of energy when I passed 30 km" are extracted. Based on these key points, "30 km~" is displayed in the "Progress point" column in the second column 232 of the second row 222, "Dissatisfied" is displayed in the "Satisfaction" column in the third column 233, and "Yes" is displayed in the fourth column 234.

[0075] From the third sentence, "I also felt there were issues with the heat countermeasures in the second half," the main points of "second half," "heat," and "issues" are extracted. Based on these main points, "heat" and "second half" are displayed in the "Environment" and "Progress" columns in the second column 232 of the third row 223, and "Dissatisfaction" is displayed in the third column 233.

[0076] From the fourth sentence, "In terms of form, I think I put a little too much force into my arm swing, and I feel fatigue around my shoulders," the key points of "arms," ​​"shoulders," and "fatigue" are extracted. Based on these key points, "arms and shoulders" is displayed in the "motion analysis index or body part" column in the first column 231 of the fourth row 224, and "dissatisfaction" is displayed in the third column 233.

[0077] From the fifth sentence, "I'm trying to use an efficient form that uses as little energy as possible, but my right calf is about to cramp, and I feel like I'm still only halfway there," the key points of "efficient form," "right calf," "about to cramp," and "still only halfway there" are extracted. Based on these key points, "efficient form, calf," and "right" are displayed in the "Motion analysis index or body part" and "Left / Right" columns in the first column 231 of the first row 221, and "Dissatisfied" and "Ambition" columns in the third column 233 are displayed.

[0078] The key points of "slightly bad condition or slightly dissatisfied" indicated by the second facial expression mark 210b and "up to 30km" indicated by the third distance mark 213c in the review information of the feedback information are identified. Based on these key points, "up to 30km" is displayed in the "Progress" field in the second column 232 of the sixth row 226, and "Dissatisfied" is displayed in the "Satisfaction" field in the third column 233.

[0079] FIG. 7 is a schematic diagram of a change list table created by the change detection unit. The change list table 240 shows the motion analysis indexes in the measurement sections where changes were detected by the change detection unit 96 among the classified measurement values, and the degree of change in those measurement values. In the change list table 240, the first column 251 shows the type of motion analysis index. The second column 252 shows the body parts related to the motion analysis indexes. The third column 253 shows the environment such as the course situation related to the change in the index measurement value. The fourth column 254 shows the point where the index measurement value changed. The fifth column 255 shows the difference between the left and right index measurement values. The sixth column 256 shows the direction of the change in the index measurement value. The seventh column 257 shows the normalized value of the magnitude of the change in each index measurement value. The eighth column 258 shows the number of matches with the key points included in the judgment table as a normalized value of the degree of match with the feedback information.

[0080] The motion analysis indexes and their measured values ​​in the measurement section where the change was detected by the change detection unit 96 are shown in the first row 241 to the n-th row 244. The first row 241 shows the measured values ​​of the indexes for the motion analysis index "pitch" and data on the changes. The first column 251 shows the motion analysis index "pitch", and the second column 252 shows "thighs, legs" as the corresponding body parts. The fourth column 254 shows "20km" as the point where the measured pitch value changed, and the fifth column 255 shows "none" as the difference between the left and right pitches. The sixth column 256 shows "improving" as the direction of the change in pitch, and the seventh column 257 shows "0.5" as the normalized value of the magnitude of the change in pitch. The eighth column 258 shows the number of matches with the main points "0" as the normalized value of the degree of match with the feedback information.

[0081] The second row 242 shows data on the index measurement value and its change for the motion analysis index "vertical movement". The first column 251 shows the motion analysis index "vertical movement", and the second column 252 shows "hip joint, trunk, buttocks, front thigh" as the corresponding body parts. The third column 253 shows "downhill" as the environment where the vertical movement measurement value changed. The fourth column 254 shows "20km" as the point where the vertical movement measurement value changed, and the fifth column 255 shows "left" as the difference between the left and right vertical movement. The sixth column 256 shows "getting worse" as the direction of the change in the vertical movement, and the seventh column 257 shows "0" as the normalized value of the magnitude of the change in the vertical movement. The eighth column 258 shows "0" as the number of matches with the main points as the normalized value of the degree of match with the feedback information.

