JP3298793B2 - Walking pattern processing method and apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a walking pattern processing method and apparatus, and more particularly to a walking pattern processing method and apparatus for measuring a human walking motion, analyzing the measurement result, and displaying the result as a parameter. .

[0002]

2. Description of the Related Art Conventionally, as means for analyzing and processing a walking pattern, there have been a floor reaction force meter and a foot size pressure distribution sensor.

[0003]

Among them, a floor reaction force meter can measure a change in force at the time of walking in a time series, but cannot measure a pressure distribution in a foot contact area, and thus cannot measure at a walking time. I couldn't measure where my feet were.

In the case of a foot-size pressure distribution sensor, a pressure distribution sensor of about the size of a foot is laid in shoes to measure the pressure distribution inside the last and to ground the foot.
Although it was possible to measure the time of takeoff, it was not possible to measure the position of the foot when walking.

As a method of measuring the position where the foot is attached, a very complicated measurement method such as a method of applying ink to the foot and walking on paper, a method of taking a video with a scale and measuring while reproducing the video, etc. There was no other way.

As a conventional method of displaying the analysis result of a walking pattern, there has been a method of directly displaying each parameter value or a statistical display. However, a large amount of data has been obtained due to insufficient measurement means. It was difficult to get. Therefore, a display method for displaying a large amount of data so as to be easily grasped has not been developed. As a conventional general method of displaying walking parameters, for example, FIG.
There are display methods as shown in FIG. 13 and FIG. 13, but these are diagrams for conceptual explanation, and there is no means for realizing a display method corresponding to these drawings for actual measurement / analysis results.

As described above, in the conventional analysis method,
There is a problem that it is difficult to simultaneously and easily obtain parameters relating to the position of the foot and the time. In addition, regarding the display of measurement results, it is difficult to grasp intuitively when a walking motion, which is a time-series phenomenon performed in a three-dimensional space, is described on a two-dimensional paper surface with a small number of parameters and a simple diagram. There was a problem that understanding required skill.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and an object of the present invention is to automatically and stably and easily put a large burden on a subject in the spatial and temporal parameters of a walking motion. It is an object of the present invention to provide a gait pattern processing method and a gait pattern processing method capable of obtaining the gait pattern without any inconvenience and facilitating intuitive grasp and quantitative comparison of spatial and temporal factors of the walking motion. .

[0009]

According to a first aspect of the present invention, there is provided a pressure sensor for acquiring a two-dimensional pressure distribution based on walking, and a method for acquiring an output of the pressure sensor as time-series data at fixed time intervals. Series pressure distribution image detection means, superimposition image creation means for creating a superimposition image by superimposing the time series pressure distribution image in the time direction, and a foot pressure mass area cutout for extracting a plurality of foot pressure mass areas from the superimposition image Means, step correspondence detecting means for associating the time series pressure distribution image with each of the foot pressure mass regions, and a parameter for detecting a parameter representing the characteristic of the walking based on the correspondence of the step corresponding means. A walking pattern processing apparatus comprising: a detection unit; and an output unit that outputs the parameter.

According to the present invention, the walking parameters are obtained from the two-dimensional pressure distribution obtained in time series and the superimposed image thereof, so that the basic walking parameters can be obtained automatically and easily. In addition, the foot pressure block pattern on the superimposed image enables intuitive grasp of the walking pattern,
It becomes easy to understand, compare, and record walking patterns that required skill in the past.

According to a second aspect of the present invention, the parameter detecting means comprises a characteristic position detecting means for detecting a predetermined characteristic position from the superimposed image based on the correspondence of the step correspondence means.

According to a third aspect of the present invention, the predetermined characteristic position is a rearmost portion, a frontmost portion, a center, or a center of gravity of each foot pressure mass in the walking direction.

According to a fourth aspect of the present invention, the parameter detecting means detects an initial frame having a pressure value for each foot pressure mass region based on the correspondence of the step correspondence means; Based on the correspondence of the step correspondence means, a final frame detecting means for detecting a frame having the last pressure value for each foot pressure mass region, and a frame detected by the initial frame detecting means and the final frame detecting means. And time interval detecting means for detecting a time parameter based on the time information.

According to a fifth aspect of the present invention, the time parameters are a stride time, a step time, a simultaneous fixing time, and a free leg time.

According to a sixth aspect of the present invention, the parameter detecting means is a center-of-gravity position detecting means for obtaining a center of gravity of a pressure value for each pressure distribution image based on the correspondence of the step correspondence means.

According to a seventh aspect of the present invention, the center-of-gravity position detecting means determines the center of gravity for both feet.

