JP2009247529A - Exercise evaluation apparatus and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a mechanism allowing those who carry out the training of the exercising form of sports to objectively perceive the degree of upgrade of the same. <P>SOLUTION: An acoustic processing section 14 and a sound generating section 15 emits first to third guiding sounds showing the rhythm of the exercising form of a subject from a loudspeaker 16. An acceleration sensor 11 detects the acceleration while the subject does the exercise according to the guiding sounds. The sample of the accelerating is fed to an evaluation section 22 through signal processing by an ADC 17, a delay section 18, and a filter 20. The evaluation section 22 obtains the time interval between the time obtaining the sample in the specific point of the time waveform of the acceleration and the time obtaining the sample of the moment when shooting balls and shows the displacement amount between the time interval and the exemplified time interval shown by the ideal value parameter as the evaluation result. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an exercise evaluation apparatus and program suitable for evaluating exercise such as tennis swing and swimming.

Many hours of sports training are spent learning the basic exercise form. Patent Documents 1 and 2 disclose a mechanism for supporting the acquisition of a golf swing, which is one of the sports widely used regardless of gender and age.
In the golf practice device disclosed in Patent Document 1, load cells are provided at the four corners of a swing base where the athlete takes an address, and the load cell detects a load speed at which the load center moves while the athlete performs a swing. Determine based on. When the moving speed at the center of the load becomes equal to or lower than the set value, a sound indicating that the club has reached the top of swing position is output. The golf swing correction apparatus disclosed in Patent Document 2 detects the weight applied to the heel side, the thumb side, and the little finger side of both feet of the athlete by a sensor. Then, when it is detected by the sensor of the thumb of the toe that the athlete has taken the address as the preparation posture for the swing, a rhythm sound suitable for the swing is generated.
Japanese Patent Laid-Open No. 04-307076 Japanese Patent Laid-Open No. 09-083917

By the way, in order to increase the efficiency of sports training, it is extremely important for the athletes who perform training to understand the level of progress in their skills. However, the devices disclosed in Patent Document 1 and Patent Document 2 merely guide the optimal start timing of the exercise form by sound. Therefore, there is a problem that the athlete cannot know how good the exercise form performed according to the guidance sound is, and cannot perform efficient training.
The present invention has been devised under such a background, and it is an object of the present invention to provide a mechanism that allows a person who trains a sports exercise form to objectively grasp the degree of progress. To do.

A motion evaluation apparatus according to a preferred embodiment of the present invention includes a sounding unit that sequentially generates a series of guide sounds at time intervals, and a detection unit that detects a physical quantity indicating a motion state of a part that receives the action of the force of the subject. And a time waveform feature point indicating a change in the physical quantity detected by the detection means, and an evaluation means for evaluating the feature point.
In this aspect, a series of guidance sounds is generated, feature points are extracted from a physical quantity time waveform indicating the motion state of a part that receives the effect of the subject's force, and the extracted feature points are evaluated. Therefore, it is possible to accurately evaluate the movement of the subject performed in accordance with the rhythm of the series of guidance sounds. The subject can grasp how much his / her exercise form has improved by confirming the evaluation presented by the exercise evaluation device.

(Embodiment of the Invention)
Embodiments of the present invention will be described below with reference to the drawings.
The exercise evaluation apparatus according to the present embodiment evaluates an exercise form of tennis swing. The tennis swing is a continuous motion consisting of a take-back motion that pulls the racket and a forward swing motion that swings the pulled racket in accordance with the arrival of the ball at its pocket. The exercise evaluation apparatus according to the present embodiment generates first to third guide sounds for facilitating a person who receives an evaluation of an exercise form (hereinafter referred to as “subject”) to take a swing rhythm, and The exercise form is evaluated based on the feature points extracted from the time waveform of the acceleration of the hip while the subject swings according to a series of guidance sounds. In the present embodiment, the “feature point” means a point where a change in the exercise state of the exerciser appears significantly in the time waveform of acceleration, and corresponds to a maximum value or a minimum value in the acceleration waveform.

