KR102433074B1 - Balance training system with heterogeneous sensors and method for determining correct posture using the same - Google Patents

Abstract

A method for determining a correct posture performed by a balance training system including a heterogeneous sensor module according to an embodiment includes: acquiring image information of a user based on a depth sensor module; obtaining skeleton position information for a plurality of skeletons preset for a user based on image information; generating at least one piece of first coordinate information corresponding to at least one bone of interest based on the bone location information; generating at least one pair of inclination information based on at least one piece of first coordinate information and second coordinate information that is output information of the body sensor module; determining whether the user's posture is normal based on at least one pair of inclination information and preset normal data; and generating alarm information for the user based on the determination of whether the posture is normal.

Description

BALANCE TRAINING SYSTEM WITH HETEROGENEOUS SENSORS AND METHOD FOR DETERMINING CORRECT POSTURE USING THE SAME

The present specification relates to a balance training apparatus for a rehabilitation patient, and more particularly, to a balance training system equipped with heterogeneous sensors and a method for determining a correct posture using the same.

In the case of patients with cranial nervous system diseases including stroke, such as hemiplegic patients, problems such as body twisting may occur due to an abnormal sense of balance.

If there is an abnormality in the sense of balance, it is necessary to evaluate the degree of balance abnormality, etc. to determine the state of balance, and to perform rehabilitation training for balance as a treatment method for the abnormality.

There are various devices for balance evaluation and balance training, but most of them are heavy and can perform balance evaluation and balance training only in a designated location, or even devices that are easily moved and used only provide limited balance evaluation and balance training methods. There is a problem.

As a conventional proposal, reference is made to a fall risk assessment system in Korean Patent Publication No. 10-1932553.

An object of the present specification is to more accurately analyze the posture of a rehabilitation patient by providing a balance training system equipped with heterogeneous sensors and a method for determining a correct posture using the same.

A method for determining a correct posture performed by a balance training system including a heterogeneous sensor module according to an embodiment includes: acquiring image information of a user based on a depth sensor module; obtaining skeleton position information for a plurality of skeletons preset for a user based on image information; generating at least one piece of first coordinate information corresponding to at least one bone of interest based on the bone location information; generating at least one pair of inclination information based on at least one piece of first coordinate information and second coordinate information that is output information of the body sensor module; determining whether the user's posture is normal based on at least one pair of inclination information and preset normal data; and generating alarm information for the user based on the determination of whether the posture is normal.

According to the present embodiment, based on the balance training system provided with heterogeneous sensors, it is possible to more accurately analyze whether the patient in the balance training system trains in the correct posture. Accordingly, it is possible to provide a more effective rehabilitation program to the rehabilitation patient.

1 is a view showing a balance training system according to an embodiment of the present specification.
2 is a view showing in detail the sensor plate according to the present embodiment.
3 is a diagram illustrating a method of performing static evaluation according to an exemplary embodiment.
4 is a diagram illustrating a method of performing dynamic evaluation according to an exemplary embodiment.
5 is a diagram illustrating a method of performing pressure evaluation according to an exemplary embodiment.
6 is an evaluation screen regarding the measurement result of COP.
7 is an evaluation screen regarding weight support rate and pressure distribution.
8 is an evaluation screen regarding the evaluation result of the limit of stability.
9 is a diagram illustrating a result value of the inclination sensor data obtained by the body sensor module 120 .
10 is a view showing a problem of the prior art according to an embodiment of the present invention.
11 is a view for explaining a skeleton of interest associated with a depth sensor module and a torso sensor module of a user according to an exemplary embodiment.
12 is an example of skeleton image information indicating location information on a skeleton of interest managed by the balance training system according to the present embodiment.
13 is a flowchart of a method for determining a correct posture performed by the balance training system according to an embodiment of the present invention.
14 is a more detailed flowchart of a method of determining a correct posture performed by the balance training system according to an embodiment of the present invention.

The foregoing characteristics and the following detailed description are all exemplary matters for helping the description and understanding of the present specification. That is, the present specification is not limited to such an embodiment and may be embodied in other forms. The following embodiments are merely examples for fully disclosing the present specification, and are descriptions for conveying the present specification to those skilled in the art to which the present specification belongs. Therefore, when there are several methods for implementing the elements of the present specification, it is necessary to make it clear that the implementation of the present specification is possible in any one of these methods or the equivalent thereto.

