KR102564945B1 - Exercise management healthcare service system using tag - Google Patents

Abstract

An exercise management healthcare service system and method using a tag are disclosed.
The exercise management health care service system using tags has a web server and database for health, fitness, yoga, and sports exercises, and provides expert trainers' videos and exercise contents to user terminals for each exercise and expert trainer. Exercise management server providing care service; And it is connected to the exercise management server through a wired/wireless communication network, attaches a shiny tag attached to each part of the human body, and with respect to the exercise image of the user captured by a camera that exercises according to the trainer image, each part of the human body Recognizes the shiny tag attached to the tag, detects the motion using alpha blending, connects the joint point nodes of the trainer's exercise video and the user's exercise video image, and analyzes the exercise video to determine whether the motions of each part match. The measured points of the human body are compared with the solid line angle between the apex and the vertex of each joint, and the motion concordance between the trainer exercise image and the user's exercise video is scored and compared with each other, and the motion concordance score and motion concordance with the expert trainer image of the exercise A user terminal with an exercise app for health, fitness, yoga, crossfit, and sports exercise that outputs a scored graph and provides exercise management healthcare service for posture correction in terms of exercise dynamics of physical education science include

Description

Exercise management healthcare service system using tag {Exercise management healthcare service system using tag}

The present invention relates to a health care service system for exercise management, and more particularly, in terms of exercise dynamics of sports science during health, fitness, yoga, crossfit, and sports exercises using exercise equipment, a computer web or smartphone / tablet PC , The exercise app of the STB/TV display terminal is connected to the exercise management server through wired/wireless communication networks to provide the trainer's video and exercise contents to the user terminal, and each part of the human body (measurement points of the human body) ), tags or sensors attached to each part of the body (measurement points of the human body) for the exercise image of the user captured by the camera that watches and follows the image of the trainer , detects motion using alpha blending, detects motion of the spine and joints of the human body, detects the position of the exercise posture, calculates the amount of exercise (p = mv), and trains the exercise image of the trainer. Connects the joint point nodes of the user's exercise image and through comparative analysis of the exercise image, outputs a graph in which the motion match score and motion match score of the user image with the trainer image are scored, and provides exercise management healthcare service for posture correction It relates to an exercise management healthcare service system using tags.

The conventional exercise posture correction method used a method of correction using a three-dimensional motion sensor. A number of motion sensors were used for 3D motion capture, and a processor with excellent performance was required for the sensor module to process the large amount of data generated by the number of sensors. The method of correcting exercise posture using a 3D motion sensor required a complex system with high module cost because a large number of 3D motion sensors were required.

As a prior art 1 related to this, Patent Registration No. 10-2123743 "motion analysis method for fitness posture correction" is registered, and an exercise posture analysis method, a motion posture analysis device, and a recording medium are provided. The exercise posture analysis method receives a measurement signal from the textile sensor, extracts an effective area signal, which is a signal measured during the user's exercise, from the input measurement signal, and analyzes the user's exercise posture using the extracted effective area signal. Thus, the user's exercise posture can be more accurately analyzed and the wrong exercise posture can be corrected.

1 is a configuration diagram of a conventional exercise posture analysis device. 2 is a view showing the entire process of determining the accuracy of a user's exercise posture by a conventional exercise posture analysis method.

The exercise posture analysis method using the exercise posture analysis device is

Receiving a measurement signal from a fiber sensor; extracting an effective area signal, which is a signal measured when a user exercises, from the input measurement signal; and analyzing a user's exercise posture using the extracted valid area signal.

In the step of extracting the effective range signal, the effective range signal is extracted by using the effective peak and average period of the measurement signal change rate over time in the input measurement signal;

In the step of extracting the effective area signal, an effective peak, which is a peak having an average period among all peaks and an average period obtained by averaging peak intervals corresponding to a specific range of all peak intervals in a graph of change rate of the measurement signal over time is selected. doing; and

and extracting, as an effective area signal, a section from the first effective peak to the last effective peak until the lapse of the average period.