According to an eighth aspect of the present invention, the center-of-gravity position detecting means obtains the center of gravity for each foot.

According to a ninth aspect of the present invention, the parameter detecting means detects the position of the center of gravity of the pressure value for each pressure distribution image detected by the center of gravity position detecting means.
It further comprises a moving direction detecting means for detecting a moving direction of the center of gravity in time.

According to a tenth aspect of the present invention, in the extraction of each foot pressure mass, the foot pressure mass region extracting means enlarges a region having a pixel value by a predetermined number of pixels, and removes an overlapping region when the enlarged. It is regarded as footprint areas of the same step.

According to this aspect, each step can be accurately extracted from the superimposed image of a plurality of steps.

According to an eleventh aspect of the present invention, a two-dimensional pressure distribution based on walking is acquired as a time-series pressure distribution image at fixed time intervals, and the time-series pressure distribution image is superimposed in a time direction to form a superimposed image. Create and extract a plurality of foot pressure mass regions from the superimposed image, associate each foot pressure mass region with the time-series pressure distribution image, and represent the characteristics of the walking based on the association. A walking pattern processing method characterized by detecting a parameter and displaying or printing the parameter.

[0022] According to a twelfth aspect of the present invention, as the parameter, the last part, the most extreme part, and the center of the foot pressure mass region in each step are detected, and the stride length and the step length are set in parallel with the superimposed image. The length information is displayed or printed, and the step information is displayed or printed on the superimposed image.

According to this aspect, the magnitude of each parameter of the stride length, the step length, and the step can be grasped as a whole.

According to a thirteenth aspect of the present invention, as the parameter, the center of the pressure value for each pressure distribution image is detected, and the position of the center of the pressure value for each pressure distribution image is determined in parallel with the superimposed image. Display or print.

According to this aspect, since the movement of the position of the center of gravity is displayed or printed, a dynamic speed change can be easily grasped.

According to a fourteenth aspect of the present invention, when the center of gravity of the pressure value is detected, the moving direction of the position is detected, and when the center of gravity moves in a direction opposite to the walking direction, the display or printing is performed. Is to change the method.

According to this aspect, it is possible to easily find an abnormality in the movement of the center of gravity.

According to a fifteenth aspect of the present invention, when the center of gravity moves in the direction opposite to the walking direction, the color of display or printing is changed.

According to a sixteenth aspect of the present invention, the center of gravity is detected for each foot and its position is displayed or printed.

According to a seventeenth aspect of the present invention, the center of gravity is detected for both feet and the position is displayed or printed.

According to the present invention, the center of gravity of a pressure distribution image having a pressure value is first detected in each step as the parameter, and the position of the center of gravity is displayed or printed on the superimposed image. It is to do.

According to this aspect, it is possible to easily find an abnormal walking in which the foot comes into contact with the toe rather than the heel when the foot starts to stick.

According to a nineteenth aspect of the present invention, a stride time, a step time, a simultaneous fixing time, and a free leg time are detected as the parameters, and for each of the two feet, the contact time is represented by a line segment along the time axis. Display or print.

According to this aspect, it is easy to grasp the time parameter.

According to a twentieth aspect of the present invention, the sum or average of the pressure values at each time point is displayed or printed in different colors or densities, superimposed on the line segment.

According to this aspect, it is easy to understand the relationship between the temporal pressure change and the walking motion.

According to a twenty-first aspect of the present invention, a pressure level scale is added to the line segment.

According to a twenty-second aspect of the present invention, the stride time, the step time, the simultaneous fixing time, and the free leg time are detected based on the time of the pressure distribution image having the first and last pressure values in each step. That is.

According to a twenty-third aspect of the present invention, a stride length, a step length, a step, a stride time, a step time, a free leg time, and a simultaneous fixing time are detected as the parameters, and the left and right feet are compared. Stride length, step length, step, stride time, step time, swing time,
The simultaneous fixing time is displayed or printed on a scale.

According to this aspect, it is easy to grasp the degree of the left-right difference that is important in abnormal walking, and at the same time, it is easy to determine whether the left-right difference is significant by comparing the magnitude of the left-right difference. it can.

According to a twenty-fourth aspect of the present invention, a plurality of data of the stride length, the step length, the step, the stride time, the step time, the free leg time, and the simultaneous fixing time are printed or displayed. Average value,
In other words, the values of the mean ± standard deviation, the maximum value, and the minimum value are displayed or printed in different colors.