In the present specification, when it is stated that a configuration includes specific elements, or when a process includes specific steps, it means that other elements or other steps may be further included. That is, the terms used in this specification are only for describing specific embodiments, and are not intended to limit the concepts of the present specification. Furthermore, the examples described to help the understanding of the invention also include complementary embodiments thereof.

Terms used in this specification have meanings commonly understood by those of ordinary skill in the art to which this specification belongs. Commonly used terms should be interpreted in a consistent sense according to the context of the present specification. In addition, the terms used herein should not be construed in an overly idealistic or formal meaning unless the meaning is clearly defined. Hereinafter, embodiments of the present specification will be described with reference to the accompanying drawings.

1 is a view showing a balance training system according to an embodiment of the present specification.

Referring to FIG. 1 , the balance training system 100 according to the present embodiment includes a sensor plate 110 , a trunk sensor module 120 , a base module 130 , a handrail 140 , and a depth. ) may include a sensor module 150 .

A plurality of sensors may be disposed in the sensor plate 110 plate. For example, the plurality of sensors may include a sensor that measures pressure.

A printed circuit board (hereinafter, 'PCB') and a housing 110a for the circuit board (not shown) may be provided on one or both sides of the sensor plate 110 .

The sensor plate 110 may include one or more magnets. In this case, one or more magnets may be used to fix the sensor plate 110 disposed on the base module 130 .

In addition, one or more magnets are disposed at both ends of the sensor plate 110 (that is, on one side and the other side, and the polarities of the magnets disposed on one side and the other side are different from each other so that one side and the other side of the sensor plate 10 are attached to each other. It can be configured to

The sensor plate 110 may be attached to the base module 130 and used, or the sensor plate 110 may be used alone. That is, the sensor plate 110 is detachable on the base module 130 .

The sensor plate 110 may include a configuration for connecting a wired power source to connect an external power source. In this case, the wired power connection configuration may be a USB type or may include a detachable internal power source in the sensor plate 110 itself. Alternatively, the wired power connection configuration may be understood as a configuration that is charged using a charging adapter.

The sensor plate 110 may be connected to the control module 160 based on wired or wireless communication. In addition, the sensor plate 110 may be connected to the body sensor module 120 based on wired or wireless communication. Here, it will be understood that the wired or wireless communication method may be implemented based on various communication protocols including Bluetooth communication.

The sensor plate 110 may include, for example, indicating the state of the power source, and may indicate a charging state or a connection state. In addition, the sensor plate 110 may include a computer or a body sensor module 120 and the like for displaying a connection state.

According to this embodiment, the sensor plate 110 may be used alone or together with the base module 130 and the handrail 140 during balance evaluation/training, so it is a movable form and can be used in various patient situations. There is an effect that can be used for customization.

The body sensor module 120 may include a sensor capable of measuring inclination. In particular, as a sensor for measuring inclination, the body sensor module 120 is an IMU (Inertial Measurement Unit) sensor implemented based on a 3-axis acceleration sensor and a 3-axis gyro sensor. can be understood

The inclination of the user's upper body may be measured using the torso sensor module 20 . This is because it is difficult to determine whether the user is training in the correct posture only with the foot pressure data obtained from the sensor plate 110 during the user's balance evaluation or balance training.

During balance evaluation or balance training, the user may tilt the upper body to obtain an incorrect posture or pressure sensor data required for evaluation or training, so that the pressure on the inclined side on the sensor plate 110 can be measured higher. .

In this case, when the user's balance evaluation or balance training is performed using only the sensor plate 110 , the result may be derived as correct even though the posture is incorrect.

Accordingly, the torso sensor module 20 may be used together to determine whether the user applies pressure by using an inclination of the upper body during balance evaluation or balance training.

In order to measure the user's upper body inclination, the most preferable position is between the 5th and 7th vertebrae, and when the torso sensor module 120 is disposed at the corresponding position, the upper body inclination may be derived as '0'.

According to this embodiment, the torso sensor module 120 may be disposed between the fifth and seventh vertebrae. However, the arrangement position of the torso sensor module 120 is not limited to the above example, and it may be arranged at a position such that the inclination of the user's upper body becomes '0' or at a position serving as a reference even if the inclination is not '0'.

In addition, only one body sensor may be included in the body sensor module 120 , and a plurality of body sensors may be included.