  FIG. 1 is a diagram illustrating a configuration of an exercise evaluation apparatus 10 according to the present embodiment. As shown in FIG. 1, the motion evaluation apparatus 10 includes an acceleration sensor 11, a main control unit 12, a parameter storage memory 13, an acoustic processing unit 14, a sound generation unit 15, a speaker 16, an ADC (Analog Digital Converter) 17, A delay unit 18, an exercise start determination unit 19, a filter 20, a sample memory 21, and an evaluation unit 22 are included. Of these, each part excluding the acceleration sensor 11 is built in a housing of a size that fits in the palm of the subject, and is accommodated, for example, in a pocket of the subject's clothes. The acceleration sensor 11 is attached to the waist, which is one of the parts that are accelerated while the subject swings, and is connected to the ADC 17 inside the housing by a cable.

The acceleration sensor 11 is a uniaxial acceleration sensor. The acceleration sensor 11 outputs the acceleration applied in the first direction as a positive analog signal having a level corresponding to the magnitude, and the acceleration applied in the second direction opposite to the first direction is the magnitude of the acceleration. It is output as a negative analog signal with a level corresponding to that. In this embodiment, the acceleration sensor 11 is fixed to the subject's waist so that the direction of movement of the waist when the subject is performing a forward swing motion is the first direction.
The main control unit 12 is a device that controls the operation of each unit built in the housing.
The parameter storage memory 13 stores four types of parameters: a guidance sound parameter, a time difference parameter, an evaluation voice parameter, and an evaluation reference value parameter. The main control unit 12 sets the guidance sound parameter and the evaluation sound parameter among these four types of parameters in the acoustic processing unit 14, sets the time difference parameter in the delay unit 18 and the evaluation unit 22, and evaluates the evaluation reference value parameter. Set in section 22.

  The guide sound parameter is a parameter related to the guide sound that suggests the timing of movement to the subject, and each of the first, second, and third temporally independent guide sounds and the pronunciation of each of the guide sounds It is a parameter which shows the time interval of. Of the series of guide sounds composed of these three guide sounds, the third guide sound is a guide sound for instructing the timing of hitting the ball. It is desirable that the timbres of the first to third guidance sounds are those that draw out rhythmical movements by the subject. For example, the timbre of the third guidance sound for instructing the timing of hitting is changed to a timbre that gives a light feeling like “cha”, and the timbres of the first and second guidance sounds before that are set to “Zun”. The sound that gives such a profound feeling is pronounced as a series of guidance sounds called “Zun Dung Cha”. While listening to the first and second guidance sounds “Zun Zun”, the subject moves from the takeback motion to the forward swing motion and hits the ball at the timing of the third guidance sound “Cha”. Prompted.

The time difference parameter is a parameter indicating a time difference between the timing at which the ball is hit by the subject's swing in accordance with the first to third guidance sounds and the timing at which the third guidance sound is generated. This time difference parameter is acquired for each subject through the initial setting operation described below.
In the initial setting work, the three guide sounds indicated by the parameters are generated in order at intervals of the time indicated by the guide sound parameter, and the ball is hit according to the pronunciation of the third guide sound. A test subject is made to make a test using the apparatus 10. Then, each part of the body of the subject and the position of the racket during the trial swing are measured by an optical measuring device or the like. At this time, it is advisable to have the practice partner placed near the subject toss the ball to the pocket. Of course, there are individual differences in the sense of rhythm of each subject, and some subjects may have a sense of rhythm that makes the timing of the swing hitting to match the pronunciation of the third guidance sound earlier than that pronunciation. For example, some subjects have a sense of rhythm that would be delayed.
When the trial hitting is over, the time difference between the time when the third guidance sound was pronounced and the time when the subject actually hit the ball was calculated based on the measured values measured during that time, and the time difference was calculated for the subject. Is stored in the parameter storage memory 13 as a time difference parameter. Here, the sign of the time difference parameter of the subject whose time of hitting in the trial hit is too early than the time of pronunciation of the third guidance sound is negative, and the sign of the time difference parameter of the subject who is too late is positive.

  The evaluation voice parameter is a parameter related to the evaluation voice for notifying the subject of the contents of the evaluation result of the exercise form, and when the evaluation result of the exercise form of the subject was bad or good, it was very good. It is a parameter indicating evaluation voices to be output at times, specifically, evaluation voices of “Practice More”, “Good”, and “Excellent”.