[0082] The third row 243 shows data on the index measurement value and its change for the motion analysis index "landing impact". The first column 251 shows the motion analysis index "landing impact", and the second column 252 shows "thighs, legs" as the corresponding body part. The third column 253 shows "downhill" as the environment where the landing impact measurement value changed. The fourth column 254 shows "40km" as the point where the landing impact measurement value changed, and the fifth column 255 shows "none" as the difference between the landing impacts on the left and right. The sixth column 256 shows "getting worse" as the direction of the change in the landing impact, and the seventh column 257 shows "1.5" as the normalized value of the magnitude of the change in the landing impact. The eighth column 258 shows the number of matches with the main points "1" as the normalized value of the degree of match with the feedback information.

[0083] The nth row 244 shows data on the index measurement value and its change for the motion analysis index "backward lean of the trunk". The first column 251 shows the motion analysis index "backward lean of the trunk", and the second column 252 shows "back, shoulder, hip joint, trunk" as the corresponding body parts. The third column 253 shows "uphill" as the environment in which the measurement value of the backward lean of the trunk changed. The fourth column 254 shows "30km" as the point at which the measurement value of the backward lean of the trunk changed, and the fifth column 255 shows "right" as the difference between the left and right backward lean of the trunk. The sixth column 256 shows "getting worse" as the direction of the change in the backward lean of the trunk, and the seventh column 257 shows "2.2" as the normalized value of the magnitude of the change in the backward lean of the trunk. The eighth column 258 shows the number of matches with the main points "2" as the normalized value of the degree of match with the feedback information. For example, the main points “shoulder” and “dissatisfied” extracted from the fourth sentence in the comments of the feedback information match “shoulder” in the second column 252 and “getting worse” in the sixth column 256, and this is counted as a match of “1.” The value obtained by counting the matches with the feedback information for each motion analysis indicator in this way is shown in the eighth column 258.

[0084] The evaluation determination unit 97 determines the total evaluation value for each motion analysis index by adding the magnitude of change shown in the seventh column 257 to the number of matches shown in the eighth column 258. As a modified example, the total evaluation value may be determined by multiplying the magnitude of change by the number of matches as a weighting. The evaluation determination unit 97 preferentially selects the motion analysis indexes with the highest total evaluation value from among the multiple motion analysis indexes, and determines them as the evaluation points to be presented to the user 10 as described below. In the example shown in FIG. 7, the motion analysis index "backward lean of the trunk", which is the n-th row 244, has the highest total evaluation value, so that the evaluation on "backward lean of the trunk" is preferentially presented to the user 10. Note that there may be cases where there are multiple motion analysis indexes with the highest total evaluation value in the change list table 240, and the respective total evaluation values ​​are the same. In such a case, the motion analysis index with the larger "magnitude of change" shown in the seventh column 257 may be preferentially presented to the user 10, or the motion analysis index with the larger "number of matches" shown in the eighth column 258 may be preferentially presented to the user 10.

[0085] FIG. 8 shows an example of classifying index measurements using the first classification method. In the first classification method, the classification processing unit 85 classifies the index measurements using the average value and the standard deviation. In the graph of FIG. 8, the time series data of the index measurements indicating the amount of rotation of the pelvis are plotted in correspondence with the travel distance points as the index measurements to be classified. The vertical axis indicates the amount of rotation of the pelvis [deg], and the horizontal axis indicates the distance points [km]. The solid line 110 indicates the average value of the time series data of the amount of rotation of the pelvis, and the dashed line 112 indicates the range of the average value ± standard deviation of the time series data of the amount of rotation of the pelvis. The index measurements that are within the range of the two dashed lines 112 are plotted as circles, and the index measurements that are outside the range of the dashed lines 112 are plotted as diamonds. In the example of this figure, the diamond plots of the plot group 114a to k are index measurements that are outside the range of the dashed lines 112, and are classified in a manner that distinguishes them from the index measurements of the circular plots that are within the range of the dashed lines 112. The index measurement values ​​shown as circular plots are classified as index values ​​indicating a stable running form that is within the standard deviation, and the index measurement values ​​shown as diamond plots are classified as index values ​​indicating an unstable or disruptive running form that is outside the standard deviation.