When a plurality of torso sensors are included in the torso sensor module 120, the torso sensors may be disposed at a distance of more than a certain distance so as to determine whether the user's posture is a bent posture during balance evaluation or balance training for the user. can

The control module 160 may determine whether the user is standing in the correct posture based on the inclination value of the body sensor module 120 .

For example, in the case of using only one body sensor, if the user's posture is bent within the error range of the body sensor module 120, the result may be derived that the user's posture is a correct posture.

That is, since incorrect results are derived even when the user's posture is not correct, the user may continuously perform training or evaluation with the incorrect posture.

For example, when a plurality of sensors are disposed within the body sensor module 20 , the plurality of sensors may be disposed within the body sensor module 20 spaced apart by a predetermined distance to accurately measure the user's torso inclination.

As another example, one sensor is included in the torso sensor module 20 , and a band worn on the user's torso is wide, so that a plurality of torso sensor modules may be included on the band separated by a predetermined distance or more.

As another example, one sensor is included in the body sensor module 120 , a band worn on the user's torso and a plurality of body sensor modules 20 are configured to include one body sensor module 120 . A plurality of bands may be attached to the user.

In this case, if the band is divided and attached to the upper body and lower body, there is an effect of making it easier to determine whether the user has completely sat down or has completely stood up in 'sit to stand' training, which will be described later.

The body sensor module 120 may be connected to the control module 160 to transmit the inclination data obtained by the body sensor module 120 to the control module 160 .

For example, the body sensor module 120 may acquire X-axis coordinate information, Y-axis coordinate information, and Z-axis coordinate information based on a 3-axis gyroscope sensor disposed along the X-Y-Z axes.

In addition, the body sensor module 120 provides tilt information in the left and right direction (X-X') (ie, Roll information), tilt information in the front-back direction (Y-Y') direction based on the 3-axis tilt sensor ( That is, pitch information) and inclination information in the vertical direction Z (ie, yaw information) may be acquired.

Alternatively, the inclination data obtained by the body sensor module 120 may be transmitted to the sensor plate 110 , and then transmitted to the control module 160 together with the pressure data obtained by the sensor plate 110 .

For example, the slope data and pressure data transmitted to the control module 160 may be transmitted as raw data itself, or may be transmitted in the form of preprocessed data according to a predetermined method.

Power supply to the body sensor module 120 may include a wired power connection configuration related to external power, and the wired power connection configuration at this time may be a USB type. In another embodiment, the body sensor module 20 itself may include a detachable internal power source, and in another embodiment, the body sensor module 20 may be charged and used using a charging adapter.

The body sensor module 120 includes a sensor for measuring the inclination of the user's upper body, and is configured to be detachable from the user. As a method of attaching and detaching to the user, a band worn on the body can be used. The band may be detachable by the user, and the body sensor module 120 disposed on the band may also be detached from the band.

The band may include a pocket shape for inserting the body sensor module 20 therein, or the body sensor module 120 may be detachably attached to the band itself.

The body sensor module 120 may include, for example, displaying a state of power, and may indicate a charging state or a connection state. In addition, the body sensor module 120 may include, for example, displaying a connection state with the sensor plate 110 , and may indicate that the sensor plate 110 is connected, connected, waiting, and the like.

As the balance training system, only the sensor plate 110 may be used, and the sensor plate 110 and the body sensor module 120 may be used together.

Although the base module 130 and the handrail 140 are shown in an attached form, the handrail 140 is detachable from the base module 130 . Therefore, the handrail 140 can be assembled and used only when necessary for use, and only the base module 130 can be utilized to use it.

The base module 130 is supported on the ground, and a groove for mounting the sensor plate 110 may be included. For example, when a magnet for fixing the sensor plate 110 to the base module 130 is included, the base module 130 may be formed of a magnetic material.

As described above, the handrail 140 may be separated from the base module 130 and may be constructed in a prefabricated manner. The handrail 40 may be used by a user to hold onto and stand up in a sit-up training to be described later. In addition, the handrail 40 may be used to support the evaluation or training performed while standing on the sensor plate 10 .

The depth sensor module 150 may include an RGB camera, an infrared depth sensor, and an infrared projector.

