JP2002233517A - Device and method for evaluating beauty of walking - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable high speed processing by simple technique and to give a guide for approaching to ideal walking by feature amount easy to be intuitively understood. SOLUTION: In a foot pressure measuring part 11, the distribution of foot pressure during walking of a subject is continuously measured by a pressure sensor, and in a beauty mark giving part 13, the beauty of the subject's walking is given marks according to the walking feature of the subject obtained from the measurement result and the ideal walking feature normalized on the basis of a foot rectangular area. The result of comparison between the walking feature of the subject and the ideal walking feature is displayed by a comparison display part 14.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a walking beauty evaluation device, and more particularly to a walking device for measuring the foot pressure of a subject during walking using a pressure sensor and evaluating the beauty of walking using obtained data. The present invention relates to a beauty evaluation device.

[0002]

2. Description of the Related Art Hitherto, research for objectively evaluating the beauty and health of a walking object such as a person has been advanced.
As a method of objectively evaluating the beauty and the degree of health of walking, there is a method of visually judging. However, in the method of judging the beauty and health of the walking of the moving object by visual observation of the judge, it is difficult to make an objective and constant judgment based on the subjective judgment of the judge. For example, factors unrelated to walking itself, such as the hobby of the judge for the body shape of the subject, may affect the judgment result.

For this reason, conventionally, as a method of objectively measuring the posture of a moving object, the position and posture of the moving object are estimated from the contact pressure of the moving object, and the information is compared with a camera image. In order to measure the posture of a moving object by taking the following method, for example, Japanese Patent Laid-Open No. 11-235328
No., etc.). Further, as a motion analysis device for diagnosing the motion state of a subject by parameterization, a device for analyzing motion analysis data of a subject by using a neural network technique has been proposed (for example, Japanese Patent Laid-Open No.
30505023). Furthermore, as a gait analysis method for quantifying gait characteristics from the foot pressure distribution of the subject,
A device that measures a foot pressure distribution during walking, analyzes a walking motion from the measurement result, and outputs the characteristic as a parameter has been proposed (for example, Japanese Patent Application Laid-Open No. H11-113884).
No., etc.).

[0004]

However, such a conventional walking beauty evaluation apparatus is related to the measurement of the operation state, and either outputs the measurement result itself from the sensor or sets the characteristic parameter. However, there is a problem that the beauty and health of the walking of the moving object cannot be quantified and objectively determined. The present invention is intended to solve such a problem, and provides a guideline for approaching an ideal gait using a feature amount that can be processed at high speed with a simple method and that is easy to understand intuitively. It is an object of the present invention to provide a walking beauty evaluation device that can perform walking.

[0005]

In order to achieve the above object, a walking beauty evaluation apparatus according to the present invention uses a pressure sensor to continuously and temporally distribute the foot pressure distribution during walking of a subject. Pressure measurement means for measuring the foot pressure, foot rectangular area extraction means for extracting the subject's foot rectangular area from the measurement results of the foot pressure measurement means, and the walking characteristics of the subject himself obtained from the measurement results of the foot pressure measurement means And a beauty scoring means for scoring the beauty of the subject's walking using the ideal walking feature normalized based on the foot rectangular area extracted by the foot rectangular area extracting means; A comparison display unit for comparing and displaying the characteristic and the ideal walking characteristic.

[0006] The beauty scoring means includes: a model database in which model parameters of the foot pressure center trajectory are recorded in advance; An ideal parameter generation unit that generates an ideal foot pressure center trajectory by normalizing according to the subject and outputs it as an ideal parameter, and obtains the actual foot pressure center trajectory of the subject from the measurement result of the foot pressure measurement unit A parameter comparison unit that outputs a personal parameter, and a parameter comparison that scores the beauty of the subject's walking based on a comparison between the ideal parameter generated by the ideal parameter generation unit and the parameter generated by the personal parameter generation unit. And a unit.

