KR20160035497A

KR20160035497A - Body analysis system based on motion analysis using skeleton information

Info

Publication number: KR20160035497A
Application number: KR1020140127172A
Authority: KR
Inventors: 김광수
Original assignee: (주)이튜
Priority date: 2014-09-23
Filing date: 2014-09-23
Publication date: 2016-03-31

Abstract

The present invention relates to a body shape analysis system. More specifically, the body shape analysis system senses motions of a user during exercise or at normal times by using a Kinect which is a motion recognition apparatus, and extracts characteristic skeleton information corresponding to skeleton of a human body from the sensed motions to match the same with body information required for motion analysis. Therefore, the motion analysis-based body shape analysis system using extracted skeleton information may easily analyze motions and the body shape of the user by digitizing the same.

Description

TECHNICAL FIELD [0001] The present invention relates to a body analyzing system based on motion analysis using skeleton extraction information,

The present invention relates to a body-type analyzing system, and more particularly, to a body-type analyzing system that detects movement of a user during normal or motion using a motion recognition device Kinect, The present invention relates to a body-based analysis system based on motion analysis using skeleton extraction information, which enables a user to easily analyze the motion and body shape of a user by extracting information and matching the body information necessary for motion analysis.

In general, to analyze the body shape of the human body, the angle of the ankle is measured by observing the angle of the ankle from the back side, or the distance between the knee and the ankle is used as a scale for judging whether the leg is O-shaped or X-shaped. In addition, when looking at the whole body side, the body shape is analyzed by the degree of the inclination following the shoulder, pelvis, knee, and ankle based on the base of the ear, or the back of the body and the inclination of the left, right shoulder, pelvis, I calculate the degree of load and analyze the body shape.

Therefore, in the past, depending on the naked eyes of the observer who leaned against the wall of the grid pattern for analyzing the body, or analyzed the image taken by the camera, Methods for determining body shape and determining treatment are still being performed at many domestic and overseas clinics.

Thus, there is a problem in that when the observer or the doctor depends only on the judgment of the naked eye, there is a problem that the user can not grasp the part of the eye beyond the limit of the eye, and the judgment may be changed every time the examination is performed.

Accordingly, in recent years, it has become possible to obtain various indices by measuring the body shape of a person who has an abnormality in his or her body shape or during exercise using various expensive equipments, and also to quantify the various indices obtained by measuring the body shape In addition, research on various body shape analyzes has been carried out, which not only raised the objectivity of the body shape analysis but also made it possible to select an effective treatment plan most suitable for each body shape.

In order to analyze human body motion, 3D motion capture equipment such as Vicon Motion Analysis System was used to analyze human motion. However, this system consists of conventional equipment and software composed of infrared camera, infrared marker, and force measuring plate, which is a surface reaction force measuring device. In order to install this system, a large space has to be secured. There is a disadvantage that a professional manpower is required.

Accordingly, there is a growing need for a device that can more easily and inexpensively inspect patient behavior abnormalities than such conventional, expensive, and expensive equipment.

Korean Patent No. 10-1402781

The present invention utilizes Kinect, an inexpensive motion recognition device, to extract 20 skeletons corresponding to a human skeleton 30 times per second, and automatically matches body information necessary for motion analysis, And to provide a body composition analysis system based on motion analysis using skeleton extraction information that enables measurement and body composition analysis to be performed.

According to an embodiment of the present invention, there is provided a system for analyzing a body based on motion analysis using skeleton extraction information,

An operation recognition unit for sensing a movement of a user performing an operation to be analyzed; An operation analyzer for analyzing the posture of the body based on the body balance from the motion of the user acquired by the motion recognition unit; And a body type analyzing unit for storing body shape information analyzed by the motion analyzing unit, providing treatment information suitable for each body shape, and comparing and displaying body shape information before and after treatment.

At this time, the motion recognizing unit is constituted by a Kinect;

Wherein the motion analyzing unit includes: a skeleton extracting unit that extracts skeleton information from operation information formed by a motion of a user acquired by the motion recognizing unit; And a numerical analysis unit for digitizing and storing the image extracted by the skeleton extracting unit.

According to the present invention, it is possible to grasp a motion or an operation as a skeleton information by using a Kinect without using a complicated and expensive equipment.

In addition, the present invention enables to quantify, analyze and store images of a body or posture photographed using a Kinect, thereby improving the accuracy and promptness of diagnosis or evaluation of experts who analyze the body shape or motion using the measured information There is an effect that can be improved.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a motion analyzing based body composition analyzing system using skeleton extraction information according to the present invention. FIG.
FIG. 2 is a view showing an example of body composition analysis information to be extracted using skeleton extraction information and skeleton extraction information acquired using a Kinect according to the present invention; FIG.
FIG. 3 is a configuration diagram showing that motion analysis is performed by photographing motion during motion; FIG.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of a body composition analyzing system based on an operation analysis using skeleton extraction information according to the present invention. FIG. 2 is a block diagram of a body composition analyzing system using skeleton extraction information and skeleton extraction information Is a screen showing an example of body composition analysis information to be derived.