For example, an infrared projector may emit infrared rays from numerous points toward a person at regular intervals, and an infrared depth sensor may sense infrared rays reflected by colliding with a person to obtain distance information between a camera and an object at each point. Meanwhile, the RGB camera may acquire human image information. The depth sensor module 150 is a device for measuring a user's skeleton, and generates predetermined skeleton image information for a skeleton of interest based on the acquired image information. can do. Here, the skeleton image information will be described in more detail with reference to FIG. 14 to be described later.

In this specification, the depth sensor module 150 may capture an arbitrary object to generate an image and location information about the object.

The depth sensor module 150 may be disposed above the monitor 160 located in front of the user (Y direction) or at a predetermined position. For example, the depth sensor module 150 may be disposed on the Y-Y'-Z plane.

According to the present embodiment, there is provided a method of using position information obtained from the depth sensor module 120 to prevent a case in which the user's incorrect posture is determined as the correct posture within the error range of the torso sensor module 120 . can be

For example, the depth sensor module 150 provides inclination information biased from the center (O) to the left and right directions (X-X') and the front-back direction (Y-Y' from the center O) based on the input image information of the user. ) direction biased slope information can be obtained.

The control device 160 may receive information obtained from a plurality of pressure sensors in the sensor plate 110 , the body sensor module 120 , and the depth sensor module 150 .

For example, the control device 160 may calculate the COP for determining the user's balance level by using the pressure values measured by the plurality of pressure sensors. Here, the control device 160 may be understood as a PC box based on the Android operating system.

Meanwhile, the control device 160 may determine the user's balance level by comparing the inclination measured from the body sensor module 120 with a preset reference inclination value. The result information of the balance level determined by the control device 160 may be reflected in the content stored in the memory and displayed through the monitor 160 .

Furthermore, the control device 160 may determine whether the user's posture is a correct posture based on the inclination information obtained from the body sensor module 120 and the inclination information obtained from the depth sensor module 150 . This will be described in more detail with reference to FIGS. 15 and 16 to be described later.

In addition, the control device 160 induces the user to maintain a balance on various contents to be described later, thereby conducting training for cultivating balance ability.

Meanwhile, in order to extract skeletal data of a patient based on a deep learning model in an Android environment, a deep learning lightweight technology may be applied to the control device 160 according to the present embodiment.

For example, deep learning lightweight technology can be implemented based on the optimized deployment of the existing verified network layer. Alternatively, the layer detection operation and the layer removal operation may be performed through the distribution of the activation function of each layer.

A 3D single recognizable body posture estimation technique may be applied to the control device 160 according to the present embodiment. For example, for the 3D single recognizable body posture estimation technique, a part based learning technique using similar shaped finger features and a layer shuffling technique considering the movement direction of each finger will be used. can

The balance training system referred to herein may further include a movable monitor 160 that provides image content for balance evaluation and balance training.

2 is a view showing in detail the sensor plate according to the present embodiment.

2, when the circuit board and the housing 210a for the circuit board are provided on both sides of the sensor plate 2100, the flexible material of the sensor plate 210 and both ends of the sensor plate 210 According to a configuration in which magnets having different polarities are included, the sensor plate 210 can be conveniently carried by folding and fixing the sensor plate 210 with magnets disposed at both ends.

As another example, when the circuit board and the housing 210a for the circuit board are provided on only one side of the sensor plate 210 , the user moves the front and rear left and right sides of the sensor plate 210 based on the position where the housing of the circuit board is disposed. It can help to clearly distinguish between

Even if the circuit board and the housing of the circuit board are not provided on both sides, but only on one side, portability can be improved because the user can hold and move the housing while carrying it.

3 is a diagram illustrating a method of performing static evaluation according to an exemplary embodiment.

3 (a) is an evaluation using the eyes as an evaluation of a standing state, and the body sensor module 20 can be used together, the number of times of measurement and a measurement time can be set, and the evaluation result is COP ( Center of Pressure) movement length, movement length image of COP, movement area of COP, and average speed of COP can be derived.

3 (b) is an evaluation of a state in which one leg is raised, an evaluation using eyes and a leg, the leg separates the right and left sides, the body sensor module 20 can be used together, and the number of measurements and a measurement time can be set, and as an evaluation result, it is possible to derive the moving length of the COP, the moving length image of the COP, the moving area of the COP, and the average speed of the COP.

3 (c) is an evaluation of a sitting state, an evaluation using the eyes, the body sensor module 20 can be used together, the number of times of measurement and a measurement time can be set, and the evaluation result is COP The moving length of , the moving length image of the COP, the moving area of the COP, and the average speed of the COP can be derived.