When generating ideal parameters, the ideal parameter generating unit spatially normalizes the model parameters by enlarging / reducing the model parameters so as to coincide with the rectangle inscribed in the subject's footprint, and at the same time, walks the subject. It is also possible to normalize temporally by interpolating and obtaining the center of the foot pressure of the example in accordance with each measurement timing from landing to kicking in, and to generate ideal parameters from the example parameters by these normalizations.

When the beauty of the walking of the subject is scored,
The parameter comparison unit calculates the distance between the ideal center of foot pressure and the center of the subject's foot pressure at each measurement timing from landing to kicking out, and sums the distances at each measurement timing to improve the beauty of the subject's walking. May be scored. Also, the comparison display means may display the ideal foot pressure center locus and the actual foot pressure center locus obtained by the measurement so as to overlap the footprint.

Further, in the method for evaluating beauty of walking according to the present invention, a foot pressure distribution during walking of a subject is continuously measured using a pressure sensor over time, and from the measurement result, the foot rectangular area of the subject is measured. Is extracted, and the beauty of the subject's walking is scored using the walking characteristics of the subject himself and the ideal walking features normalized on the basis of the rectangular region of the foot. This is to compare and display a typical walking feature.

When scoring the beauty of walking, ideally, the model parameters of the center trajectory of the foot pressure stored in advance in the model database are normalized according to the subject based on the rectangular region of the foot. Generating a typical foot pressure center trajectory and outputting it as an ideal parameter, obtaining the actual foot pressure center trajectory of the subject himself from the measurement results and outputting it as a subject parameter, based on a comparison between the ideal parameter and the subject parameter, Beauty may be scored.

When generating ideal parameters, the model parameters are spatially normalized by enlarging / reducing the model parameters so as to match the rectangle inscribed in the subject's footprint, and kicking from the landing when the subject walks. The center of the foot pressure of the sample may be interpolated and obtained in accordance with each measurement timing up to the measurement, thereby normalizing the model temporally, and the normalization may be used to generate ideal parameters from the model parameters.

Further, the distance between the ideal center of foot pressure and the center of foot pressure of the subject is determined at each measurement timing from landing to kicking out, and the distance at each measurement timing is summed up, so that the beauty of the walking of the subject is obtained. May be scored. Further, the ideal foot pressure center locus and the actual foot pressure center locus may be displayed so as to be superimposed on the last.

[0013]

Next, an embodiment of the present invention will be described.
This will be described with reference to the drawings. FIG. 1 shows an embodiment of the present invention.
It is a block diagram of the beauty evaluation apparatus of the walk which concerns on a state.
The walking beauty evaluation device includes a foot pressure measuring unit 11 and a foot square.
Shape area extraction unit 12, beauty score conversion unit 13, and comparison table
An indicator 14 is provided. The foot pressure measurement unit 11
Time distribution of the foot pressure distribution of a walking subject using a sensor.
Time series foot pressure image sequence that continuously measured and visualized foot pressure distribution
(X₁, X_Two, ..., x _T) Appeared in beauty scoring section 13
Power. x_TIs observed at time t (measurement timing).
This is the measured feature vector.

FIG. 2 shows a specific example of the foot pressure measuring unit 11.
You. The foot pressure measurement unit 11 is a pressure sensor for measuring foot pressure.
Image of the foot pressure detected by the pressure sensor 21
And a computer 22 for processing. pressure
The sensor 21 can measure foot pressure for several consecutive steps.
Wear. The subject walks naturally on the pressure sensor 21.
It takes about 3 to 5 steps. The pressure sensor 21
The detected foot pressure is taken into the computer 22,
When the intensity of pressure at each measurement timing is used as a gradation value
Sequence foot pressure image sequence (x₁, X_Two, ..., x _T), Eg 3 per second
0 foot pressure image frames (hereinafter referred to as frames)
To the foot rectangular area extraction unit 12 and the beauty score conversion unit 13
Is output.

In the foot rectangular area extracting section 12, the foot pressure measuring section 1
The footprint image is obtained by superimposing all the frames of the time-series foot pressure image sequence (x ₁ , x ₂ ,..., X _T ) from ₁ .
Then, from this footprint image, foot rectangle area information {p ₁ , indicating the lengths of the long axis and the short axis of the rectangle inscribed in the sole portion,
Extract p ₂ }.