According to this aspect, it is easy to grasp the degree of the variation which is important in abnormal walking a plurality of times, and at the same time, by comparing the magnitude of the variation, it is easy to determine whether the left-right difference is significant. Can be determined.

According to a twenty-fifth aspect of the present invention, a stride length, a step length, and a step are detected as the parameters, and the stride length, the step length, and the step are displayed or printed as a radar chart. is there.

According to a twenty-sixth aspect of the present invention, a step length and a stride length which are lengths in the walking direction are selected as parameters in the vertical direction of the radar chart, and a length orthogonal to the walking direction is selected as a parameter in the horizontal direction of the radar chart. That is, a step distance is selected, and parameters relating to the left and right feet are arranged corresponding to the left and right positions of the radar chart.

According to this aspect, it is possible to intuitively grasp the balance of each spatial parameter and the balance between them.

According to the present invention, a stride time, a step time, a simultaneous fixing time, and a free leg time are detected as the parameters, and the stride time, the step time, the simultaneous fixing time, and the free leg time are detected. It is displayed or printed as a radar chart.

According to a twenty-eighth aspect of the present invention, a step in a walking direction and a time required for a stride are selected as parameters in a vertical direction of a radar chart, and a simultaneous fixing time of both right and left legs and a free leg are selected as parameters in a horizontal direction of the radar chart. The time is selected, and the respective parameters relating to the left and right feet are arranged corresponding to the left and right positions of the radar chart.

In this embodiment, the balance of each spatial parameter and the balance between them can be intuitively grasped.

[0049]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of a walking pattern processing apparatus according to the present invention. In the figure, 1 is a foot of a pedestrian, 2 is a pressure sensor unit for acquiring a two-dimensional pressure distribution, and 3 is a time-series pressure distribution image detection for acquiring the output of the pressure sensor unit 2 as time-series data at fixed time intervals. , 4 is a superimposed image creation unit that superimposes the time-series pressure distribution image in the time direction, 5 is a foot pressure mass region cutout unit that extracts a foot pressure mass region from the superimposed image, and 6 is a time direction image of the foot pressure mass image. , A step correspondence detecting unit for detecting the correspondence of each step with the original time-series image sequence, a center-of-gravity position detecting unit 7 for detecting the position of the center of gravity of the image sequence, and 8 a moving direction of the center-of-gravity position sequence. A moving direction detecting unit 9 for detecting a feature position detecting unit 9 for detecting a feature position of each image associated with the step correspondence detecting unit 6; The initial frame of the step Initial frame detection unit for output, 11 last frame detector for detecting the last frame of the steps of each image after being compatible in step correspondence detection unit 6, 1
Reference numeral 2 denotes a time interval detection unit that detects a time interval between the initial frame and the last frame, and 13 denotes a walking parameter display unit that displays the respective walking parameters.

FIG. 2 is a diagram showing an example of the flow of processing of a pressure distribution image by the walking pattern processing apparatus shown in FIG. In the figure, 21 is an acquired time-series pressure distribution image, 22 is a superimposed image, a square frame in the superimposed image is a one-step foot pressure image block area, and 23 is a feature point superimposed on the superimposed image. This is the image displayed. 25 is a step length measured on the image 23, and 25 is an image 23
Represents the stride length measured above. 26,27
Represents the first frame and the last frame of one foot pressure image block area, respectively.

FIGS. 3, 4, and 5 are flowcharts showing details of the processing of the pressure distribution image. Hereinafter, FIGS.
A description will be given based on FIGS.

First, the foot 1 walks on the pressure sensor 2. At this time, the pressure sensor unit 2 acquires a unique two-dimensional pressure distribution based on the walking pattern of the foot 1 (step S5).
01). That is, the pressure sensor unit 2 obtains a pressure value output corresponding to the coordinate value of each pressure sensor.

The time-series pressure distribution image detecting section 3 detects a two-dimensional pressure distribution pattern continuously obtained by the pressure sensor section 2 for a predetermined period of time at predetermined time intervals, and obtains T time-series pressure distribution images. (T = 1, 2,..., T) is acquired (step S502). This fixed time interval is desirably 1/10 second to 1/100 second. For example, 1/1
When measurement is performed for 5 seconds at a time interval of 0 second, t = 1 to 5
Fifty time series pressure distribution images numbered 0 are obtained.

The superimposed image creating section 4 superimposes the time-series pressure distribution image acquired by the time-series pressure distribution image detecting section 3 on the time method (step S503). In this superimposition processing, the following two processing methods are effective. One is to use the maximum value of the pressure value output of each sensor as a value as a superimposed image of the position corresponding to that sensor, and the other is to use the total value of the pressure value output of each sensor as the sensor Is a method of setting a value as a superimposed image at a position corresponding to.