4 is a diagram illustrating a method of performing dynamic evaluation according to an exemplary embodiment.

4 (a) is an evaluation of the limit of stability, an evaluation of measuring how far it can move in all directions in one place, and as an evaluation result, it is possible to derive an image of the movement area of the COP and the area of the COP.

4 (b) is an evaluation of the user's walking state, and the right and left legs are divided, and as an evaluation result, the movement direction and pressure distribution of the COP can be derived.

The evaluation of the user's walking state of FIG. 4 (b) is defined as a gate evaluation, and in the gate evaluation, the sensor plate 110 is placed and the sensor plate 110 is stepped on and passed.

At this time, the evaluation result should be derived by dividing the user's right leg and left leg, and in the case of performing evaluation with the other foot despite the instruction to step on the sensor plate 110 with a specific foot during evaluation, the evaluation result go

There is a problem that cannot be accurately derived.

Therefore, by displaying an arrow indicating the direction of use on the sensor plate 110 , the user confirms this, and after distinguishing the direction of the sensor plate 110 , and then performing evaluation, the right side and left side on the sensor plate 110 . By detecting which side of the side is stepping on and passing, there is an effect that the user can distinguish which side of the foot the user is evaluating.

However, even if the position on the sensor plate 110 is divided in order to determine which foot is passed, the right part on the sensor plate 110 is stepped on with the left foot, or the left part on the sensor plate 110 is stepped on with the right foot. There may be errors in the evaluation results.

Therefore, in order to solve the above problems, both feet are placed on the sensor plate 110 before gate evaluation, and the shape of each user's feet is measured through COP and/or plantar pressure of both feet. Then, based on the measurement result, without the placement position of the sensor plate 110 or the guide message of the foot to perform the gate evaluation, the user can automatically recognize the foot performing the evaluation on either foot and derive the evaluation result. can

5 is a diagram illustrating a method of performing pressure evaluation according to an exemplary embodiment.

5A is an evaluation of plantar pressure, a measurement time may be set, and as an evaluation result, a weight support rate and a pressure distribution may be derived.

5 (b) is an evaluation of the sitting pressure, the measurement time can be set, and the movement area of the COP, the weight support rate, and the pressure distribution can be derived.

6 is an evaluation screen regarding the measurement result of COP. Fig. 6 (a) shows the movement path of the COP, and shows the movement length of the COP, the movement area of the COP, and the average speed of the COP. can confirm that there is

7 is an evaluation screen regarding weight support rate and pressure distribution. It can be seen that FIG. 7(a) shows the weight bearing rate and pressure distribution of plantar pressure, and FIG. 7(b) shows the weight bearing rate and pressure distribution of the sitting posture pressure.

Referring to FIG. 7 , the sensor plate 110 is divided into 4 equal parts on the screen of the same shape as the sensor plate 110 to indicate the weight support rate, and the pressure distribution is displayed in color, corresponding to a higher pressure from blue to red. indicates

8 is an evaluation screen regarding the evaluation result of the limit of stability. That is, FIG. 8 is a view showing the evaluation result of the stability limit of FIG. 6( a ). The movement area of the COP is shown on the screen of the same shape as the sensor plate 110 , and the movement area of the COP is also numerically shown. It can be seen that the .

9 is a diagram illustrating a result value of the inclination sensor data obtained by the body sensor module 120 . It can be seen that FIG. 9 shows the coordinate position of the inclination sensor data on the same screen and numerical value as the sensor plate 110 .

10 is a view showing a problem of the prior art according to an embodiment of the present invention. Referring to FIG. 10 , when one torso sensor is used to obtain the user's inclination information as shown in FIG. 10 , even if the user's posture is bent within the error range of the torso sensor module 120 , the user's posture is correct. The result can be derived from the attitude.

11 is a view for explaining a skeleton of interest associated with a depth sensor and a torso sensor of a user according to an exemplary embodiment.

1 to 11, the cervical vertebrae 1110 are composed of 1 to 7, the thoracic vertebrae 1120 is composed of 1 to 12, and the lumbar vertebrae 1130 is composed of 1 to 5, and , the sacrum 1150 may be composed of numbers 1 to 4.

12 is an example of skeleton image information indicating location information on a skeleton of interest managed by the balance training system according to the present embodiment.

