WO2020085699A1

WO2020085699A1 - Body shape analyzing method and apparatus

Info

Publication number: WO2020085699A1
Application number: PCT/KR2019/013439
Authority: WO
Inventors: 박재현
Original assignee: (주)엠지솔루션스
Priority date: 2018-10-25
Filing date: 2019-10-14
Publication date: 2020-04-30
Also published as: KR102165429B1; KR20200046637A

Abstract

An embodiment of the present invention provides a body shape analyzing apparatus and method in which a plurality of markers are used with an exact analysis algorithm to precisely analyze the body shape of a person to be measured, so as to prevent a disease due to body shape imbalance and allow the person to be measured to maintain an ideal body shape suitable for him or her. In the apparatus of the embodiment, a camera is configured to capture images of the front, lateral, and rear sides of a person to be measured, to which the markers are attached, and a laser foot plate is configured to display the measurement position of the foot of the person to be measured, by using a laser beam. A calculation means receives front, lateral, and rear side images of the person to be measured from the camera, recognizes the markers in the images, analyzes muscle imbalance for each type of disease, and outputs a result of the analysis to a display. The calculation means may comprise: an image input unit; a marker recognizing unit; a disease-specific estimation unit for analyzing an image in which a marker is recognized, according to a disease-specific analysis algorithm to estimate the muscle imbalance of the person to be measured; and a correction information generating unit for generating customized information for correcting the imbalance of the person to be measured according to a result of the estimation.

Description

Body type analysis method and device

The present invention relates to a body shape analysis device that analyzes a body shape after photographing a subject with a camera, and more specifically, uses a plurality of markers and precisely analyzes the body shape of the subject with an accurate analysis algorithm, resulting in body shape imbalance. It relates to a body shape analysis method and apparatus for preventing a disease and maintaining a customized body shape suitable for a patient.

In recent years, the obesity has rapidly increased due to the westernized eating habits, and the imbalance of body shape is deepened due to lack of exercise due to urban life, and the disease due to the imbalance of body shape is increasing. In other words, if the body imbalance is intensified, pelvic imbalance, knee 휜 leg, muscle imbalance development around the knee, tortoise / straight neck vertebral deformity, vertebral pelvis warping, thoracic muscle / ligament tissue instability, knee overextension, pelvic-leg imbalance, back of the knee It is known that body type diseases such as muscle imbalance development and spine (head bone, spine) bending are likely to occur.

In order to prevent such a body type disease, body type analysis is required. The body type analyzer using a web camera disclosed in Korean Patent Office Publication No. 10-2010-0092555 is an electric motor and a web camera that can move the web camera up and down. The body type is analyzed, including a personal computer capable of acquiring image data captured by a web camera, and an algorithm that quantitatively automatically analyzes the position value of the marker and the degree of inclination of the body in the image.

In addition, the 'Analysis Prescription Service System and Method Based on Body Type Analysis' announced in Publication No. 10-1402781 is taken by uploading the user's whole body, front, back, side, lower body, upper body, etc., and analyzing it in 3D The program analyzes the user's body type by using the program, compares it with the standard body type, and provides exercise prescription information for conversion to the standard body type, if necessary, with the comparison result.

However, such a conventional body type analyzer has a problem in that it is difficult to accurately measure the body type, and thus the reliability of the analysis result is poor, and the size of the body type analysis device is large or expensive. In particular, the conventional body type analyzer has a problem of insufficient evaluation and analysis for diseases caused by body type imbalance.

The present invention has been proposed to solve the conventional problems as described above, and the object of the present invention is to accurately correct a user's body imbalance through a unique analysis algorithm for each disease in order to prevent diseases frequently occurring to modern people due to body imbalance. It is to provide a body type analysis method and apparatus that can be measured and analyzed.

In addition, another object of the present invention is to provide a body analysis method and apparatus that simplifies hardware configuration through laser scaffolding to reduce costs, and to propose effective activities for body correction based on the measured results.

The embodiment of the present invention uses a plurality of markers and accurately analyzes the body shape of the subject with an accurate analysis algorithm to prevent a disease caused by body shape imbalance and a body shape analysis device and method for maintaining a customized body shape suitable for the subject Disclosed.