The evaluation reference value parameter is a parameter related to the time of occurrence of the feature point in the time waveform of the acceleration of the waist between the start of the swing and the time when the racket hits the ball, and the maximum and minimum values closest to the moment of hitting This is a parameter indicating an exemplary time interval (referred to as “exemplary time interval”) between each and the moment of impact. The evaluation reference value parameter may be obtained from an actual measurement value of an acceleration time waveform obtained by causing professional tennis players to perform an exemplary swing. Further, the evaluation reference value parameter may be acquired from a theoretical value (calculated value) of a time waveform of acceleration of a model swing.
Here, in the time waveform of the acceleration of the waist in the swing of a senior player such as a professional tennis player, as shown in FIG. 2, a maximum value P1 appears a predetermined time D1 before the moment of hitting, and the moment of hitting There is a common feature that the minimum value P2 appears before the predetermined time D2. There is a rational way of accelerating the waist to efficiently convert the body force into a ball striking force, and all advanced players have learned how to accelerate such a waist. Because it is.

In FIG. 1, the acoustic processing unit 14 outputs a sound waveform audio signal of first to third guide sounds indicated by the guide sound parameter, and a sound waveform audio of the evaluation sound indicated by the evaluation sound parameter. It plays a second role of outputting a signal.
In the first role, when the sound processing unit 14 is instructed by the main control unit 12 to generate a guide sound, the sound processing unit 14 first generates a first guide sound and then generates a first guide sound. The second guidance sound is generated at a time interval t1, the second guidance sound is generated, the third guidance sound is generated at a time interval t2, and the third guidance sound is generated. The cycle of returning to the sound of the first guide sound after a time interval t3 is repeated. The time intervals t1, t2, and t3 are indicated by guide sound parameters, and may be specified based on the clock count.
In the second role, when the sound processing unit 14 gives an instruction for pronunciation of any of the evaluation sounds “Practice More”, “Good”, and “Excellent” from the main control unit 12, the sound processing unit 14 evaluates the instruction. Output audio signal.
The sound generator 15 emits the sound indicated by the audio signal output from the sound processor 14 from the speaker 16.

The ADC 17 samples the analog signal output from the acceleration sensor 11 with a sampling clock φ having a predetermined period, and outputs a digital signal indicating the sample.
The delay unit 18 delays the digital signal of the acceleration sample output from the ADC 17 by the time difference indicated by the time difference parameter when the sign of the time difference parameter set by the main control unit 12 is negative. And then output.
The exercise start determination unit 19 plays a role of determining the start of a swing by the subject. In this role, the exercise start determination unit 19 uses a threshold th corresponding to the rise of the acceleration of the waist when the subject starts swinging for determination. Specifically, each time an acceleration sample is output from the delay unit 18, the exercise start determination unit 19 obtains a moving average of accelerations of the sample and a plurality of samples preceding the sample, and the obtained moving average and threshold th Compare When the moving average exceeds the threshold th, the acceleration sample output from the delay unit 18 is output to the filter 20 through itself.

The filter 20 removes a component corresponding to noise and a component corresponding to gravitational acceleration from the acceleration sample output from the motion start determination unit 19 and outputs the result.
The sample memory 21 is a memory having a storage address enough to write a sample sequence of acceleration corresponding to time T from the start to the end of the swing. Samples output from the filter 20 are written in the memory 21 in the order of output.

The evaluation unit 22 acquires the time when a sample of acceleration at the moment of impact (referred to as “reference sample”) is acquired, and a sample of maximum and minimum values as characteristic points (referred to as “evaluated sample”). A time interval (referred to as an “evaluated time interval”) between each of the determined times is obtained, and a signal indicating a deviation amount between the evaluated time interval and the exemplary time interval indicated by the evaluation reference value parameter is output. . In the following description, the evaluated time interval corresponding to D1 in FIG. 2 is referred to as “evaluated time interval D1 ′”, and the exemplary time interval corresponding to D1 is referred to as “exemplary time interval D1 ″”. Further, the evaluated time interval corresponding to D2 in FIG. 2 is referred to as “evaluated time interval D2 ′”, and the exemplary time interval corresponding to D2 is described as “exemplary time interval D2 ″”.
In this role, the evaluation unit 22 waits for the third guide sound to be generated while synchronizing with the output cycle of the audio signal in the sound processing unit 14. When the third guide sound is generated, the evaluation unit 22 reads from the sample memory 21. After searching for the reference sample, the sample sequence preceding the reference sample in the sample memory 21 is scanned one by one, and two evaluated samples corresponding to the maximum value and the minimum value are searched. Then, the evaluation unit 22 calculates the evaluated time intervals D1 ′ and D2 ′ by performing a predetermined calculation on the number of written samples between the reference sample and each of the evaluated samples, and these evaluated time intervals D1 ′. And D2 ′ and the total amount of deviation (| D1′−D1 ″ | + | D2′−D2 ″ |) between the exemplary time intervals D1 ″ and D2 ″ as a deviation amount signal. Output to.