[0086] In this way, by classifying the time series data of index measurements based on whether they are within the range of the average value ± standard deviation, it is possible to distinguish between index measurements that indicate a stable running form and those that indicate an unstable running form.

[0087] FIG. 9 shows a preparatory process for classifying index measurements using the second classification method. In the second classification method, the classification processing unit 85 classifies index measurements using cluster analysis. In the graph of FIG. 9, time series data of index measurements indicating running pace as the index measurements to be classified is plotted in correspondence with the distance traveled. The vertical axis indicates running pace [min / km], and the horizontal axis indicates distance traveled [km]. In FIG. 9(a), the index measurements are plotted as they are. Here, as a preparatory process, a window W of a predetermined length is set as shown in the figure, and window processing is performed to calculate the average value and standard deviation of the index measurements included in the window W while sliding the window W from the start point to the goal point. The window W covers a range that may include multiple, for example, five, consecutive index measurements in a chronological order. If the index measurements are measured every 100 m, a window of a 500 m section is set. Alternatively, if the index measurements are measured in units of time, for example, every 10 seconds, a time window of 60 seconds per section is set.

[0088] 1st Window W 1 covers the first to fifth index measurements, and the second window W 2 covers the 2nd to 6th index measurements, and sets the window to slide by one index measurement at a time. It slides toward the goal point, and the n-1th window W n-1 to the last nth window W n When the window processing is completed, the window processing is completed. In Fig. 9(b), the time series data of the average value of the index measurement value in each window calculated by the window processing is plotted in correspondence with the travel distance point of each window. In Fig. 9(c), the time series data of the standard deviation of the index measurement value in each window calculated by the window processing is plotted in correspondence with the travel distance point of each window.

[0089] FIG. 10 shows a process of classifying index measurements using the second classification method. The graph in FIG. 10(a) is a scatter plot in which the average value in FIG. 9(b) calculated for each window in FIG. 9 is plotted on the horizontal axis and the standard deviation in FIG. 9(c) is plotted on the vertical axis. All plots are clustered, for example, by the K-means method. The number of clusters K may be set to an estimated value using the elbow method, for example, or a fixed value such as "4" may be set when it is desired to uniformly divide the plots into four stages regardless of the type of driving evaluation value. As shown in the figure, all plots are clustered according to the distance from K centroids, and are classified into a first group 120 with a centroid 121 as the core, a second group 122 with a centroid 123 as the core, a third group 124 with a centroid 125 as the core, and a fourth group 126 with a centroid 127 as the core. Each plot is labeled with the group label into which it is classified. The graph in FIG. 10(b) shows the average value in FIG. 9(b). The plots are classified into a first group 120, a second group 122, a third group 124, and a fourth group 126, which are the group labels labeled in Fig. 10(a). The number of clusters may vary depending on the index measurement value, or may vary depending on the variability of the index measurement value for each user 10 or for each run.

[0090] In this way, by classifying the time-series data of index measurements using cluster analysis, it is possible to categorize the index measurements based on the degree of similarity between their characteristics, and to objectively grasp and detect the variation in the index measurements.

[0091] FIG. 11 is a bar graph showing the proportion of classification for each motion analysis index. For each motion analysis index, the index measurement value is classified by the first or second classification method described above, and the time series data is divided at the change point of the classification. The vertical axis shows each index measurement value, and the horizontal axis shows the distance traveled [km]. In the case of a marathon, the entire distance is set to 42 km, and the bar graph of the index measurement value of each motion analysis index is divided by the proportion of each classification. For example, the motion analysis index "stiffness" is divided into two classifications at the border around 34.5 km. The motion analysis index "pelvic rotation timing" is divided into three classifications at the borders around 4 km and 14.5 km. For example, in the case of a motion analysis index classified by the first classification method, the time series data of the index measurement value is divided at the border of whether it is within or outside the standard deviation range. In the case of a motion analysis index classified by the second classification method, the time series data of the index measurement value is divided at the border of the change point of the group label.