The fiber sensor is plural, and the step of extracting the effective area signal is a fiber sensor for detecting a user's movement among a plurality of fiber sensors by comparing each average energy of the measurement signal input from each fiber sensor with a specific threshold value. Select and extract the effective area signal from the measurement signal input from the selected fiber sensor.

As prior art 2 related to this, Patent Registration No. 10-1962578 "Fitness exercise service using VR" is registered.

As prior art 3 related to this, Patent Publication No. 10-2020-0055243 discloses “a method for providing a fitness scheduling service linked to an exercise test and a server device therefor”.

Recently, exercises performed in gyms, fitness, yoga, and sports exercises are performed by accessing an exercise management server through a wired or wireless communication network using a web or mobile terminal to exercise in terms of sports science and exercise mechanics. You are providing content.

However, in the case of health, fitness, yoga, and sports exercise, the existing exercise management system provides the trainer's video and exercise contents to a smartphone/tablet PC from the exercise management server, and attaches shiny tags to each part of the human body. It is attached to provide the user's movement image captured by the camera, detect the movement of the spine and joints of the human body by motion recognition using alpha blending, detect the position of the movement posture, and capture the movement image with the camera for analysis. Comparison and analysis of the user's exercise video with the trainer's video, linking the joint nodes of the trainer's exercise video and the user's exercise video, and comparing them through image analysis to express a graph in which the motion coincidence is scored, and analyzing big data during exercise It did not provide exercise management healthcare services and home training services that visualize data and provide exercise-related big data analysis and data visualization, time-series statistical data in bar graphs and line graphs, and posture correction during exercise.

Patent Registration No. 10-2123743 (Registration Date: June 10, 2020), "Motion Analysis Method for Fitness Posture Correction", Electronic Components Research Institute Patent Registration No. 10-1962578 (registration date March 20, 2019), "Fitness Exercise Service Provision System Using VR", Jong-Beom Kim Patent Publication No. 10-2020-0055243 (published on May 21, 2020), "Method of providing fitness scheduling service linked to exercise test and server device therefor", Pete Co., Ltd.

An object of the present invention to solve the above problems is a computer web or smartphone/tablet PC, STB/TV display terminal in terms of exercise dynamics of sports science during health, fitness, yoga, crossfit, and sports exercise using exercise equipment. The exercise app (App) accesses the exercise management server through a wired/wireless communication network to provide the trainer's video and exercise contents to the user terminal, and attach tags or sensors to each part of the human body (measurement points of the human body). Recognize tags or sensors attached to each part of the body (measurement points of the human body) of the exercising person in relation to the exercise image of the user captured by the camera and exercise along with the video of the trainer, and perform alpha blending (alpha blending) is used to detect the motion of the spine and joints of the human body, detect the position of the posture of the motion, calculate the amount of exercise (p = mv), and compare the trainer's motion image and the user's motion image Exercise using tags that connects joint point nodes and compares and analyzes exercise images, outputs a graph in which the motion match score and motion match score of the user image with the trainer image are scored, and provides exercise management healthcare services for posture correction Provides a managed healthcare service system.

In order to achieve the object of the present invention, an exercise management healthcare service system using tags includes a web server and a database for health, fitness, yoga, and sports exercises, and includes expert trainers' images and exercise contents for each exercise and expert trainer. An exercise management server that provides an exercise management healthcare service by providing a user terminal; And it is connected to the exercise management server through a wired/wireless communication network, attaches a shiny tag attached to each part of the human body, and with respect to the exercise image of the user captured by a camera that exercises according to the trainer image, each part of the human body Recognizes the sparkling tag attached to and detects the motion using alpha blending, connects the joint point nodes of the trainer's motion image and the user's motion image image, and determines whether the motions of each part match through motion image analysis. The measured points of the human body are compared with the solid line angle between the apex and the vertex of each joint, and the motion consistency of the trainer's motion image and the user's motion image are scored and compared with each other, and the motion coincidence with the expert trainer's image of the corresponding exercise It includes a user terminal in which an exercise app for health, fitness, yoga, crossfit, and sports exercise, which outputs a graph in which scores and motion coincidences are scored, is installed.