Here, the search of the reference sample that is the starting point for determining the evaluated time intervals D1 ′ and D2 ′ is performed by, for example, the following rules a. And b. It is good to do according to.
a. When the sign of the time difference parameter set in the evaluation unit 22 by the main control unit 12 is positive, a subject having a rhythmic feeling that the moment of hitting is slower than the pronunciation of the third guide sound is used by the subject. using. Therefore, a sample written after the time difference indicated by the time difference parameter is used as a reference sample after writing at the same time as the third guidance sound is generated.
b. When the sign of the time difference parameter set in the evaluation unit 22 by the main control unit 12 is negative, a subject having a rhythmic sense that the moment of hitting is earlier than the pronunciation of the third guide sound is used by the subject. using. However, in this case, since the sample sequence is delayed by the delay unit 18 by the time difference indicated by the time difference parameter, the acceleration sample at the moment of impact in the sample sequence that has undergone the delay is written into the sample memory 21. There is no significant time difference between the time and the time when the third guidance sound is generated. Therefore, a sample written almost simultaneously with the pronunciation of the third guidance sound is used as a reference sample.

  The main control unit 12 performs first control for setting the time difference parameter to the delay unit 18 and the evaluation unit 22, and causes the acoustic processing unit 14 to generate audio signals of the first to third guidance sounds indicated by the guidance sound parameter. And a third control for receiving the deviation amount, which is an evaluation result of the swing in accordance with the guide sounds, from the evaluation unit 22 and causing the acoustic processing unit 14 to generate an audio signal of the evaluation sound according to the deviation amount. I do.

  Among the components described above, the acoustic processing unit 14, the sound generation unit 15, and the speaker 16 constitute “a sound generation unit that sequentially generates a series of guide sounds with a time interval”, and the acceleration sensor 11 includes “ The detection unit configured to detect a physical quantity indicating a motion state of a part subjected to the force of the subject's force is configured, and the evaluation unit 22 extracts a feature point of a time waveform indicating a change in the detected physical quantity, and determines the feature point. "Evaluation means for evaluation" is configured.

Next, the operation of this embodiment will be described.
FIG. 3 is a flowchart showing the operation of the present embodiment. The series of processes shown in FIG. 3 is performed by using the acceleration sensor 11 of the motion evaluation apparatus 10 that has set the time difference parameter in the delay unit 18 and the evaluation unit 22 and the evaluation reference value parameter in the evaluation unit 22. When the start of evaluation is instructed by an operator (not shown), it is executed.

  When the start of evaluation is instructed, the main control unit 12 starts sounding the guide sound (S100). Specifically, the sound processing unit 14 is instructed to generate a guide sound. Receiving this instruction, the sound processing unit 14 repeats a cycle of outputting the audio signals of the first to third guidance sounds to the sound generation unit 15 with a time interval indicated by the parameter. The subject grasps the rhythm by listening to the first to third guidance sounds repeatedly generated from the speaker 16 for several cycles, and swings according to the pronunciation of the first to third guidance sounds of an appropriate cycle. I do.

When determining that the swing has started (S110: Yes), the exercise start determination unit 19 outputs an acceleration sample string (S120). These sample trains are subjected to extraction of feature point samples by the evaluation unit 22 after components corresponding to noise and gravitational acceleration are removed by the filter 20 (S130). Further, the evaluation unit 22 obtains evaluated time intervals D1 ′ and D2 ′ that are time intervals between the acquisition time of the sample of the feature points and the acquisition time of the acceleration sample at the moment of impact, and the evaluated time interval D1. Deviation of the total deviation amount (| D1′−D1 ″ | + | D2′−D2 ″ |) between “and D2” and the exemplary time intervals D1 ″ and D2 ″ indicated by the evaluation reference value parameter It outputs to the main control part 12 as a quantity signal (S150). And the main control part 12 which acquired deviation | shift amount produces | generates the evaluation sound according to the magnitude | size of the deviation | shift amount (S160). Specifically, when the amount of deviation is smaller than the first threshold, the main control unit 12 generates an “Excellent” evaluation sound. If the amount of deviation is larger than the first threshold value but smaller than the second threshold value, the evaluation sound of “Good” is sounded, and if larger than the second threshold value, the evaluation sound of “Practice More” is sounded.
When the evaluation sound is output in step S160, the sample in the sample memory 21 is deleted, and then the process returns to step S110 to wait for the start of the next swing. The series of processing shown in FIG. 3 ends when the end of evaluation is instructed by the operator.