The beauty scoring unit 13 includes a foot pressure measuring unit 11
Based on the time series foot pressure image sequence (x ₁ , x ₂ ,..., X _T ) from the user and the foot rectangle area information {p ₁ , p ₂ } from the foot rectangle area extraction unit 12. Is scored. FIG. 3 shows a configuration example of the beauty scoring unit 13. The beauty scoring unit 13 includes a personal parameter generation unit 131, an ideal parameter generation unit 132, an example database 133, and a parameter comparison unit 134.

The personal parameter generation unit 131 converts the time-series foot pressure image sequence (x ₁ , x ₂ ,..., X _T ) into personal parameters (r ₁ , r ₂ ,.
r _T ) and outputs the result to the parameter comparison unit 134.
Here, the foot pressure center coordinates (C _x (t), C _{y at time} t)
(T)) and the principal parameters (r ₁ , r ₂ ,..., R _T )
Is the coordinates (x, y) of the foot pressure image frame at time t
Is defined as P (x, y, t) when the pixel value (foot pressure) at is P (x, y, t).

[0018]

(Equation 1)

In the model database 133, the model parameters (i ₁ , i ₂ ,..., I _{T ′} ) indicating the time change of the coordinates of the trajectory of the center of the foot pressure of the teacher serving as the walking model are stored in the user parameters (r ₁ , r ₂ ,..., r _T ). This model parameter is converted into a numerical value per unit size of the leg rectangular area. The ideal parameter generation unit 132 uses the foot rectangle area information {p ₁ , p ₂ } from the foot rectangle area extraction unit 12 to convert the model parameters (i ₁ , i ₂ ,..., I _{T ′} ) into the foot of the subject. Normalized according to the type, and output as ideal parameters (i ′ ₁ , i ′ ₂ ,..., I ′ _T ).

These personal parameters (r ₁ , r ₂ ,..., R
_T ) and ideal parameters (i ′ ₁ , i ′ ₂ ,..., I ′ _T ) are composed of foot pressure center trajectories, and have the same order. Here, the principal parameters (r ₁ , r ₂ ,..., R _T ) are actual parameters of the principal obtained from the measurement, whereas ideal parameters (i ′ ₁ , i ′ _{2 ,.} Is obtained by normalizing the model parameters (i ₁ , i ₂ ,..., I _{T ′} ) of the teacher who becomes the walking model according to the last of the subject.

FIG. 4 shows that the ideal parameter generator 132
7 shows a normalization process of a foot pressure center locus. Ideal parameter raw
In the forming unit 132, the subject's foot rectangular area information {p₁, P_Two｝
Based on the model parameters (i₁, I_Two, ..., i_{T '})
In the long axis direction of the inscribed rectangle of the sole ₁Double and sole
P in the short axis direction of the inscribed rectangle_TwoDouble, heel (landing) and second finger (kicks)
By combining the reference points A and B provided in
Perform spatial normalization. Also, from landing to kicking out
In order to represent by a predetermined number T (for example, 20) frames,
Example parameters (i₁, I_Two, ..., i_{T '}) T '
Ideal parameters from foot pressure center coordinates for each frame
(I '₁, I '_Two, ..., i '_T) Feet for each frame in
By interpolating the pressure center coordinates, temporal normalization can be performed.
Do.

In the parameter comparing section 134, the ideal parameters (i ′ ₁ , i ′ ₂ ,..., I ′ _T ) obtained by the normalization in the ideal parameter generating section 132 in this way, and the person obtained by the measurement are obtained. The parameters (r ₁ , r ₂ ,..., R _T ) are compared, and a score y (scalar value) indicating the beauty of walking is calculated and output based on the distance between these foot pressure center trajectories.
FIG. 5 shows the scoring process in the parameter comparison unit 134.
In the parameter comparison unit 134, the ideal parameters (i ′ ₁ ,
i ′ ₂ ,..., i ′ _T ) and the principal parameters (r ₁ , r ₂ ,.
_{From T} ), the distance between the foot pressure center trajectories is calculated.