The exercise management healthcare service system using the tag of the present invention is a computer web or smartphone/tablet PC, STB/TV in terms of exercise dynamics of sports science during health, fitness, yoga, crossfit, and sports exercises using exercise equipment. The exercise app on the display terminal connects to the exercise management server through wired/wireless communication networks to provide the trainer's video and exercise contents to the user terminal, and tags or Recognize tags or sensors attached to each part of the human body (measurement points of the human body) in relation to the user's exercise image captured by the camera that attaches the sensor and exercises according to the trainer's image, Detect motion using alpha blending to detect the movement of the spine and joints of the human body, detect the position of the motion posture, calculate the amount of motion (p = mv), and use the trainer's motion image and the user's motion By connecting the joint point nodes of the image and through comparative analysis of the exercise image, the motion matching score of the user image with the trainer image and the motion matching score are output, and the exercise management healthcare service for posture correction is output. .

Fitness using exercise equipment, PT (Personal Training), fitness, yoga, crossfit, sports exercise, in terms of exercise mechanics in physical education science, equipment exercise in the gym, lifting weights, push-ups based on images of expert trainers It is used for straightening, fitness movement, yoga posture correction, tennis posture correction, and screen golf posture correction, and provides exercise management healthcare service while the user watches the smartphone and TV display with the exercise app installed.

1 is a configuration diagram of a conventional exercise posture analysis device.
2 is a view showing the entire process of determining the accuracy of a user's exercise posture by a conventional exercise posture analysis method.
3 is a block diagram of an exercise management healthcare service system using a tag.
4 is a block diagram of an exercise management healthcare service system using a tag according to the present invention.
5 is a main screen and a login screen of a fitness exercise application (App) of a smart phone according to an embodiment of the present invention.
6 is a start screen and a challenge screen of a fitness exercise application (App) of a smartphone.
7 is an exercise selection screen and a trainer selection screen of a fitness exercise application (App) of a smartphone.
8 is an exercise image screen of a fitness exercise application (App) of a smart phone and a comparison screen (I) with a trainer image.
9 is an exercise image screen of a fitness exercise application (App) of a smartphone and a comparison screen (II) with a trainer image.
10 is an exercise image screen of a fitness exercise application (App) of a smart phone and a comparison screen (III) with a trainer image.
11 shows the test results of the weight measurement and body composition detector (In Body scale), skeletal muscle fat analysis (weight, muscle mass, body fat mass), obesity analysis (BMI, body fat rate), and visceral fat analysis (visceral fat level) with measured values and bars This is a screen displayed in the exercise app of the user terminal in the form of a graph (bar chart).
Figure 12 is a line chart of the body balance change process of measured values of body weight, muscle mass, body fat mass, and body fat percentage according to time of day/week/month as a result of examination of body weight measurement and body composition detection device (In Body Scale). Expressed as , weight measurement and body composition detection device (In Body scale) scores (In Body scores) are displayed on the standard normal distribution curve, and weight measurement of a sample of a population having a standard body shape of a certain number of people measured in advance and the top percentage in the statistical graph having a normal distribution by the average of body composition detection device (In Body scale) scores.
13 is a diagram according to neck joints/trunk joints/elbow joints/knee joints of user images attached with shiny tags or sensors attached to each part of the human body based on trainer images during health, fitness, yoga, and sports exercises. It is a diagram explaining the measurement principle of the motion agreement score of the torso and limb motion model with the trainer image.
14 connects the apex of the joint of the human body measurement points in the exercise image of the user who watches and follows the exercise image of the expert trainer during health, fitness, yoga, and sports exercise and recognizes the exercise action. It is a picture.