[0092] FIG. 12 shows a state where the index measurement values ​​divided in the horizontal bar graph of FIG. 11 are evaluated by another criterion. The comparison processing unit 94 evaluates the classification of each motion analysis index using the comparison target set by the comparison target setting unit 93 as a reference value. For example, when the average value of each group label classified by the classification processing unit 85 is used as the comparison target, the comparison processing unit 94 compares the average values ​​of each group label. The evaluation determination unit 97 evaluates each group of index measurement values ​​classified by the classification processing unit 85 based on the comparison result by the comparison processing unit 94, and displays the groups of index measurement values ​​in different colors according to the order of the average values ​​of each group as shown in FIG. 12 in order to distinguish them from other groups of index measurement values. The first color 130 indicates the best value classification (for example, shown in blue). The second color 131 indicates a slightly better value classification (for example, shown in green). The third color 132 indicates a slightly worse value classification (for example, shown in yellow). The fourth color 133 indicates the worst value classification (for example, shown in red). As a modified example, the comparison processing unit 94 may calculate the average value of the index measurement value for the entire race, compare it with the average value for the entire race for each group label, and display the comparison results in different colors to distinguish them.

[0093] FIG. 13 shows an example of displaying the output target narrowed down to the index measurement value detected as the noteworthy value. In FIG. 13(a)(b), the comparison result by the comparison processing unit 94 is displayed limited to a classification significantly better than a predetermined reference value to be compared and a classification significantly worse than the predetermined reference value to be compared, and the significantly better classification is shown in a fifth color 134 (e.g., red), and the significantly worse classification is shown in a sixth color 135 (e.g., blue). The comparison target in FIG. 13(a) is, for example, the data of the user 10 himself or other runners. The comparison target may be set from the index measurement values ​​of the same runner in the same race, from the index measurement values ​​of the same runner in a past race, or from the index measurement values ​​of other runners. When the index measurement values ​​of other runners are used as the comparison target, the index measurement values ​​of other runners in the same measurement section may be used as the comparison target. For example, the comparison target in FIG. 13(b) may be the value of the previous window for each window in FIG. 9(a), or the value of the previous lap for each lap.

[0094] The change detection unit 96 detects the motion analysis index having the longest period in which a significantly good trend continues (in the example of FIG. 13(a) , "left / right tilt of the pelvis"), and the motion analysis index having the longest period in which a significantly bad trend continues (in the example of FIG. 13(a) , "deceleration amount").

[0095] The change detection unit 96 may detect a combination of motion analysis indexes having the longest section in which all of the multiple motion analysis indexes continue to show a significantly good trend, or a combination of motion analysis indexes having the longest section in which all of the multiple motion analysis indexes continue to show a significantly bad trend. For example, the change detection unit 96 detects a combination of landing impact / kicking acceleration (grounding with less strain) and lateral impact (smooth center of gravity movement) as a combination of motion analysis indexes having the longest section in which the significantly good trend continues in the final stage of the race. In this case, the evaluation determination unit 97 may determine, as the evaluation result, advice that the athlete is able to run while exerting his / her power without waste. In addition, the combination of the amount of rotation of the pelvis (power of movement) and kicking time (linkage of the whole body around the pelvis) is detected as a combination of motion analysis indexes having the longest section in which the significantly bad trend continues in the final stage of the race. In this case, the evaluation determination unit 97 may determine, as the evaluation result, advice that the athlete should reduce the kicking movement and run by using the pelvis to put his / her feet forward.