In FIG. 5, (X _t , Y _t ) indicates an ideal barycenter position in the t-th frame indicated by ideal parameters (i ′ ₁ , i ′ ₂ ,..., I ′ _T ). (X _t , y _t ) indicates the actual position of the center of gravity in the t-th frame indicated by the principal parameters (r ₁ , r ₂ ,..., R _T ). At this time, the distance L between the ideal foot pressure center locus and the actual foot pressure center locus is obtained as follows. The smaller the distance L, the closer to the ideal walking and the more beautiful the walking.

[0024]

(Equation 2)

FIG. 6 is a graph showing the relationship between the distance L obtained by the above equation 2 and the subjective evaluation by the teacher. The teacher's evaluation is based solely on the foot pressure, and the scoring is given on a 5-point scale (5 for an ideal beautiful walk). As data, six beginners of walking took lessons at the walking school for half a year,
The data measured every month and the data measured by the teacher's walking were used.

From this graph, it can be seen that the closer the distance L between the ideal foot pressure center locus and the actual is, the higher the beauty score. The results of comparing and displaying the change in foot pressure of a certain student whose score has increased from the first, second, and third times are shown as data 1 (first time), data 2 (second time), and data 3 (third time) in FIG. Was. According to this, it is understood that the foot pressure gradually approaches the ideal foot pressure center locus. In this way, by superimposing and displaying the ideal foot pressure center locus and the actual foot pressure center locus, it is possible to know the parts of the feet to be used when moving the center of gravity, and also to know the timing of using them.

The score of the beauty of walking can be obtained from the regression equation obtained from the data in FIG. Now, let L be the distance between the ideal foot pressure center trajectory and the actual locus, and let P be the score of the subjective evaluation by the teacher. Here, when the data plotted in FIG. 6 is applied to the linear regression equation P = A × L + B, the coefficients A and B are A = 0125 and B = 5.5, respectively. Therefore, by calculating L as the distance between the ideal foot pressure center locus and the actual from the measured foot pressure, the score y of the beauty of walking can be obtained by the following equation. In addition, the following Expression 3 derives the expression of the score y based on the data in FIG. 6, and is not limited to this.

[0028]

(Equation 3)

As described above, the foot pressure measuring unit 11 continuously measures the foot pressure distribution during walking of the subject using the pressure sensor, and, based on the measurement result, determines the rectangular region of the subject's foot in the foot. The beauty scoring unit 13 extracts the subject's walking using the walking characteristics of the subject himself extracted by the rectangular region extracting unit 12 and the ideal walking features normalized on the basis of the leg rectangular region. Since the beauty is scored and the comparison result between the subject's own walking characteristics and the ideal walking characteristics is displayed on the comparison display unit 14, high-speed processing can be performed by a simple method. In addition, since the foot pressure center locus is used as the feature quantity, the order is low, so that it is easy to handle and intuitively understandable. Further, since the ideal and actual foot pressure center trajectories are displayed for comparison, it is possible to provide a guideline for approaching ideal walking.

Prior to the present invention, the present inventors have proposed a walking judgment apparatus for quantifying and outputting the beauty and health of walking by using foot pressure data in the same manner as in the present invention (Japanese Patent Application 2000). -035061) is different from the present invention in the following four points. First, the output of the device is different. The walking judgment device is a device for quantifying beauty or health, but the present invention specializes in beauty and evaluates it with scores. Second, the criteria are different. In the above walking judgment device, the beauty criterion is determined by using the dynamic rationality such as balance, symmetry, and temporal continuity as a criterion, and comparing the feature value with a threshold value stored in the kinetic rationality judgment criterion unit. Compare their taste and health. The present invention evaluates beauty by comparing it with the parameters of a human model as an example.