1 to 12 , the balance training system according to this embodiment acquires a plurality of predetermined skeletal position information based on the user's image information obtained through the depth sensor module (eg, 150 in FIG. 1 ). can do.

Specifically, the balance training system according to the present embodiment may extract the respective positions (H, T, n1 to n8) of the human joints from the user's image information. In this case, the attachment position (T) of the torso sensor module is preset, and the positions of the remaining joints may be extracted based on an algorithm preloaded in the balance training system according to the present embodiment.

On the other hand, the balance training system according to the present embodiment may form the skeleton image information based on the extracted position information of each of the joints. In this case, the reference point (or starting point) for generating the skeleton image information of FIG. 12 may be the attachment position T of the body sensor module.

In other words, since the skeleton image information of FIG. 12 is formed while extending to the positions (H, n1 to n8) of each joint with the body sensor attachment position (T) as a reference point, relatively high accuracy while using one depth sensor module Skeleton image information with may be obtained.

As an example, the skeleton image information may be generated by applying a body pose estimation technique to the user's image information. Meanwhile, at least one predetermined position of the skeleton of interest (eg, n1 to n4 in FIG. 12 ) At least one piece of coordinate information for can be obtained. In this case, the at least one piece of coordinate information may include a value on the X-axis and a value on the Y-axis with respect to the position of the corresponding skeleton of interest.

For example, n1 of FIG. 12 may correspond to the cervical vertebrae 1110 , and positions T to which the torso sensor module 120 of FIG. 12 is attached may correspond to Nos. 5 to 7 of the thoracic vertebrae 1120 . Also, n2 and n3 of FIG. 12 may correspond to the lumbar vertebrae 1130 , and n3 and n4 of FIG. 12 may correspond to the sacrum 1140 .

13 is a flowchart of a method for determining a correct posture performed by the balance training system according to an embodiment of the present invention.

1 to 13 , in step S1310, the balance training system according to the present embodiment may receive a user input image (ie, a rehabilitation training image) obtained based on the depth sensor module.

In step S1320, the balance training system according to the present embodiment may extract a skeletal position value associated with the user's input image information based on the depth sensor module.

In step S1330, the balance training system according to the present embodiment may extract position values corresponding to a plurality of predetermined positions (ie, n1, n2, n3 in FIG. 12).

In step S1340, the balance training system according to the present embodiment may calculate a left and right (X-X') inclination value and a front-back (Y-Y') inclination value obtained from the torso sensor module.

In addition, the balance training system according to the present embodiment includes the left and right (X-X') inclination values and the front and back (Y-Y') of each of the skeletons of interest (ie, n1, n2, n3 in FIG. 12) predetermined from the depth sensor module. ) to calculate the slope value.

For example, the balance training system according to the present embodiment may calculate the X-coordinate inclination value of each skeleton by using the X-coordinate value of each sensor module. Similarly, the balance training system according to the present embodiment may calculate the Y coordinate inclination value of each skeleton by using the Y coordinate value of each sensor module.

In this specification, the inclination value may be understood as a value obtained by dividing the difference value between the position value of the body sensor module and the position value of the corresponding main skeleton by the position value of the body sensor module.

In step S1350, the balance training system according to the present embodiment may determine whether the posture with respect to the X coordinate is normal by comparing the inclination value of the X coordinate of each skeleton and the first reference value.

In step S1360, the balance training system according to the present embodiment may determine whether the posture with respect to the Y coordinate is normal by comparing the Y coordinate inclination value and the second reference value.

According to the present embodiment, it may be determined whether the user's posture is normal without attaching a plurality of sensors to each joint.

For a patient with an uncomfortable body, attaching a plurality of sensors may cause fatigue and discomfort. . Furthermore, since a depth sensor for measuring the patient's skeleton is applied, the patient's posture can be measured, and thus, there is an advantage in that the patient's posture can be determined more conveniently and accurately than before.

14 is a more detailed flowchart of a method of determining a correct posture performed by the balance training system according to an embodiment of the present invention.

1 to 14 , in step S1410 , the balance training system 100 according to the present embodiment may acquire image information of the user from the depth sensor module 150 .

In step S1420, the balance training system 100 according to the present embodiment provides a plurality of preset skeletons (eg, H, T, n1 to Skeletal position information about n8) can be obtained.