The foot pressure lump area cutout unit 5 cuts out a region regarded as a footprint of the same step on the superimposed image as a foot pressure lump area (step S504). At this time, there is no problem in that adjacent pixels each having a value on the superimposed image are considered to be regions of the same step, but even in the same step as in the case of each finger, the pressure distribution image The following processing is effective because there may be a part separated above.

That is, in the superimposed image, a region where a pixel has a value is enlarged by 1 pixel to 10 pixels, and a region overlapping when the region is enlarged is regarded as a footprint region of the same step. The processing for enlarging the area is described in FIGS.
As shown in (b), this can be realized by including pixels near the region as the same region. Here, FIG. 6A is a diagram illustrating an original image, and FIG. 6B is a diagram illustrating an image obtained by performing an enlargement process for one pixel, for example. More specifically, it is assumed that the original image shown in FIG. 6C is obtained. When this original image is subjected to enlargement processing, FIG.
The image shown in (d) is obtained. 6D is regarded as a footprint region of the same step, and a region corresponding to the region is cut out from FIG. 6C to obtain a superimposed image as shown in FIG. 6E. Is obtained for each step.

Since the shape of the foot region is close to a rectangle that is long in the traveling direction, it is effective to change the enlargement ratio according to the direction in the enlargement processing shown in FIG. For example, the processing is such that the image is enlarged by 10 pixels in the traveling direction and is enlarged by 5 pixels in the direction orthogonal thereto. According to such processing, even when the left and right footprints are close to each other in the superimposed image, the footprint region can be stably cut out.

It is also effective to have an interface that allows a human to visually observe the superimposed image and interactively designate a footprint area at the same step.

The foot pressure mass area cutout unit 5 also counts the number N of foot pressure masses (step S505).

Next, the step correspondence detecting section 6 searches the time-series pressure distribution image from t = 1 for each step cut out by the foot pressure lump area cutting section 5, and finds at least one pixel in the step area. A time-series pressure distribution image having a pressure value is detected (steps S506 to S510). That is, for example, an array M (Ψ (t),
n) is prepared. Here, n is a parameter for specifying the order of the step. M = 1 is set for a time-series pressure distribution image having a pressure value in at least one pixel (step S507); otherwise, M =
It is set to 0 (step S508). In practice, for example, M
(Ψ (5), 1) to M (Ψ (25), 1) = 1, M (Ψ
(20), 2) to M (Ψ (40), 2) = 1, M (Ψ
(35), 3) to M (Ψ (55), 3) = 1,...
・ It becomes like this.

The center-of-gravity position detecting section 7 detects each pressure distribution image ｛Ψ
(T), the center of gravity of the pressure value is obtained for t = 1 to T｝ (step S511). At this time, there are the following two types of target ranges for obtaining the center of gravity. One is obtained for the entire pressure image, that is, for both feet, and the other is obtained for each step, that is, for each foot.

The moving direction detector 8 detects the direction in which the center of gravity moves in time based on the center of pressure of each pressure distribution image sequence obtained by the center of gravity position detector 7 (steps S512 to S515). ). That is,
It is determined whether the center of gravity has moved in the same direction as the walking direction or in the opposite direction with respect to the immediately preceding pressure distribution image. For example, an array G (Ψ (t)) is prepared, and a pressure distribution image Ψ (t
From -1), if the moving method of the center of gravity in the pressure distribution image Ψ (t) matches the walking method, G (Ψ (t)) =
1 and G (１ (t)) = 0 in the opposite direction. The direction of movement of the center of gravity can also be detected for both feet or for each foot.

The initial frame detecting section 10 selects one of the images for each step detected by the step correspondence detecting section 6.
A frame having a pressure value first in the step, that is, a pressure distribution image Ψ _{n, s} is detected (step S516). Specifically, in M (Ψ (t), n), a frame where M (Ψ (t), n) ≠ 0 is first detected starting from t = 1.

The final frame detecting section 11 selects a frame having the last pressure value within one step, that is, a pressure distribution image Ψ _{n, e} from the images for each step detected by the step correspondence detecting section 6. Detect (Step S51)
7). Specifically, a frame in which M (Ψ (t), n) = 0 is detected first, starting from a frame having a pressure value first.