[0096] FIG. 14 shows an example of a motion analysis index having the longest section as a section where a significantly bad tendency continues. In this figure, the amount of pelvic rotation, which has the longest section as a section where a significantly bad tendency continues, is shown in a graph at the end of the race. As the index measurement value, time series data of the index measurement value indicating the amount of pelvic rotation is plotted in correspondence with the distance traveled. The vertical axis indicates the amount of pelvic rotation [deg], and the horizontal axis indicates the distance traveled [km]. In particular, the amount of pelvic rotation of the user 10 in section 139, which is the end of the race, continues to show a significantly bad tendency, and is detected by the change detection unit 96 as an important change. In many cases, the user 10 himself recognizes the occurrence of such an important change by sensation during the race, but the user 10 himself may not be able to recognize at what stage the change actually begins as a sign. According to this embodiment, important changes in objective data and the timing at which the change begins can also be detected as a sign, which becomes useful data for improving the level of the user 10.

[0097] FIG. 15 shows an example of a screen displaying the race results. When the user 10 issues an instruction to display an overview of the race results, a race summary screen 140 generated by the mode determination unit 90 is displayed. In the race summary screen 140, in addition to information showing the running condition such as the running distance, running time, average pace, average stride, and average pitch, a running form score based on the measured values ​​of various movement analysis indexes is displayed in a radar chart format. In addition, course information based on the running position information is displayed. The bottom column displays the evaluation contents including advice for the user 10.

[0098] Fig. 16 shows an example of a screen displaying a race evaluation. When the user 10 gives an instruction to display the race evaluation, a race review screen 142 generated by the manner determining unit 90 is displayed. In the race review screen 142, a first column 143 displays an evaluation of the running pace. A second column 144 displays a graph showing the progress of the running pace. A third column 145 displays an evaluation of the pace divided into the early, middle, and late stages of the race, and an evaluation of the form.

[0099] The fourth column 146 displays an evaluation based on an important change detected by the change detection unit 96, an evaluation on the motion analysis index determined by the evaluation determination unit 97, and an evaluation on the motion analysis index corresponding to the viewpoint selected by the user 10. The fourth column 146 displays a plurality of evaluation points determined by the evaluation determination unit 97, and at the instruction of the user 10, the screen switches to one as shown in the following figure, in which detailed explanations and graphs on each evaluation point are displayed.

[0100] 17 shows an example of a screen displaying a detailed explanation of the evaluation points. When the user 10 instructs to display a detailed explanation of the evaluation points, a point explanation screen 150 generated by the mode determination unit 90 is displayed. In the point explanation screen 150, an explanation of the motion analysis index itself, an explanation of a change when an important change is detected, an explanation of the comparison result with a specified comparison target, etc. are displayed in text and graphs.

[0101] For example, in the change list table 240 of FIG. 7, since the overall evaluation value of "backward lean of the trunk" shown in the nth row 244 is the highest, an evaluation of "backward lean of the trunk", which is one of the motion analysis indicators, is presented to the user 10 in FIG. 17. In the point explanation screen 150, as "point explanation", a point or time when an important change occurred in the motion analysis indicator "backward lean of the trunk", and how the body and posture have specifically changed are presented. In addition, a sentence explaining the meaning of the motion analysis indicator "backward lean of the trunk" itself is displayed. In addition, the process by which the motion analysis indicator "backward lean of the trunk" changed during the race is displayed in graphs and numerical values. In addition, as advice from the coach, a sentence is presented on how to improve or take measures in the future.

[0102] The present disclosure has been described above based on the embodiments. These embodiments are merely examples, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component and each processing process, and that such modifications are also within the scope of the present disclosure. Furthermore, the above-described embodiments can be generalized to obtain the following aspects.

[0103] [Aspect 1] an information acquisition unit that acquires position information and movement information of a user who is a runner, the position information and movement information being measured at each progress point during a run by a predetermined measuring device; a measurement acquisition unit that acquires measurement values ​​of a plurality of types of motion analysis indexes that indicate the running motion state of the user for each measurement section of a predetermined unit based on the position information and the motion information that are continuous in time series; a classification processing unit that classifies the measurement values ​​of the plurality of measurement sections by a predetermined classification method for classifying the measurement values ​​by characteristics for each type of the motion analysis index; a feedback acquisition unit that acquires feedback information input by a user regarding the running; an analysis processing unit that extracts key points from the feedback information based on predetermined extraction conditions; a mode determination unit that determines an output mode of the measurement values ​​by preferentially extracting, from the classified measurement values, the motion analysis indexes and their measurement values ​​that match the key points in a measurement section in which a change corresponding to a predetermined change condition is detected, as output targets; an output unit capable of outputting at least information relating to a measurement value that matches an analysis result of the feedback information based on the determined output mode; A driving analysis system comprising:

[0104] In the driving analysis system of the first aspect, the changes in driving that the user himself / herself perceives or notices based on the feedback information from the user can be regarded as particularly important changes and can be prioritized as evaluation targets. Also, the information that the user himself / herself particularly wants to know can be effectively presented.

[0105] [Aspect 2] The feedback acquisition unit acquires feedback information including comments input by a user regarding the running, The driving analysis system of aspect 1, wherein the analysis processing unit extracts text indicating noteworthy points that the user subjectively recognizes regarding the driving condition from comments included in the feedback information through language analysis as the key points.

[0106] In the driving analysis system of aspect 2, important changes can be captured more accurately based on linguistic information from the user, and these can be given priority for evaluation and presented to the user.

[0107] [Aspect 3] The driving analysis system of aspect 1 or 2, wherein the analysis processing unit extracts, from the comments included in the feedback information, text containing at least one of information indicating noteworthy points subjectively recognized by the user regarding the driving condition, information regarding driving form, information regarding progress points, information regarding mental state, and information regarding comparison with past driving, as the key points through language analysis.

[0108] In the driving analysis system of aspect 3, by analyzing language from multiple specific perspectives to capture important changes based on linguistic information from the user, it is possible to more accurately prioritize important changes and present them to the user.

[0109] Aspect 4 The driving analysis system according to any one of aspects 1 to 3, wherein the analysis processing unit stores a plurality of predefined keywords for extracting information indicating noteworthy points subjectively recognized by the user regarding the driving condition, such as information regarding driving form, information regarding progress points, information regarding mental state, and information regarding comparison with other drives, and extracts the key points by extracting at least one of the plurality of keywords from text included in the feedback information.

[0110] In the driving analysis system of aspect 4, by performing language analysis using multiple specific keywords to capture important changes based on linguistic information from the user, it is possible to more accurately prioritize important changes as evaluation targets and present them to the user.

[0111] Aspect 5 A driving analysis system as described in any one of aspects 1 to 4, characterized in that the aspect determination unit extracts as output targets from the classified measurement values ​​in order of largest calculated value that combines a normalized value of the degree of change of the motion analysis index and its measurement value in a measurement section in which a change corresponding to a predetermined change condition was detected, and a normalized value of the degree of match between the key points extracted from the feedback information and the motion analysis index and its measurement value.

[0112] In the driving analysis system of aspect 5, by calculating an objective evaluation value using specific numerical values ​​and a simple calculation formula, it is possible to accurately capture important changes and present them to the user.

[0113] Aspect 6 acquiring position information and movement information of a user who is a runner, the position information and movement information being measured at each progress point during a run by a predetermined measuring device; acquiring measurement values ​​of a plurality of types of motion analysis indexes that indicate the running motion state of the user for each measurement section of a predetermined unit based on the position information and the motion information that are continuous in time series; classifying the measured values ​​of the plurality of measurement sections for each type of the motion analysis index according to a predetermined classification method for classifying the measured values ​​according to characteristics; obtaining feedback information input by a user regarding the trip; extracting key points from the feedback information based on a predetermined extraction condition; detecting a change corresponding to a predetermined change condition from the classified measured values; a step of matching the measurement section in which the change is detected among the classified measurement values ​​with the key point; extracting, as output targets, the motion analysis indexes and their measurement values ​​that match the key points in the measurement section in which the change is detected; determining an output mode of the measurement value based on the extraction result of the output object; outputting at least information related to the classified measurement values ​​based on the determined output manner; A driving analysis method comprising:

[0114] In the driving analysis method of the sixth aspect, the change in driving that the user himself / herself perceives or notices based on the feedback information from the user can be regarded as a particularly important change and can be given priority for evaluation. Also, the information that the user himself / herself particularly wants to know can be effectively presented. [Explanation of symbols]

[0115] 10 user, 20 measuring device, 30 information acquisition unit, 40 measurement value acquisition unit, 70 information acquisition unit, 80 measurement value acquisition unit, 85 classification processing unit, 86 feedback acquisition unit, 87 analysis processing unit, 90 behavior determination unit, 99 output unit, 100 driving analysis system.