Third, the difference in the criterion causes a difference in the feature amount, its display method, and the processing speed. In the above-mentioned walking judgment device, dynamic rationality was used as a judgment standard. Therefore, it is necessary to absolutely normalize the feature amount, and for the foot pressure feature amount, it is necessary to provide a sole portion dividing unit and perform absolute spatial normalization of the sole position. In addition, the change over time of the load for each part used as the foot pressure feature amount is
It was difficult to understand intuitively and the amount of calculation was large. In the present invention, the normalization of the feature amount is performed relatively between the individual and the model. For this reason, a plantar part division is not required. Also,
The feature amount uses a foot pressure center locus which can be displayed so as to be superimposed on the footprint, and is easy to understand intuitively.

Therefore, according to the present invention, as compared with the above-described walking determination device, there are advantages that the characteristic amount is easy to understand, the configuration of the device is simple, and the processing speed is high. In the walking determination device, what is finally displayed is only the determined parameters of the subject.
In the present invention, the guideline of how to approach the ideal walking can be provided by displaying the centerline of the person's foot pressure and the centerline of the model's foot pressure by superimposing and displaying the centerline of the model. .

Fourth, the configuration of the device is different due to the different judgment criteria. In the above-described walking determination device, since balance, symmetry, and temporal continuity are used as determination criteria, a plurality of pieces of sensor information are extracted and processed using a pressure sensor or a camera. In the present invention, since the determination criterion is a comparison with the model, only the pressure sensor is used as the sensor, and only the foot pressure center locus in which the model data is clarified is extracted.

Although the walking judgment device makes it possible to score the beauty of walking and the degree of health, it is necessary to cut out the sole portion and the degree of the feature quantity is large, so that the device is complicated. Processing was heavy. In addition, it was difficult for the subject to obtain information for correcting the walking of the white body of the subject. The present invention is intended to solve such a problem, and provides a guideline for approaching an ideal gait using a feature amount that can be processed at high speed with a simple method and that is easy to understand intuitively. It is an object of the present invention to provide a walking beauty evaluation device that can perform walking.

[0035]

As described above, according to the present invention,
The foot rectangular area extraction unit extracts the foot type information of the subject individual,
Since the ideal parameter generator generates a score based on the distance between the ideal feature generated according to the footprint of the subject and the actual feature, a simple method of comparing the distance between features is adopted. The configuration is simplified, and high-speed processing becomes possible. Also, since the foot pressure center locus is used as the feature value, it is easy to intuitively understand. Further, by providing the comparison display section, the ideal and actual foot pressure center trajectories can be displayed so as to be superimposed on the subject's footprint, and a guideline for approaching ideal walking can be given.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a walking beauty evaluation device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a foot pressure measuring unit.

FIG. 3 is a configuration diagram showing a beauty scoring unit.

FIG. 4 is an explanatory diagram illustrating normalization processing in an ideal parameter generation unit.

FIG. 5 is an explanatory diagram showing a scoring process in a parameter comparing unit.

FIG. 6 is a graph showing a relationship between an ideal distance of a foot pressure center locus and an evaluation point.

FIG. 7 is an explanatory diagram showing a change in a foot pressure center locus due to a lesson.

[Explanation of symbols]

11: Foot pressure measurement unit, 12: Foot rectangular area extraction unit, 13: Beauty score conversion unit, 14: Comparison display unit, 21: Pressure sensor,
22: computer, 131: principal parameter generation unit,
132: ideal parameter generation unit; 133: sample database; 134: parameter comparison unit.

Continuation of the front page (72) Inventor Sakuichi Otsuka 2-3-1 Otemachi, Chiyoda-ku, Tokyo F-term in Nippon Telegraph and Telephone Corporation (reference) 4C038 VA11 VB14 VC20 5L096 EA13 FA18 FA69 HA07 JA11

Claims (10)