In the present embodiment described above, the subject is caused to swing in accordance with the rhythms of the first to third guidance sounds, the waist acceleration is sampled, and the feature points are extracted from the sample sequence during the swing. Then, evaluated time intervals D1 ′ and D2 ′ that are time intervals between the acquisition time of the acceleration sample of the feature point and the acquisition time of the acceleration sample at the moment of impact are obtained, and the evaluated time intervals D1 ′ and D2 are obtained. Evaluation result according to the total amount of deviation (| D1′−D1 ″ | + | D2′−D2 ″ |) between the exemplary time intervals D1 ″ and D2 ″ indicated by the evaluation reference value parameter Is output. Thereby, the reliable evaluation result which shows the improvement degree of a test subject's exercise | movement form can be shown.
In the present embodiment, the time difference parameter for each subject acquired through the initial setting work is set in the delay unit 18 and the evaluation unit 22 of the motion evaluation apparatus 10. If the sign of the time difference parameter is negative, the acceleration sample output from the ADC 17 is delayed by the time difference indicated by the time difference parameter, and then written into the sample memory 21. On the other hand, when the sign of the time difference parameter is positive, such a delay is not performed, and a sample written in the sample memory 21 after the time difference parameter indicated by the time difference parameter after the sounding time of the third guidance sound is used as a reference. Using the samples, the evaluated time intervals D1 ′ and D2 ′ are obtained. That is, in the present embodiment, the sounding time of the third guidance sound is delayed or retroactive by an amount corresponding to the time difference indicated by the time difference parameter, and the sounding time of the third guidance sound is delayed or retroactively. Assume that the time intervals between the time points at which the feature point acceleration samples are acquired are evaluated time intervals D1 ′ and D2 ′. Therefore, the process required until the time intervals D1 ′ and D2 ′ to be evaluated are specified is simplified, and the time required for the subject to present the evaluation result of the exercise form after swinging can be shortened.
Further, in the present embodiment, since the first to third series of guidance sounds are taught, the motions of a plurality of subjects are constant, individual differences are minimized, and the accuracy of motion evaluation is improved. In addition, the subject can also repeatedly perform the exercise to be evaluated at a certain timing with the assistance of the first to third series of guidance sounds, so there is less variation in the exercise form when the swing is performed a plurality of times. Become.

(Other embodiments)
The present invention can be modified in various ways.
(1) In the above embodiment, not only the timbre (frequency characteristics, harmonic structure, etc.) of the first to third guide sounds, but also the pitch, volume (including rising and falling characteristics), or a combination thereof is changed. May be.
(2) In the above embodiment, the series of guidance sounds output according to the guidance sound parameter need not be three, and may be two or four or more depending on the characteristics of the sports exercise form to be evaluated. Good.
(3) In the said embodiment, the 1st-3rd guidance sound was sounded according to the guidance sound parameter. On the other hand, when the evaluation result of the acceleration sample detected by the acceleration sensor 11 from when the first guide sound is pronounced until when the second guide sound is pronounced is not good, the pitch of the third guide sound is set. By lowering, the subject is given an unpleasant impression, and the exercise form of the exercise to be performed between the first guidance sound and the second guidance sound (for example, take back exercise) is not evaluated well. Also good. Further, if the evaluation result of the acceleration sample detected by the acceleration sensor 11 from when the first guide sound is pronounced until when the second guide sound is pronounced is not good, the acoustic processing unit 14 generates the third The audio signal of the guide sound may be filtered by a low-pass filter and then supplied to the sound generation unit 15, and the timbre of the third guide sound may be changed to give an unpleasant impression. These modified examples are suitable as training tools for intermediate and advanced players who are not satisfied with the guide tone of a regular tone.