In the device of the embodiment, the camera is for photographing the front, side, and rear of the subject to which the marker is attached, and the laser scaffolding is for displaying the position of the foot by the laser beam when measuring the subject, and the display is a measurement process and measurement results It is to indicate. The calculation means receives the front, side, and back images of the subject from the camera, recognizes the markers in the picture, analyzes muscle imbalance for each disease, and outputs the analysis results to the display.

The calculation means analyzes an image input unit for receiving image data captured from the camera, a marker recognition unit for recognizing a marker from the input image, and an image in which the marker is recognized according to a predetermined disease-specific analysis algorithm. It includes a disease-specific evaluation unit for evaluating muscle imbalance, a calibration information generation unit for generating customized information for correcting an imbalance of a patient according to the evaluation result, and an output unit for outputting body shape analysis results and body shape correction information. You can.

The method of the implementation example includes receiving an image photographed from the front, side, and back of the subject to which the marker is attached, recognizing the marker from the input image, and evaluating a predetermined disease-specific evaluation algorithm based on the recognized marker. Accordingly, it may include the step of analyzing the body type of the person to be measured, and outputting the result of the body type analysis.

Recognizing the markers includes clicking a marker in a picture, generating a 51x51 pixel color mask around the click coordinates, processing in grayscale, and Sobel Mask edge detection. ) And recognizing the learned support vector machine (SVM) kernel marker.

The steps to analyze the body type include muscle imbalance above the chest, pelvic imbalance, knee extremity, unbalanced development of the muscles around the knee, knee overextension, pelvic tilt, sway back, flat back, turtle neck round shoulder deformity, thoracic scoliosis, lumbar scoliosis At least one of them is analyzed according to the corresponding algorithm.

According to the body shape analysis technology of the present invention, it is possible to reduce the hardware production cost and volume by increasing the practicality by positioning the scaffolding using a cross laser as well as removing the massive grid background occupying only the volume by software image processing.

In addition, according to the body shape analysis technology of the present invention, a number of markers are used to accurately measure the position and angle of bones located at the front, side, and back, and the body's imbalance is accurately measured using a unique algorithm for each disease. There is an advantage that can increase the reliability of the analysis.

In addition, according to the present invention, by implementing a participatory interface in which a user can participate in the body shape analysis process in real time while viewing a large LCD panel, the user can provide a familiar feeling away from the feeling of being unilaterally photographed.

1 is a schematic layout of a body type analysis device according to an embodiment of the present invention,

2 is a block diagram of a body type analysis device according to an embodiment of the present invention,

Figure 3 is a flow chart of the operation of the body analysis apparatus according to an embodiment of the present invention,

4 is a detailed flowchart illustrating a marker recognition procedure in the picture shown in FIG. 3,

Figure 5 is a detailed flow chart showing the muscle imbalance analysis procedure for each disease shown in Figure 3,

6 to 8 are exemplary views for explaining the detailed procedure shown in FIG. 4,

9 and 10 is a schematic diagram showing the position of the marker and the measurement posture when measuring the front surface according to an embodiment of the present invention,

11 and 12 are schematic diagrams showing the position and measurement posture of the marker during side measurement according to an embodiment of the present invention,

13 and 14 are schematic diagrams showing the position of the marker and the measurement posture when measuring the rear surface according to an embodiment of the present invention.

The present invention and the technical problems achieved by the practice of the present invention will be more apparent by the preferred embodiments of the present invention described below. The following examples are merely illustrated to illustrate the present invention, and are not intended to limit the scope of the present invention.

1 is a schematic layout diagram of a body type analysis device according to an embodiment of the present invention, and FIG. 2 is a block diagram of a body type analysis device according to an embodiment of the present invention.

Body shape analysis device 100 according to an embodiment of the present invention, as shown in Figures 1 and 2, the frame 102, 22 markers 104 attached to the body of the subject 10, laser scaffolding ( 110), a camera 120, a computer body 130, an operation unit 140, a storage 150, an LCD 160, and a printer 170.

1 and 2, the frame 102 is for supporting the LCD 160, the computer main body 130, the camera 120, the laser 110, etc., and the marker 104 is the subject 10 ) To accurately recognize the location of the muscle (skeletal) in the image taken by attaching it to the body.