Hereinafter, the configuration and operation of a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the description of the present invention, if it is determined that a detailed description of a related known function or known configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. In addition, the attached drawing numbers refer to the same drawing numbers in other drawings when indicating the same configuration.

In gyms, fitness, yoga, and crossfit, exercise management servers that provide home training and healthcare services and computers connected through wired/wireless communication networks, smartphones/tablet PCs, and TV displays use exercise applications to select / exercise Once a trainer is selected,

The exercise management server is an exercise application (App) on the user terminal. It provides exercise information and exercise contents including video, audio, image, and text related to basic exercise postures, and images of expert trainers for each exercise, and the human body of a person exercising attaching a shiny tag or sensor to each part (measurement points of the human body), and after selecting an exercise/trainer, taking an exercise image of a user exercising according to the video of an expert trainer with a camera of a user terminal; By using the exercise analysis program of the exercise scientific exercise app that analyzes the user's exercise image attached with the tag that reflects each part of the exerciser's body, the motion using alpha blending is detected and the spine of the human body Detect the movement of the joint and the position of the movement, and calculate the momentum (p = mv); - If necessary, enlarge/reduce the video/photo by scaling the user's video according to the standard size (width x height) of the expert trainer's video, - Joint joints between the trainer's exercise video/photo and the user's exercise video/photo image Through image analysis, the motion consistency of user images is scored based on the trainer image and compared with each other, and alpha blending accuracy is determined through comparative analysis of the motion image of the user image based on the expert trainer image for each spine/joint node. It calculates and provides motion concordance scores and motion concordance graphs, provides exercise management healthcare services for posture correction, and is used for health, PT (Personal Training), fitness, yoga, and sports exercises using exercise equipment. .

According to the motion model of the exercise motion, the motion consistency of the user image based on the trainer image is compared with the angle of the solid line connected between the vertices of each joint of the measurement points of the human body, and the motion consistency of the user image based on the trainer image is are scored

For example, in terms of exercise mechanics in physical education science, gym equipment exercise, weight lifting, push-up, fitness motion, yoga posture correction, tennis posture correction, and screen golf posture correction based on the images of expert trainers in the exercise used, etc.

It uses shiny tags or sensors attached to each part of the body such as wrists, shoulders, limbs and torso, knees, and thighs of the exercising person, and the sparkling tags attached to the body parts of the user's exercise video taken by the camera are used. It recognizes tags, detects motions using alpha blending, detects motions of the spine and joints of a person exercising, and recognizes the posture and position of an exerciser.

As the sensor, a plurality of three-dimensional motion detection sensors provided for each part of the human body, or a wrist-worn wearable band having a three-axis acceleration sensor, a control unit, and a communication unit may be used, and Bluetooth (BLE) as a user terminal synchronized with settings during exercise Alternatively, 3D motion detection data or 3-axis acceleration data is transmitted through Wi-Fi communication.

Provides an exercise management healthcare service that detects the motion of the exercise video and provides posture correction of the human body motion of the exercise through motion analysis. Health using exercise equipment, PT (Personal Training), fitness, yoga, crossfit , used for sports exercise.

Exercise includes, but is not limited to, gym exercise (treadmill running, weight lifting, machine exercise, etc.), fitness, yoga, crossfit, sports exercise (table tennis, tennis, golf, taekwondo, judo, kendo, etc.) A variety of exercises can be used.

Fitness (physical fitness) provides a fitness exercise program consisting of whole-body aerobic exercise using all limbs, and provides exercise for a healthy body by promoting blood circulation through movement of the spine and muscles.

For reference, yoga is the movement of the joints of the neck/shoulder/wrist/limb/spine by stretching of the spine, muscles and joints with movement of the spine, muscles and joints through deep abdominal breathing, according to the yoga movement and breathing method of the correct posture. It is an exercise to keep the balance of the body according to the resistance and stretching of the body and properly align the muscles and joints.