For example, one skeleton image information may be formed based on skeleton position information regarding a plurality of skeletons (eg, H, T, n1 to n8 in FIG. 12 ).

Here, the attachment position (T) preset for the body sensor module may be understood as a reference point for forming the skeleton image information and a starting point for generating the skeleton image.

In step S1430, the balance training system 100 according to the present embodiment generates at least one first coordinate information corresponding to at least one skeleton of interest (eg, n1 to n3 in FIG. 12 ) based on the skeletal position information. can do.

For example, the at least one piece of first coordinate information may include an X-coordinate value (x1) and a Y-coordinate value (y1) corresponding to the n1 position in FIG. 12, and an X-coordinate value (x2) and Y corresponding to the n2 position in FIG. 12 . It may include a coordinate value y2, and an X-coordinate value (x3) and a Y-coordinate value (y3) corresponding to the n3 position of FIG. 12 .

In step S1440 , the balance training system 100 according to the present embodiment generates at least one pair of inclination information based on at least one piece of first coordinate information and second coordinate information that is output information of the torso sensor module 120 . can do.

Here, the second coordinate information may include an X-coordinate value (T_x) and a Y-coordinate value (T_y) corresponding to the T position of FIG. 12 .

For example, the pair of inclination information corresponding to the n1 position of FIG. 12 may include X-axis inclination information (X_n1) and Y-axis inclination information (Y_n1).

As an example, the X-axis inclination information (X_n1) associated with the n1 position of FIG. 12 is the X-coordinate value (T_x) corresponding to the T position of FIG. 12 and the X-coordinate value (x1) corresponding to the n1 position of FIG. It may be understood as a value divided by the X-coordinate value (T_x) corresponding to the T position of FIG. 12 .

As an example, the Y-axis inclination information (Y_n1) associated with the n1 position of FIG. 12 is the difference between the Y-coordinate value (T_y) corresponding to the T position of FIG. 12 and the Y-coordinate value (y1) corresponding to the n1 position of FIG. It may be understood as a value divided by the Y coordinate value (T_y) corresponding to the T position of FIG. 12 .

For example, the pair of inclination information corresponding to the n2 position of FIG. 12 may include X-axis inclination information (X_n2) and Y-axis inclination information (Y_n2).

As an example, the X-axis inclination information (X_n2) associated with the n2 position of FIG. 12 is the X-coordinate value (T_x) corresponding to the T position of FIG. 12 and the X-coordinate value (x2) corresponding to the n2 position of FIG. It may be understood as a value divided by the X-coordinate value (T_x) corresponding to the T position of FIG. 12 .

As an example, the Y-axis inclination information (Y_n2) associated with the n2 position of FIG. 12 is the difference between the Y-coordinate value (T_y) corresponding to the T position of FIG. 12 and the Y-coordinate value (y2) corresponding to the n2 position of FIG. 12 . It may be understood as a value divided by the Y coordinate value (T_y) corresponding to the T position of FIG. 12 .

For example, the pair of inclination information corresponding to the n3 position of FIG. 12 may include X-axis inclination information (X_n3) and Y-axis inclination information (Y_n3).

As an example, the X-axis inclination information (X_n3) associated with the n3 position of FIG. 12 is the X-coordinate value (T_x) corresponding to the T position of FIG. 12 and the X-coordinate value (x3) corresponding to the n3 position of FIG. It may be understood as a value divided by the X-coordinate value (T_x) corresponding to the T position of FIG. 12 .

As an example, the Y-axis inclination information (Y_n3) associated with the n3 position of FIG. 12 is the Y-coordinate value (T_y) corresponding to the T position of FIG. 12 and the Y-coordinate value (y3) corresponding to the n3 position of FIG. It may be understood as a value divided by the Y coordinate value (T_y) corresponding to the T position of FIG. 12 .

In step S1450, the balance training system 100 according to the present embodiment may determine whether the user's posture is normal based on at least one pair of inclination information and preset normal data.

Here, at least one pair of slope information may be associated with at least one skeletal interest (eg, n1 to n3 of FIG. 12 ). Also, the preset normal data may be associated with at least one skeletal interest (eg, n1 to n3 of FIG. 12 ).

For example, the balance training system 100 according to the present embodiment may determine whether the difference between the pair of inclination information corresponding to the n1 position and the normal data associated with the n1 position is greater than a preset threshold value.