The laser scaffold 110 is implemented as a laser beam generator that generates a cross-shaped laser beam to indicate the position of the subject's foot during measurement, and the camera 120 stops the subject 10 to which the marker 104 is attached. It is a video input means for taking a video or video.

The computer 130 executes software loaded according to the operation of the operation unit 140 such as a keyboard or mouse to execute an image input unit 131, a marker recognition unit 132, a disease-specific evaluation unit 133, and correction information generation unit ( 134), the output unit 135, and the communication unit 136 are implemented to output the body shape analysis result and body shape correction information to the LCD 160 screen or the printer 170.

The image input unit 131 is for receiving image data captured from the camera 120 and storing it in the storage 150, and the marker recognition unit 132 is for recognizing the marker 104 from the input image, and disease The star evaluation unit 133 evaluates the muscle imbalance of the subject by processing the image recognized by the marker 104 according to a predetermined disease-specific analysis algorithm.

The correction information generation unit 134 generates customized information for correcting the imbalance of the person to be measured according to the evaluation result of the evaluation unit 133 for each disease, and the output unit 135 communicates the body shape analysis result and body shape correction information ( It is configured to transmit to the outside through 136) or print to the printer 170 or display it on the LCD 160.

3 is a flowchart of an operation of the body type analysis device according to an embodiment of the present invention

The body shape analysis procedure according to the embodiment of the present invention, as shown in FIG. 3, when the computer 130 is turned on, the body shape analysis program is executed to display the main screen (S1). By selecting the menu on the main screen, if the user is the first user (user), a user database is established, and if the user is a registered user, previously registered user data is loaded (S2).

In the user database, at least items such as a user name, height, age, email address, gender, user bone location information, and measurement date are generated and the corresponding information is recorded. User bone location information includes (Right / Left) Sternal end, (Right / Left) Acromion end, Umbilicus (belly), (Right / Left) Anterior Superior iliac spine (top) Anterior osteoporosis), (Right / Left) mid point of knee, (Right / Left) mid point of ankle, Occipital protuberance, Mid point of acromial process, Body of Lumbar Vertebrae (mid pelvic center), Mid point between knee and popliteus, Lateral malleolus (outer peach bone), (Right / Left) Superior angle, (Right / Left) Inferior angle (lower shoulder angle), (Right / Left) End of scapula spine.

Subsequently, when the user selects a static muscle imbalance measurement, the user with the marker 104 attached is photographed, and the marker 104 is recognized in the photographed picture, and then muscle imbalance by disease is analyzed according to a specific algorithm described later. (S3 ~ S6).

If the user selects the dynamic muscle imbalance measurement, the motion of the user with the marker 104 attached in the form of a video is captured, and the motion of the marker 104 in the captured image is recognized, and then dynamic muscle imbalance by disease Analyze (S7 ~ S9).

Subsequently, when the analysis is completed, body shape correction information is generated to correct imbalance according to the body shape analysis result of the person to be measured, and the analysis result and body shape correction information are output to the LCD 160 or the printer 170 or transmitted by e-mail. (S10 ~ S12).

4 is a detailed flowchart illustrating a procedure for recognizing a marker in a picture illustrated in FIG. 3.

The procedure of recognizing the marker 104 in the picture taken by the camera 120 is as shown in FIG. 4, in a step of clicking a marker in the picture on the screen (S51), and a color mask of 51x51 pixels based on the click coordinates. Generating step (S52), processing in gray scale (S53), performing Sobel Mask (Sobel Mask) edge detection (S54), and learned SVM (Support Vector Machine) kernel (kernel) It consists of a step (S55) for recognizing the marker.

Referring to FIG. 4, in step S51, as shown in FIG. 6, a marker in a picture is clicked, and 22 markers 104 are clicked because 22 markers 104 are attached when shooting. In step S52, a mask of 51x51 color pixels is generated around the clicked marker 104 as shown in FIG. 7, and in step S53, gray scale is performed according to Equation 1 below.

In step S54, an edge is detected with a Sobel mask according to the following Equation 2 to calculate an image as shown in FIG.

Thereafter, in step S55, the SVM kernel marker is recognized according to the following equation (3).