The time interval detecting section 12 determines the time from the start of a certain step to the start of the next step (step time), the right foot or the right foot based on the information detected by the initial frame detecting section 10 and the last frame detecting section 11. Time from the start of the left foot step to the start of the next left or right foot step (stride time), the time during which both the left and right feet touch the ground (simultaneous fixation time), and the time during which one foot is floating (swing time) ) Is detected (step S519). Specifically, detected as the step time by multiplying the frame time interval (Ψ _{n + 1, s} number _{(t n + 1, s)} -Ψ n, s number (t _{n, s)).} (Ψ _{n + 2, s} number of _{(t n + 2, s)} -Ψ n, s
(T _{n, s} ) is multiplied by the frame time interval and detected as a stride time. (Ψ _{n + 1, s} number (t _{n + 1, s} )
-Ψn _{, e} number (t _{n, e} )) is multiplied by the frame time interval and detected as a simultaneous fixing time. (Ψn _{+ 2, s} number (t
_{n + 2, s) -Ψ n} , e number of (t _{n, e))} to be multiplied by the frame time interval is detected as the free leg time.

The characteristic position detecting section 9 detects a predetermined characteristic position from the superimposed image representing the footprint area of each step (steps S520 to S526). As the feature position, the rearmost portion, the tipmost portion, the center, or the center of gravity in the walking direction in the superimposed image representing the footprint region of each step is effective.

For a pedestrian who normally walks, the position of the rearmost portion can be detected most stably. However, in an abnormal walking in which the heel does not touch the ground in the first image in one step, the position detection of the rearmost portion may not be stable. Therefore, when the center of gravity of the pressure distribution of the first image in the step is not included in the rear third area of the superimposed image, a switching process such as using the center or the center of gravity instead of the last part is effective. .

Specifically, first, the last position h _n of the n-th foot pressure mass is calculated (step S520). It calculates the position s _n of the cutting edge portion of the n-th foot圧塊(step S
521). The center of gravity g _n of the region having the pressure value in the pressure distribution image Ψ _{n, s} having the first pressure value in one step detected in step S516 is calculated (step S52).
2).

Next, it is determined whether or not equation (1) is satisfied (step S523).

H _n − (s _n −h _n ) / 3> g _n (1) That is, a distance of (the total length of the foot pressure mass) / 3 from the last position h _{n of the} foot pressure mass in the walking direction. position, it is determined whether the front of the center of gravity g _n of. This is determined for all foot pressure masses (step S524). as a result,
For all foot圧塊when the determination is affirmative determination in step S523 is the feature points for every foot圧塊position h _n of the last section (step S525). On the other hand, if the determination in step S523 is negative for any of the foot pressure masses, the center of gravity is set as a feature point for all foot pressure masses (step S52).
6).

The walking parameter display unit 13 displays information detected by the center-of-gravity position detecting unit 7, the moving direction detecting unit 8, the characteristic position detecting unit 9, the initial frame detecting unit 10, and the time interval detecting unit 12 by human beings intuitively. The information is displayed on the display device so as to be easily recognized (step S527).

FIG. 7 is a diagram showing a screen displaying a superimposed image and a movement of the center of gravity in parallel with the superimposed image, and is a so-called space chart. The walking parameter display unit 13 sets the distance in the walking direction as the horizontal axis, and displays the superimposed image created in step S503 at the center in the vertical direction.
The position of each center of gravity of the time-series pressure distribution image detected by the center-of-gravity position detection unit 7 is plotted at the lower part of the superimposed image. In this embodiment, the center-of-gravity position movement (for each square foot) 35
And the movement of the center of gravity position (both feet) 36 are displayed. In the display of the movement of the position of the center of gravity, the display method is changed depending on the moving direction according to the change in the moving direction obtained in steps S512 to S515. For example, the color is changed, or a solid line and a broken line are used. In this embodiment, as indicated by a display 37, the movement in the opposite direction is indicated by a broken line.
In the actual screen, it is easier to see the display in different colors, such that the position of the center of gravity in the walking direction is displayed in blue and the position of the center of gravity in the opposite direction is displayed in red.

Further, the walking parameter display section 13 displays a step length 24 and a stride length 25 at the top of the superimposed image based on the information of the last part and the foremost part of each step detected by the characteristic position detecting section 9. Display as a line segment in the direction. Further, the step 34 is displayed as a horizontal line segment so as to overlap the superimposed image. Here, the step length refers to an interval from the right foot to the left foot or from the left foot to the right foot in the walking direction.
The stride length is an interval from the right foot to the left foot and back to the right foot, or from the left foot to the right foot and back to the left foot. The step distance refers to a characteristic position interval in a direction orthogonal to the walking direction.