(4) In the above-described embodiment, the sample at the moment of impact is specified by using the time of pronunciation of the third guide sound among the three guide sounds that are continuously generated as the reference of the time axis in the acceleration sample sequence. However, the time of sound generation of the first guide sound and the second guide sound may be used as a reference for the time axis in the acceleration sample sequence.
(5) In the above embodiment, the total amount of deviation between the evaluated time intervals D1 ′ and D2 ′ and the exemplary time intervals D1 ″ and D2 ″ (| D1′−D1 ″ | + | D2′− In addition to D2 ″ |), a deviation amount between the magnitude of the acceleration of the feature point indicated by the sample to be evaluated and the magnitude of the acceleration prepared in advance as the evaluation reference value parameter may be obtained. Further, the obtained deviation amount may be compared with a threshold value, and an evaluation sound corresponding to the result may be generated (S160 in FIG. 3).
(6) In the above embodiment, without preparing the ideal value parameter in advance, a better evaluation result is output as the time interval between the acquisition time of the sample to be evaluated and the acquisition time of the reference sample is shorter, As the time interval is longer, a better evaluation result may be output.

(7) In the above embodiment, the evaluation result of the exercise form of the subject is evaluated in three stages, and an evaluation voice of “Excellent”, “Good”, or “Practice More” is output according to the evaluation result did. However, there are two evaluation results. If the exercise form is poor, a sound suggesting a defect such as “Boobuo” is output, and if it is good, a sound indicating a good condition such as “ping-pong” is output. Also good. Moreover, you may output the audio | voice which shows the specific content of the fault of the exercise | movement form identified based on the evaluation result like "a step is weak".
(8) In the said embodiment, the acceleration sensor 11 was mounted | worn with the test subject's waist. However, the acceleration sensor 11 may be attached to another part that receives the action of the subject's force, for example, a part or tool other than the waist of the subject's body, and a sample of acceleration applied to the part may be used for evaluating the exercise form. .
(9) In the above-described embodiment, various known motion detection sensors such as a speed sensor, a strain sensor, a pressure sensor, a geomagnetic sensor, a tilt sensor, and a sensor for detecting the muscle tension are replaced with the body of the subject instead of the acceleration sensor 11. It is also possible to mount a physical quantity indicating the motion state of these parts, and use a sample of the physical quantity to evaluate the exercise form.

(10) In the above embodiment, the exercise start determination unit 19 determines that the exercise has started when the moving average of the acceleration of the sample exceeds the threshold th. However, when the maximum value of the acceleration exceeds the threshold th, it may be determined that the exercise has started. Further, each time a time t sufficiently smaller than the time T from the start to the end of the exercise elapses, a plurality of extreme time intervals during the time t are obtained, and the obtained time intervals are preset. It may be determined that the exercise has started when it matches the pattern of the motion model.
(11) In the above embodiment, a time difference parameter unique to each of a plurality of subjects may be stored in the parameter storage memory 13 so that a subject using the apparatus 10 can select his / her time difference parameter each time. In addition, a plurality of guide sound parameters in which the number of guide sounds, timbre, sound generation interval, etc. are changed in accordance with various sports exercise forms are stored in the parameter storage memory 13, and these are changed according to the type of exercise form to be evaluated. You may make it possible to select from.
(12) In the above-described embodiment, the acceleration sample written in the sample memory 22 at the same time as the third guide sound is generated without causing the subject to make a trial hit and acquiring the time difference parameter of the subject. The time interval from the acquisition time of the sample to be evaluated may be obtained as it is as a reference sample.