5 is a detailed flowchart illustrating a muscle imbalance analysis procedure for each disease illustrated in FIG. 3.

Referring to FIG. 5, in the muscle imbalance analysis step (S601), the line segment between the upper left clavicle end (P1) and the right clavicle end (P2) is formed with the horizontal line as shown in Equation 4 below. It is evaluated by evaluation, and if the obtained angle is greater than 0 °, it is evaluated as 'right skew imbalance in the muscle above the chest', and if the calculated angle is less than 0 °, it is evaluated as 'skew imbalance in the muscle above the chest.'

In the pelvic imbalance analysis step (S602), as shown in Equation 5 below, the length of the segment connecting the left upper anterior bone pole (P3) and the navel (P4) and the length of the segment connecting the right upper anterior bone (P5) and the navel (P4) The ratio is calculated and evaluated. If the calculated ratio is greater than 1, it is evaluated as 'the pelvis turned to the left', and if the calculated ratio is less than 1, it is evaluated as 'the pelvis turned to the right'.

In the knee thigh analysis step (S603), as shown in Equation 6 below, a first angle (based on the left ankle center point; angle _left ) that a line connecting the left knee bone (P6) and the left ankle center (P7) forms a vertical line, The angle between the line between the right knee bone (P8) and the center of the right ankle (P9) and the vertical line (based on the right ankle center point; angle _right ) is evaluated. At this time, the first angle (angle _left ) is greater than or less than 0 or the second angle (angle _right ) is greater than or less than 0 is evaluated. If the first angle (angle _left <0, angle _right > 0, it is '0 leg', and if angle _left > 0, angle _right <0, it is evaluated as 'X leg'.

In the step of analyzing the unbalanced development of the muscles around the knee (S604), the length of the line connecting the left upper anterior bone pole (P3) and the left knee bone (P6) and the right upper front bone bone (P5) and the right knee bone as shown in Equation 7 below. The ratio (Ratio) of the length of the line segment (P8) is calculated, and if the ratio is greater than 1, it is evaluated as 'left thigh deformation', and if less than 1, it is evaluated as 'right thigh deformation'.

In the knee overextension state analysis step (S605), as shown in Equation 8 below, the angle between the outer peach bone P10 and the point P11 between the hamstring and the knee is obtained from the vertical line, and if the angle is greater than 0 It is evaluated as 'the state of flexion of the knee', and if it is less than 0, it is evaluated as 'the state of hyperextension of the knee'.

In the step of analyzing the total inclination of the pelvis (S606), as shown in Equation 9 below, the angle between the point P11 between the hamstring and the knee and the lateral pelvis P12 is obtained from the vertical line, and the angle is greater than 0. When it is large, it is evaluated as 'the overall inclination of the pelvis'.

In the Sway Back analysis step (S607), as shown in Equation 10 below, the first angle (angle ₁ ) of the line segment connecting the point P11 between the hamstring and the knee and the lateral pelvis P12 and the vertical line, The first angle is less than 0 and the second angle is 0 at the same time as the second angle (angle ₂ ) of the vertical line formed by the line connecting the coordinates and the outer peach bone (P10) and the shoulder blade shoulder lateral neutral part (P13) is obtained. If it is larger, it is evaluated as Sway Back.

In the flat back analysis step (S608), as shown in Equation 10 above, the first angle (angle ₁ ) that the line segment connecting the point P11 between the hamstring and the knee and the lateral pelvis P12 forms a vertical line, When the first angle is less than 0 and the second angle is less than 0 by obtaining the second angle (angle ₂ ) that the line connecting the coordinates and the outer peach bone (P10) and the shoulder blade shoulder lateral neutral part (P13) forms a vertical line Evaluate in the state of flat back.

In the turtle neck round shoulder deformation analysis step (S609), the angle obtained by obtaining an angle (angle) between the line connecting the shoulder blade lateral protrusion (P13) and the occipital protrusion (P14) with the vertical line as shown in Equation 11 is 0. If it is larger, it is evaluated as a 'round shoulder deformation of a turtle neck', and if it is less than 0, it is evaluated as a 'round shoulder deformation of a straight neck'.