The walking parameter display unit 13 refers to the center of gravity position of the initial frame obtained by the initial frame display unit 10 from the center of gravity position detection unit 7 and displays the same as the initial ground contact center position 38 on the superimposed image.

Note that these integrated displays can be printed as they are, as well as displayed on the display device.

FIG. 8 is a diagram showing a display of a time chart. The walking parameter display unit 13 sets the walking time as the horizontal axis, refers to each time parameter from the time interval detection unit 12, and displays the respective time parameters when the left foot is the starting point and when the right foot is the starting point. Separately, they are displayed as horizontal line segments together with the time axis. When displaying this,
The sum or average of the pressure values at each time point is expressed by changing the display color or density. At this time, it is effective to additionally describe the pressure level scale 41. Also,
Since the pressure level has a large individual difference, it is effective to change the color or the density scale of the pressure level for each individual with the maximum pressure value of the pedestrian to be measured as a full scale. On this pressure level scale, preferably
Larger pressure levels are displayed in a reddish color, and lower pressure levels are displayed in a bluish color.

FIG. 9 is a diagram showing a display in which the space parameter and the time parameter are compared for the left and right feet.
The walking parameter display unit 13 displays spatial parameters obtained from the characteristic position detecting unit 9 and the time interval detecting unit 12, ie, stride length, step length, and step distance, and time parameters, ie, stride time, step time, swing time, And the simultaneous fixing time are displayed on the left and right legs, respectively, as vertical lines. Regardless of whether the walk is a plurality of walks or a single walk, a plurality of sets of parameters can be obtained from the plurality of walks. Therefore, the average value 54, the average ± standard deviation value 55, 56, the maximum value 58, and the minimum value 57 are displayed in a different color and superimposed.

The example shown in FIG. 9 is data of a left-sided paralysis patient. The time parameter particularly affects the left / right balance of the average value, and the spatial parameter particularly affects the foot on the paralysis side (left side). It is shown that the parameters have large variations.

FIG. 10 is a diagram showing an example of a radar chart display relating to spatial parameters. The walking parameter display unit 13 displays the spatial parameters obtained from the characteristic position detecting unit 9, that is, the stride length, the step length, and the step distance as a radar chart separately for the left and right feet. At this time, the step length and the stride length, which are the lengths in the walking direction, are selected as the vertical parameters of the radar chart, and the step interval, which is the length orthogonal to the walking direction, is selected as the horizontal parameter of the radar chart. . Further, the respective parameters relating to the left and right feet are arranged corresponding to the left and right positions of the radar chart. Thereby, the aspect ratio and the left / right balance of the entire radar chart indicate the walking pattern.

The example shown in FIG. 10 is data of a left hemiplegic patient, and shows that the steps and stride of the left foot are smaller than those of the right.

FIG. 11 is a diagram showing an example of a radar chart display relating to time parameters. The walking parameter display unit 13 displays the time parameters obtained from the time interval detection unit 12, that is, the stride time, the step time, the free leg time, and the simultaneous fixing time, separately for the left and right feet, and displays them as a radar chart. At this time, the time required for the step and the stride in the walking direction are selected as the parameters in the vertical direction of the radar chart, and the simultaneous fixing time and the free leg time of the left and right legs are selected as the parameters in the horizontal direction of the radar chart. Further, the respective parameters relating to the left and right feet are arranged corresponding to the left and right positions of the radar chart. Thereby, the aspect ratio and the left / right balance of the entire radar chart indicate the walking pattern.

The example shown in FIG. 11 is data of a left hemiplegic patient. The simultaneous fixation time from left to right is extremely long and it is difficult to shift weight from left to right, that is, the kicking force of the left foot Has been shown to be inadequate.

[0084]

According to the present invention described above, spatial and temporal parameters of a walking motion can be automatically and stably acquired easily without imposing a large burden on the subject. In addition, it facilitates intuitive grasp and quantitative comparison of spatial and temporal factors of the walking motion. From this, according to the present invention, for example, the following effective use becomes possible.

(A) In the medical field, the present invention can be used for measuring the severity of a disease in which a walking motion is impaired.

(B) When performing rehabilitation of walking motion in the medical field, it can be used to measure the degree of progress.

(C) By comparing before and after administration of a drug effective for improving walking motion in the medical field, drug efficacy can be measured.

(D) In the medical field, it is possible to measure the walking motion of a patient having no objective symptoms such as dizziness and low back pain to measure the degree of serious injury.

(E) The information of the footprint position and its time, which can be stably detected, obtained by the present invention is used, and at the same time, an image of the walking motion is obtained. And walking motion parameters such as the angles of the knee joint and the hip joint can be stably detected.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram of an embodiment of a walking pattern processing device of the present invention.