(13) In the above embodiment, the present invention is applied to the evaluation of the exercise form of tennis swing. However, the exercise form that can be evaluated according to the present invention is not limited to this. The following exercise forms are expected to improve training efficiency by applying the present invention.
a. Swimming exercise form The swimming exercise form has a scratching movement that scratches water backward by limbs and a return movement that returns the scratched limbs to the original, and the exercise form of the scratching movement can earn as much propulsive force as possible. It is desirable that the movement form of the return movement should be as small as possible in the loss of propulsive force obtained by the scraping movement. Evaluation of an exercise form having such properties can be realized by the following configuration, for example. First, the first guide sound and the second guide sound for guiding the start timing of each of the scraping motion and the return motion are repeatedly output with a time interval, and the subject wearing the water pressure sensor on the forehead is allowed to swim. And the time waveform which shows the change of a test subject's acceleration is acquired based on the pressure value which the water pressure sensor detected during the swimming. Furthermore, feature points are extracted from the time waveform of the acceleration of each stroke from the scraping motion to the return motion, and a time interval between the time when the acceleration of those feature points is detected and the time when the guide sound is pronounced is prepared in advance. The deviation from the model time interval is output as the evaluation sound. More preferably, the evaluation sound indicates whether the exercise form is good or bad by changing the tone to a timbre with a low-pass filter when the amount of deviation exceeds a threshold value. Further, the speed of the subject may be calculated from the value of the water pressure sensor, and the volume and pitch of the next guidance sound to be generated may be changed according to the speed.
b. Golf Swing Exercise Form The golf driver swing exercise form is more desirable as the speed of the head at the moment the club head impacts the ball is greater. Evaluation of an exercise form having such properties can be realized by the following configuration, for example. First, a first guide sound for guiding the start timing of the backswing, a second guide sound for guiding the start timing of the subsequent downswing, and a third guide sound for guiding the timing of hitting the ball The test is performed with a time interval, and the subject who holds the club with the wind pressure sensor attached to the club head performs a driver swing. And the time waveform which shows the change of the speed of a club head is acquired based on the pressure value which the wind pressure sensor detected. Further, when the speed of the maximum value that is a feature point of the time waveform is detected, that is, when the speed of the club head is maximized, a fourth guidance sound indicating that is output. The subject can grasp whether the speed of the club head has been maximized at the moment of impacting the ball based on the time difference between the pronunciation of the third guide sound and the fourth guide sound.
c. Walking exercise form It is considered desirable to advance walking while maintaining a correct posture in the walking exercise form. Evaluation of an exercise form having such properties can be realized by the following configuration, for example. First, a position sensor is attached to each part of the subject's knee, waist, arm, etc., and a first guide sound for guiding the start timing of the exercise to advance the right foot forward and the start timing of the exercise to advance the left foot forward The second guidance sound for guiding the user is repeatedly output at intervals of time, and the subject is made to walk. Then, based on the position detected by the position sensor during the walking, a time waveform indicating changes in the posture of the subject, the inclination of the trunk, and the vertical movement is acquired. Then, feature points are extracted from the time waveform, and a deviation amount between the extracted feature points and a model feature point prepared in advance is output as an evaluation sound. The evaluation sound may be such that when the deviation amount exceeds the threshold, the pitch is fluctuated and an unpleasant impression is given to the subject.
Further, the subject may repeatedly perform the exercises such as swimming, golf swing, and walking as described above, and evaluate each of the repeated exercises. According to this aspect, correction of the form of those movements can be completed in a short time.

(14) In the above embodiment, when the sign of the time difference parameter obtained by performing the test hit of the subject is negative, the delay unit 18 uses the digital signal of the acceleration sample output from the ADC 17 as the time difference parameter. If the sign of the time difference parameter is positive while the output is delayed by the time difference shown, the evaluation unit 22 is later than the time of the third guide sound by the time difference indicated by the time difference parameter. The sample written in the sample memory 21 was used as a reference sample. However, the difference between the subject's sense of rhythm and the rhythms of the first to third guidance sounds may be compensated only by adjusting the delay amount by the delay unit 18. This modification can be realized by the following configuration, for example. First, when the sign of the time difference parameter obtained by the test strike of the subject is positive, the delay amount of the delay unit 18 is made smaller than the specified value d by the absolute value, and the sign of the time difference parameter is negative. The delay amount of the delay unit 18 is made larger than the specified value d by the absolute value. On the other hand, regardless of whether the time difference parameter is positive or negative, the evaluation unit 22 samples the acceleration written in the sample memory 21 only by a predetermined value d after the third guide sound is generated. Is identified as a reference sample.