In the thoracic scoliosis analysis step (S610), as shown in Equation 12 below, the angle formed by the line between the left shoulder blade spinous end (P15) and the right shoulder bone spinous end (P16) is obtained from the horizontal line and the angle is 0. If it is larger, it is evaluated as 'only the thoracic vertebrae curved to the right' and if it is less than 0, it is evaluated as 'only the thoracic vertebrae curved to the left'.

In the lumbar scoliosis analysis step (S611), as shown in Equation 13 below, the angle formed by the line segment formed by the left upper anterior bone pole (P3) and the right upper upper bone pole (P5) is obtained when the angle is greater than 0. It is evaluated as 'only lumbar vertebrae curved to the right', and if it is less than 0, it is evaluated as 'only lumbar vertebrae curved to the left'.

9 and 10 are schematic diagrams showing positions and measurement postures of markers when measuring the front surface according to an embodiment of the present invention, and FIGS. 11 and 12 are positions and measurement postures of markers during side measurement according to an embodiment of the present invention 13 and 14 are schematic views showing the position and measurement posture of the marker when measuring the rear surface according to an embodiment of the present invention.

9 and 10, the posture at the time of front measurement is looking at the camera 120 and standing in a comfortable position, placing the peach bones of both feet on the horizontal axis of the cross-shaped laser scaffold 110, and attaching the heel. Then, both thumb toes are spread within about 30 degrees, and both arms are naturally stretched to the floor and photographed with the camera 120.

11 and 12, the posture when measuring the side faces the camera 120 and looks at the front. Standing in a comfortable position so that the neck and waist do not enter the stiffness, place the peach bones of both feet on the vertical axis of the cross-shaped laser scaffold 110 and attach the heels. Then, open both big toes within approximately 30 degrees, and for both arms, move forward about 15 degrees so that the lateral pelvic markers are visible.

13 and 14, the posture at the time of the rear measurement stands in a comfortable posture while looking at the front with the camera 120 behind. Thereafter, the peach bones of both feet are placed on the horizontal axis of the cross-shaped laser scaffold 110, and the heels are attached. Then, both thumb toes are spread within about 30 degrees, and both arms are naturally stretched to the floor and photographed with the camera 120.

In the above, the present invention has been described with reference to one embodiment shown in the drawings, but those skilled in the art will understand that various modifications and other equivalent embodiments are possible therefrom.

Claims

A camera for photographing the front, side, and back of the subject to which the marker is attached;

A laser scaffold for displaying a foot position with a laser beam when measuring a subject;

A display for displaying the measurement process and measurement results; And

A body shape analysis device comprising calculation means for receiving a front, side, and rear image of a subject from the camera, recognizing a marker in a picture, analyzing muscle imbalance for each disease, and outputting an analysis result to the display.
The method of claim 1, wherein the calculation means

An image input unit for receiving image data captured from the camera;

Marker recognition unit for recognizing a marker in the input image,

A disease-specific evaluation unit for evaluating the muscle imbalance of the subject by analyzing the image in which the marker is recognized according to a predetermined disease-specific analysis algorithm,

A calibration information generation unit that generates customized information for correcting the imbalance of the measured person according to the evaluation result,

And an output unit for outputting body shape analysis results and body shape correction information.
The method of claim 2, wherein the calculation means

And a dynamic muscle imbalance evaluation means for receiving a video from the camera and analyzing dynamic muscle imbalance.
Receiving an image photographed from the front, side, and rear of the subject to which the marker is attached;

Recognizing a marker from the input image;

Analyzing the body shape of the subject according to a predetermined disease-specific evaluation algorithm based on the recognized marker; And

And outputting a body shape analysis result.
According to claim 4, Recognizing the marker is

Clicking a marker in a picture, generating a 51x51 pixel color mask around the click coordinates, processing in grayscale, and performing Sobel Mask edge detection; , Recognizing the learned support vector machine (SVM) kernel marker.
The method of claim 4, wherein analyzing the body type

At least one of muscle imbalance above the chest, pelvis imbalance, knee extremity, unbalanced development of the muscles around the knee, knee overextension, pelvic tilt, sway back, flat back, turtle neck round shoulder deformation, thoracic vertebrae, lumbar vertebrae Body type analysis method characterized in that the analysis according to the algorithm.