Referring to FIG. 1, the system for analyzing an operation analysis based on skeleton extraction information according to the present invention includes an operation recognition unit 100 for sensing a movement of a user performing an operation to be analyzed, A motion analyzer 200 for analyzing the posture of the body based on the balance of the user from the motion of the acquired user, and a motion analyzer 200 for storing the body information analyzed by the motion analyzer and providing treatment information suitable for each body, And a body type analyzer 300 for comparing and expressing the body type information.

Here, the motion recognizing unit 100 may be a sensor for detecting movement of a user, and may be configured as a depth-based camera or an infrared camera by photographing movement of a user and comparing image information with each other, It is preferable that it is composed of a Kinect which can easily grasp the movement and skeleton of the user and can be constructed at low cost.

The motion analyzer 200 includes a skeleton extractor 210 for extracting skeleton information from motion information obtained by the motion recognition unit and motion information of the user, And a numerical analysis unit 220 for storing the data.

Accordingly, it is possible to analyze the posture analysis and the range of the body based on the body balance (body balance / posture) based on the movement or operation information of the user acquired using the Kinect, It is possible to digitize and analyze images of the body shape and posture photographed by the kinetics.

The body shape and the image thus analyzed can be cumulatively stored through the history management, so that it is possible to support a systematic posture analysis. In the body shape analyzing unit, .

In addition, the motion analyzer 200 extracts 20 skeletons corresponding to the human skeleton 30 times per second using the user's operation information obtained by using the Kinect, matches the body information for motion analysis, .

Everyone from young children to adults has a musculoskeletal disease (posture disease), which is a little bit strange. However, it is common that recognition technology based on people or objects is progressing much research, but it is hard to find recognition based technology related to human posture.

FIG. 3 shows an example of using a depth-based camera or an infrared camera in connection with the recognition based on the recent posture. However, when such an expensive camera is used, there is a difficulty in requiring a high cost for constructing and maintaining the system. Accordingly, in the present invention, not only the attitude of the user is automatically measured using the kinect, but also the accuracy of the judgment can be improved by extracting the skeleton information and using it as the data of the behavior analysis and the body type analysis And the cost of building and maintaining the system can be reduced.

Accordingly, as shown in FIG. 2, skeleton extraction information according to a user's motion can be used to determine a head slope, a shoulder height difference, a pelvis height difference, a Toe in / out gap, HEAD Forward Posture Level, Kyphosis Level, Step Length, Step Length difference ratio, Arm Swing difference between right and left ratio can be analyzed and stored.

In addition, since the motion recognition section can perform moving image shooting and moving image comparison, comparison and analysis between dynamic positions can be performed.

As described above, the motion recognition unit can eliminate the hassle of attaching the sensor to the body as in the case of the conventional motion analysis system, and can provide an expensive sensor and sensor information processing device to the Kinect It can be constructed at low cost by the image processing apparatus, thereby ensuring cost competitiveness of the equipment.

In addition, since the Kinect recognizes the motion on the treadmill, such as the treadmill, the space can be reduced by about 1/4 or more as compared with the conventional motion sensor recognition system, and the installation convenience can be improved .

Also, in the present invention, when a kinetic camera is provided together with a camera so that motion and movement of a user can be sensed, a motion prescriber or a physician can perform a self motion through a video record . As a result, the treadmill, the Kinect, the cam camera and the console desk can be built as a system to experiment with accurate measurement values.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments. It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope of the present invention.

100 - Motion recognition unit (Kinect)
200 - Operation Analysis Unit 210 - Skeleton Extraction Unit
220 - Numerical Analysis Department
300 -

Claims

An operation recognition unit for sensing a movement of a user performing an operation to be analyzed;
An operation analyzer for analyzing the posture of the body based on the body balance from the motion of the user acquired by the motion recognition unit; And
And a body shape analyzing unit for storing body shape information analyzed by the motion analyzing unit, providing treatment information suitable for each body shape, and comparing and displaying body shape information before and after treatment. Based body shape analysis system.

The method according to claim 1,
Wherein the motion recognition unit comprises a Kinect;
Wherein the motion analyzer comprises:
A skeleton extracting unit for extracting skeleton information from the operation information formed by the motion of the user acquired by the motion recognizing unit; And
And a numerical analysis unit for digitizing and storing the image extracted by the skeleton extracting unit.