Claims

1. an information acquisition unit that acquires position information and movement information of a user who is a runner, the position information and movement information being measured at each progress point during a run by a predetermined measuring device; a measurement acquisition unit that acquires measurement values ​​of a plurality of types of motion analysis indexes that indicate the running motion state of the user for each measurement section of a predetermined unit based on the position information and the motion information that are continuous in time series; a classification processing unit that classifies the measurement values ​​of the plurality of measurement sections by a predetermined classification method for classifying the measurement values ​​by characteristics for each type of the motion analysis index; a feedback acquisition unit that acquires feedback information input by a user regarding the running; an analysis processing unit that extracts key points from the feedback information based on predetermined extraction conditions; a mode determination unit that determines an output mode of the measurement values ​​by preferentially extracting, from the classified measurement values, the motion analysis indexes and their measurement values ​​that match the key points in a measurement section in which a change corresponding to a predetermined change condition is detected, as output targets; an output unit capable of outputting at least information relating to a measurement value that matches an analysis result of the feedback information based on the determined output mode; A driving analysis system comprising:

2. The feedback acquisition unit acquires feedback information including comments input by a user regarding the running, The driving analysis system according to claim 1, characterized in that the analysis processing unit extracts text indicating noteworthy points that the user subjectively recognizes regarding the driving condition from comments included in the feedback information by linguistic analysis as the main points.

3. The driving analysis system described in claim 1 or 2, characterized in that the analysis processing unit extracts, from the comments included in the feedback information, text containing at least one of information indicating noteworthy points subjectively recognized by the user regarding the driving condition, information regarding driving form, information regarding progress points, information regarding mental state, and information regarding comparison with other drives, as the key points through language analysis.

4. The driving analysis system according to claim 1 or 2, characterized in that the analysis processing unit stores a plurality of pre-defined keywords for extracting information indicating noteworthy points subjectively recognized by the user regarding the driving condition, such as information regarding driving form, information regarding progress points, information regarding mental state, and information regarding comparison with other drives, and extracts the key points by extracting at least one of the plurality of keywords from text included in the feedback information.

5. The driving analysis system described in claim 1 or 2, characterized in that the behavior determination unit extracts, from the classified measurement values, as the output targets in order of largest calculated value that combines a normalized value of the degree of change of the motion analysis index and its measurement value in a measurement section in which a change corresponding to a specified change condition was detected, and a normalized value of the degree of match between the key points extracted from the feedback information and the motion analysis index and its measurement value.

6. acquiring position information and movement information of a user who is a runner, the position information and movement information being measured at each progress point during a run by a predetermined measuring device; acquiring measurement values ​​of a plurality of types of motion analysis indexes that indicate the running motion state of the user for each measurement section of a predetermined unit based on the position information and the motion information that are chronologically continuous; classifying the measured values ​​of the plurality of measurement sections for each type of the motion analysis index according to a predetermined classification method for classifying the measured values ​​according to characteristics; obtaining feedback information input by a user regarding the trip; extracting key points from the feedback information based on a predetermined extraction condition; detecting a change corresponding to a predetermined change condition from the classified measured values; a step of matching the measurement section in which the change is detected among the classified measurement values ​​with the key point; extracting, as output targets, the motion analysis indexes and their measurement values ​​that match the key points in the measurement section in which the change is detected; determining an output mode of the measurement value based on the extraction result of the output object; outputting at least information related to the classified measurement values ​​based on the determined output manner; A driving analysis method comprising:

Images