FIG. 2 is a diagram showing an example of a flow of processing of a pressure distribution image by the walking pattern processing device shown in FIG.

FIG. 3 is a flowchart illustrating details of processing of a pressure distribution image.

FIG. 4 is a flowchart illustrating details of processing of a pressure distribution image.

FIG. 5 is a flowchart illustrating details of processing of a pressure distribution image.

FIG. 6 is a diagram illustrating a process of cutting out a region regarded as a footprint of the same step on a superimposed image as a foot pressure lump region.

FIG. 7 is a diagram showing a screen on which a superimposed image and a movement of the center of gravity and the like are displayed in parallel with the superimposed image, that is, a so-called space chart.

FIG. 8 is a diagram showing a display of a time chart.

FIG. 9 is a diagram showing a display in which a space parameter and a time parameter are compared for right and left feet.

FIG. 10 is a diagram showing an example of a radar chart display relating to spatial parameters.

FIG. 11 is a diagram illustrating an example of display of a radar chart relating to a time parameter.

FIG. 12 is a diagram showing a conventional method of displaying walking parameters.

FIG. 13 is a diagram showing a conventional method of displaying walking parameters.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Pedestrian's foot 2 Pressure sensor part 3 Time series pressure distribution image detection part 4 Superposition image creation part 5 Foot pressure mass area cutout part 6 Step correspondence detection part 7 Center of gravity position detection part 8 Movement direction detection part 9 Characteristic position detection part 10 Initial frame detection unit 11 Last frame detection unit 12 Time interval detection unit 13 Walking parameter display unit

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Masanobu Arai 3-19-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo Within Nippon Telegraph and Telephone Corporation (56) References JP-A 7-160883 (JP, A) 60-108033 (JP, A) JP-A-8-145826 (JP, A) JP-A-8-145825 (JP, A) Atsushi Yamato, three others, individual from foot pressure image using pressure sensor mat Examination of Identification, IEICE Technical Report PRU94-59-63, November 17, 1994, Vol. 339, pp. 15-22 (58) Field surveyed (Int.Cl. ⁷ , DB name) G06T ⁷ /00-7/60 G06T 1/00 A61B 5/11 G01L 5/00

Claims (28)