[Claims] 1. A walking beauty evaluation apparatus for objectively evaluating the beauty of walking, comprising: a foot pressure measuring means for continuously and temporally measuring a foot pressure distribution during walking of a subject using a pressure sensor; A foot rectangular area extracting means for extracting the subject's foot rectangular area from the measurement result of the foot pressure measuring means; a walking characteristic of the subject himself / herself obtained from the measurement result by the foot pressure measuring means; and the foot rectangular area extracting means Using the ideal walking feature normalized based on the leg rectangular area extracted in
A beauty evaluation device for walking, comprising: beauty scoring means for scoring the beauty of a subject's walking; and comparison display means for comparing and displaying the walking characteristics of the subject and ideal walking characteristics. . 2. The beauty evaluation device according to claim 1, wherein the beauty scoring means includes: a model database for preliminarily recording model parameters of a foot pressure center trajectory; An ideal parameter generation unit that generates an ideal foot pressure center locus by normalizing the example parameters of the example database according to the subject based on the foot rectangular area, and outputs the generated locus as an ideal parameter; An actual parameter of the subject's actual foot pressure center trajectory obtained from the measurement result of the means, and an actual parameter generation unit for outputting as a personal parameter; an ideal parameter generated by the ideal parameter generation unit; and an actual parameter generated by the personal parameter generation unit And a parameter comparison unit that scores the beauty of the subject's walking based on the comparison with Walk of beauty evaluation device to. 3. The walking beauty evaluation apparatus according to claim 2, wherein the ideal parameter generation unit expands the model parameter so as to match a rectangle inscribed in the subject's last.
In addition to spatially normalizing the image by reducing it, it also normalizes temporally by interpolating the center of foot pressure of the model parameter in accordance with each measurement timing from landing to kicking out of the subject during walking, and these normalizations are performed. And generating the ideal parameters from the model parameters using the walking parameters. 4. The walking beauty evaluation device according to claim 2, wherein the parameter comparison unit calculates an ideal foot pressure center and a subject's foot pressure center at each measurement timing from landing to kicking. A walking beauty evaluation apparatus characterized in that the distance of each measurement timing is obtained, and the distance of each measurement timing is summed up, so that the beauty of the walking of the subject is scored. 5. The walking beauty evaluation device according to claim 2, wherein the comparison and display means superimposes an ideal foot pressure center locus and an actual foot pressure center locus obtained by measurement on a footprint. The walking beauty evaluation device characterized by displaying. 6. A walking beauty evaluation method for objectively evaluating the beauty of walking, wherein a pressure sensor is used to continuously and temporally measure a foot pressure distribution during walking of a subject using a pressure sensor. Extract the subject's foot rectangular area, and score the beauty of the subject's walking using the walking characteristics of the subject himself obtained from the measurement result and the ideal walking characteristics normalized based on the foot rectangular area. A walking beauty evaluation method that compares and displays the subject's own walking characteristics and ideal walking characteristics. 7. The beauty evaluation method according to claim 6, wherein when the beauty of walking is scored, a model of a foot pressure center locus stored in advance in a model database based on the foot rectangular area. The parameters are normalized according to the subject to generate an ideal foot pressure center locus by outputting the ideal foot pressure center locus, and output the actual foot pressure center locus of the subject himself from the measurement result, and output the same as a personal parameter, A beauty evaluation method for walking, wherein the beauty of walking of the subject is scored based on a comparison between the ideal parameter and the personal parameter. 8. The walking beauty evaluation method according to claim 7, wherein, when the ideal parameters are generated, the model parameters are enlarged or reduced so as to match a rectangle inscribed in the subject's last. In addition to spatial normalization, the center of foot pressure of the model is interpolated and obtained in accordance with each measurement timing from the landing to kicking out of the subject during walking. A walking beauty evaluation method, wherein the ideal parameter is generated from a parameter. 9. The walking beauty evaluation method according to claim 7 or 8, wherein a distance between an ideal center of foot pressure and a center of foot pressure of the subject is obtained at each measurement timing from landing to kicking out. A beauty evaluation method for walking, characterized in that the walking beauty of the subject is scored by summing the distances of the subjects. 10. The walking beauty evaluation method according to claim 7, wherein an ideal foot pressure center trajectory and an actual foot pressure center trajectory are superimposed and displayed on a footprint. Evaluation method.