If it is determined to be greater than the threshold value, the balance training system 100 according to the present embodiment may confirm that the user is taking an incorrect posture at the n1 position.

As another example, the balance training system 100 according to this embodiment may determine whether the difference between the pair of inclination information corresponding to the n2 position and the normal data associated with the n2 position is greater than a preset threshold value.

If it is determined that it is greater than the threshold value, the balance training system 100 according to the present embodiment may confirm that the user is taking an incorrect posture at the n2 position.

As another example, the balance training system 100 according to this embodiment may determine whether the difference between a pair of inclination information corresponding to the n3 position and the normal data associated with the n3 position is greater than a preset threshold value.

If it is determined to be greater than the threshold value, the balance training system 100 according to the present embodiment may confirm that the user is taking an incorrect posture at the n3 position.

It will be understood that step S1450 is not limited to the above, and may be implemented in a manner of comparing the average value of the inclination values obtained at each location with preset normal data.

In step S1460, the balance training system 100 according to the present embodiment may generate alarm information based on the determination information on whether the user's posture is normal. That is, step S1460 may be understood as a step performed when the user takes an incorrect posture at a specific location.

In the detailed description of the present specification, specific embodiments have been described, but various modifications are possible without departing from the scope of the present specification. Therefore, the scope of the present specification should not be limited to the above-described embodiments and should be defined by the claims and equivalents of the claims of the present invention as well as the claims to be described later.

100: balance training system 110: sensor plate
120: body sensor module 130: base module
140: hand rail 150: depth sensor module
160: control module

Claims (10)

A method for determining a correct posture performed by a balance training system comprising a trunk sensor module and a depth sensor module, the method comprising:
acquiring image information of a user based on the depth sensor module;
obtaining skeleton position information for a plurality of skeletons preset for the user based on the image information;
generating at least one piece of first coordinate information corresponding to at least one bone of interest based on the bone location information;
generating at least one pair of inclination information based on the at least one piece of first coordinate information and second coordinate information that is output information of the body sensor module;
determining whether the posture of the user is normal based on the at least one pair of inclination information and preset normal data; and
Comprising the step of generating an alarm (alarm) information for the user based on the determination of whether the posture is normal,
the at least one first coordinate information includes a first X coordinate value and a first Y coordinate value corresponding to a first position associated with the at least one skeletal interest;
The second coordinate information includes a fourth X coordinate value and a fourth Y coordinate value corresponding to an attachment position of the body sensor module, and
The at least one pair of inclination information includes X-axis inclination information based on the first X coordinate and the fourth X coordinate, and Y-axis inclination information based on the first Y coordinate and the fourth Y coordinate. The method of claim 1,
The skeleton image information corresponding to the user is generated based on the skeleton position information for the plurality of skeletons. 3. The method of claim 2,
The method in which the skeleton image information is generated using a preset attachment position for the torso sensor module among the skeleton position information as a reference point. The method of claim 1,
The at least one first coordinate information includes a second X coordinate value and a second Y coordinate value corresponding to a second position associated with the at least one skeleton of interest, and a second X coordinate value and a second Y coordinate value corresponding to a third position associated with the at least one skeleton of interest. A method further comprising 3 X coordinate values and a third Y coordinate value. delete delete The method of claim 1,
The step of determining whether the posture of the user is normal,
determining that the posture is abnormal when it is determined that a difference between the at least one pair of inclination information and the normal data is greater than a preset threshold value; and
and determining that the posture is normal when it is determined that a difference between the at least one pair of inclination information and the normal data is smaller than a preset threshold value. base module;
a sensor plate disposed on the base module and including a plurality of pressure sensors for measuring a user's foot pressure;
a control device connected to the sensor plate;
a depth sensor module connected to the control device, acquiring image information of the user, and acquiring at least one piece of first coordinate information corresponding to at least one skeleton of interest for the user;
a body sensor module connected to the control device, attached to a preset position for the user, and configured to obtain second coordinate information regarding the preset position,
The control device generates at least one pair of inclination information based on the at least one piece of first coordinate information and the second coordinate information, and the user's posture based on the pair of inclination information and preset normal data A balance training system for determining whether is normal, and generating skeletal position information for a plurality of skeletons preset for the user based on the image information. delete 9. The method of claim 8,
The at least one skeleton of interest is a balance training system that is determined based on the skeleton position information.

Images