(57) [Claims] 1. a pressure sensor for acquiring a two-dimensional pressure distribution based on walking; a time-series pressure distribution image detecting means for acquiring an output of the pressure sensor as time-series data at fixed time intervals; Superimposed image creating means for creating a superimposed image by superimposing images in the time direction; foot pressure mass area cutting means for extracting a plurality of foot pressure mass areas from the superimposed image; Step correspondence detection means for making correspondence with the sequence pressure distribution image, parameter detection means for detecting a parameter representing the characteristic of walking based on the correspondence of the step correspondence means, and output means for outputting the parameter, A walking pattern processing device comprising: 2. The walking method according to claim 1, wherein the parameter detecting means comprises a characteristic position detecting means for detecting a predetermined characteristic position from the superimposed image based on the correspondence of the step correspondence means. Pattern processing device. 3. The walking pattern processing device according to claim 2, wherein the predetermined characteristic position is a rearmost portion, a leading edge portion, a center, or a center of gravity of each foot mass in the walking direction. 4. An initial frame detecting means for detecting a frame having a pressure value first for each foot pressure lump area based on the correspondence of the step correspondence means, and an association of the step correspondence means. Based on, based on the time information pertaining to each frame detected by the final frame detecting means and the final frame detecting means for detecting the last frame having a pressure value for each foot pressure mass region, the initial frame detecting means,
The walking pattern processing apparatus according to claim 1, comprising: a time interval detecting means for detecting a time parameter. 5. The walking pattern processing device according to claim 4, wherein the time parameters are a stride time, a step time, a simultaneous fixing time, and a free leg time. 6. The apparatus according to claim 1, wherein the parameter detecting means comprises a center-of-gravity position detecting means for obtaining a center of gravity of a pressure value for each pressure distribution image based on the correspondence of the step correspondence means. Walking pattern processing device. 7. The walking pattern processing device according to claim 6, wherein the center-of-gravity position detecting means obtains the center of gravity for both feet. 8. The walking pattern processing device according to claim 6, wherein the center-of-gravity position detecting means obtains the center of gravity for each foot. 9. A moving direction detecting means for detecting a temporally moving direction of the center of gravity based on the position of the center of gravity of the pressure value for each pressure distribution image detected by the center of gravity detecting means. 7. The walking pattern processing device according to claim 6, further comprising: 10. A foot pressure lump region extracting means expands a region having a pixel value by a predetermined number of pixels when extracting each foot pressure lump, and regards an overlapping region when expanded as a footprint region of the same step. The walking pattern processing device according to claim 1, wherein: 11. Obtaining a two-dimensional pressure distribution based on walking as a time-series pressure distribution image at fixed time intervals, superimposing the time-series pressure distribution image in a time direction to create a superimposed image, Extracting a plurality of foot pressure mass regions, performing a correspondence with the time-series pressure distribution image for each of the foot pressure mass regions, detecting a parameter representing the characteristic of walking based on the correspondence, And displaying or printing a walking pattern. 12. As the parameter, a rearmost portion, a leading edge portion, and a center of a foot pressure mass region in each step are detected, and information on a stride length and a step length is displayed or printed in parallel with the superimposed image. 12. The walking pattern processing method according to claim 11, further comprising displaying or printing step information on the superimposed image. 13. As the parameter, detecting the center of gravity of the pressure value for each pressure distribution image, and displaying or printing the position of the center of gravity of the pressure value for each pressure distribution image in parallel with the superimposed image. The walking pattern processing method according to claim 11, wherein: 14. When detecting the center of gravity of the pressure value, detecting the moving direction of the position, and changing the display or printing method when the center of gravity moves in the direction opposite to the walking direction. The walking pattern processing method according to claim 13, wherein: 15. The walking pattern processing method according to claim 14, wherein when the center of gravity moves in a direction opposite to the walking direction, the color of display or printing is changed. 16. The walking pattern processing method according to claim 13, wherein the center of gravity is detected for each foot and its position is displayed or printed. 17. The walking pattern processing method according to claim 13, wherein the center of gravity is detected for both feet and the positions are displayed or printed. 18. The method according to claim 18, wherein a center of gravity of a pressure distribution image having a pressure value is first detected in each step as the parameter, and the position of the center of gravity is displayed or printed on the superimposed image. A walking pattern processing method according to claim 11. 19. A method according to claim 19, wherein a stride time, a step time, a simultaneous fixing time, and a free leg time are detected as the parameters, and for each of the two feet, the contact time is displayed or printed as a line segment along a time axis. The walking pattern processing method according to claim 11, wherein: 20. The walking pattern processing method according to claim 19, wherein a total or an average value of the pressure values at each time point is displayed or printed in a different color or density so as to be superimposed on the line segment. 21. The walking pattern processing method according to claim 20, wherein a pressure level scale is described together with the line segment. 22. The stride time, the step time,
20. The walking pattern processing method according to claim 19, wherein the simultaneous fixing time and the free leg time are detected based on the time of a pressure distribution image having a pressure value first and last in each step. 23. As the parameters, a stride length, a step length, a step interval, a stride time, a step time, a free leg time, and a simultaneous fixation time are detected. The walking pattern processing method according to claim 11, wherein the interval, stride time, step time, free leg time, and simultaneous fixing time are displayed or printed on a scale. 24. A plurality of data of stride length, step length, step interval, stride time, step time, swing time, and simultaneous fixing time are printed or displayed, respectively, and an average value, average ± standard deviation of the plurality of data are displayed. 24. The walking pattern processing method according to claim 23, wherein the value, the maximum value, and the minimum value are displayed or printed in different colors. 25. The apparatus according to claim 11, wherein a stride length, a step length, and a step are detected as the parameters, and the stride length, the step length, and the step are displayed or printed as a radar chart. Walking pattern processing method. 26. A step length and a stride length which are lengths in the walking direction are selected as vertical parameters of the radar chart, and a step interval which is a length orthogonal to the walking direction is selected as a horizontal parameter of the radar chart. 26. The walking pattern processing method according to claim 25, wherein respective parameters related to the left and right feet are arranged corresponding to the left and right positions of the radar chart. 27. A stride time, a step time, a simultaneous fixing time, and a free leg time are detected as the parameters, and the stride time, the step time, the simultaneous fixing time, and the free leg time are displayed or printed as a radar chart. The walking pattern processing method according to claim 11, wherein: 28. A time required for a step and a stride in a walking direction are selected as parameters in a vertical direction of a radar chart, and a simultaneous fixing time and a free leg time of right and left legs are selected as parameters in a horizontal direction of the radar chart. 28. The walking pattern processing method according to claim 27, wherein the respective parameters relating to the feet are arranged corresponding to the left and right positions of the radar chart.