(15) In the above embodiment, the total amount of deviation between the evaluated time intervals D1 ′ and D2 ′ and the exemplary time intervals D1 ″ and D2 ″ (| D1′−D1 ″ | + | D2′− When calculating D2 ″ |), one or both of the deviation amounts may be multiplied by an individual weighting coefficient, and then the total may be obtained. This weighting coefficient may be appropriately set according to the content of the exercise to be evaluated and the purpose of the evaluation.
(16) In the above embodiment, the maximum value and the minimum value in the acceleration waveform are characteristic points, and the time between the acquisition time of the sample of the maximum value and the minimum value and the acquisition time of the acceleration sample at the moment of impact The evaluated time intervals D1 ′ and D2 ′, which are intervals, were obtained. However, not only the maximum and minimum values, but also a stop point where the acceleration waveform switches from an upward convex waveform to a downward convex waveform, a stop point where the downward convex waveform switches to an upward convex waveform, etc. Evaluation points may be made based on the time interval between the sample acquisition time corresponding to them and the sample acquisition time at the moment of impact. The stopping point at which the acceleration waveform switches from an upward convex waveform to a downward convex waveform, the stationary point at which the downward convex waveform changes to an upward convex waveform, and other inflection points are the first floor of acceleration. It is possible to detect by using a differential value or a higher-order differential value. Further, an integrated value (waveform area) of the acceleration waveform may be extracted as a feature point.
(17) A program for causing a computer to realize the same function as each part of the exercise evaluation device 10 according to the above-described embodiment is transmitted from a server device on the WWW (World Wide Web) to a personal computer, a PDA (Personal Data Assistance), a mobile phone. It may be downloaded to the terminal. In this case, it is possible to control the sensors, sound sources, speakers, and the like that are provided as standard in those terminals by the downloaded program, thereby realizing the same operation as the motion evaluation apparatus 10. Further, such a program may be distributed after being stored in a storage medium.

It is a block diagram which shows the structure of the exercise | movement evaluation apparatus which is embodiment of this invention. It is a figure which shows the time waveform of the acceleration of the waist during a forward swing exercise | movement. It is a flowchart which shows the content of the process of the embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Motion evaluation apparatus, 11 ... Acceleration sensor, 12 ... Main control part, 13 ... Memory for parameter storage, 14 ... Sound processing part, 15 ... Sound generation part, 16 ... Speaker, 17 ... ADC, 18 ... Delay part, 19 ... exercise start determination unit, 20 ... filter, 21 ... sample memory, 22 ... evaluation unit

Claims (7)

A pronunciation means for generating a series of guide sounds in order at intervals, and
A detecting means for detecting a physical quantity indicating a motion state of a portion that receives the action of the force of the subject;
A motion evaluation apparatus comprising: an evaluation unit that extracts a feature point of a time waveform indicating a change in physical quantity detected by the detection unit and evaluates the feature point. The pronunciation means is:
The motion evaluation apparatus according to claim 1, further generating a sound indicating a result of the evaluation by the evaluation unit. The evaluation means includes
A time interval between a sounding time when the sounding means pronounces one of the series of guide sounds and a time when the detecting means detects a physical quantity of the feature point is obtained, and the time interval and the evaluation reference The movement evaluation apparatus according to claim 1, wherein an amount of deviation from a time interval is calculated. The evaluation means includes
The pronunciation time when the sound generation means pronounces one of the series of guide sounds is delayed or retroactive by an amount corresponding to the difference between the rhythm of the series of guide sounds and the rhythm of the subject's movement, and the pronunciation time A time interval between the time delayed or moved back and the time when the detection means detects the physical quantity of the feature point, and the amount of deviation between this time interval and the time interval used as a reference for evaluation is obtained. The motion evaluation apparatus according to claim 1, wherein: And further comprising guide sound change control means for changing a pitch, tone color, volume, or a combination of the guide sounds that the sound generation means subsequently generates according to the deviation amount obtained by the evaluation means. The exercise | movement evaluation apparatus of any one of Claim 2 to 4. A determination unit that determines whether the moving average of the physical quantity satisfies a predetermined condition each time the detection unit detects the physical quantity;
The evaluation means includes
6. The feature point is extracted from a time waveform indicating a change in physical quantity detected by the detection means after the determination means determines that the moving average of the physical quantities satisfies a predetermined condition. The exercise evaluation apparatus according to any one of the above. A sound generation function that produces a series of guidance sounds in sequence at intervals in the computer,
A detection function for detecting a physical quantity indicating a motion state of a part subjected to the action of the force of the subject;
A program for realizing an evaluation function for extracting a feature point of a time waveform indicating a change in a physical quantity detected by the detection function and evaluating the feature point.

Images