Yoga, which is based on an eight-aspect approach to reaching the highest state of yoga in Ashtanga yoga practice, means eight, multiple limbs, as recorded in the Yoga Sutras of Patanjali. Yama (abstinence), Niyama (contemplation), Asana (posture), Pranayama (breathing), Pratihara (sensory control), Dharana (concentration), Dhyana (meditation), Samadhi (samadhi) . The practice of eight posture exercises (asana) is said to establish yoga movements for the correct practice of pranayama (breathing).

Breathing methods include Ujjayi breathing, Lion's breathing, and deep stomach breathing.

Yoga practice posture practice (asana) of yoga movement, correct posture and breathing method of pranayama (breathing), correct posture according to movement of joints of spine/neck/wrist/limbs, resistance and stretching of muscles to improve sense of balance. It trains the human body to improve joint discomfort, remove chronic muscle pain and toxins from the body, and improve blood circulation according to the movement of the spine and joints to maintain a healthy body.

Yoga movements correct body shape, release stiff muscles and joints, strengthen weakened muscles, and restore flexibility to have a healthy body and mind. For example, a person who is accustomed to "leaning forward posture" can use "backward posture", "backward rotation of the head and neck", and "left and right side stretching" to correct posture and body shape such as curled shoulders and a twisted spine. It relieves tension in the muscles of the head, neck and shoulders.

Yoga is a warm-up (relaxing the neck, shoulders, arms, elbows, wrists, fingers, hip joints, knees, and toes in order) and breathing methods. holder openers, hip and leg flexors, back bends, core strengtheners, arm balances and inversions, twists Twists and the like provide yoga movements.

The forward flexion action stretches the back of the body from the head to the heel in a forward bending action at the point where the hip joint is folded.

There are dog posture, cow head posture, and back-to-back posture to open the shoulders of yoga, and it expands and relaxes muscles to reduce chronic muscle pain and joint discomfort, releasing tension in the muscles of the head, neck and shoulders.

Backbending is a yoga movement that strengthens the back of the body of the muscles surrounding the legs, hips, ribs, and upper body. It relieves fatigue, relieves fatigue, removes toxins from the body, improves blood circulation, and improves the quality of sleep.

Hereinafter, the configuration and operation of a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

3 is a block diagram of an exercise management healthcare service system using a tag.

Exercise management healthcare service system using the tag of the present invention

For health, fitness, yoga, and sports exercises, a web server, web application server (WAS), and database (DB) are provided, and video of expert trainers, basic exercise information, and exercise contents are provided to user terminals for each exercise and expert trainer. an exercise management server 200 that provides exercise management healthcare services; and

A user who is connected to the exercise management server 200 through a wired/wireless communication network, attaches a shiny tag or sensor to each part of the body of a person exercising, selects an exercise/trainer, and then watches an expert trainer's video and exercises along the way. The motion image is captured by the camera of the user terminal, and the sparkling tags or sensors attached to each part of the human body (measurement points of the human body) are recognized for the motion image of the user captured by the camera, and alpha blending is performed. Detects the motion using , detects the position of the exercise posture, calculates the amount of exercise (p = mv), connects the joint point nodes of the image of the exercise video of the user who exercises according to the expert trainer's exercise video, and analyzes the exercise image Through this, whether the motions of each part match is compared with the solid line angle between the apex and the vertex of each joint of the measurement points of the human body, and the motion coincidence between the trainer motion image and the user's motion image is scored (ex. motion coincidence 3 points, Approximately match partial motion - 2 points, motion mismatch - 1 point, motion completely inconsistent - 0 points) and compare each other, and when motion fitting is clicked, it takes the same motion as the opaque fitting silhouette shown on the camera screen, During motion, it outputs a graph in which the motion coincidence score of the motion motion of the user image with the expert trainer image of the exercise displayed according to the degree of agreement with the silhouette and the motion coincidence score are displayed, and the motion dynamics of physical education science is client/server method On the side, it includes user terminals 100, 110, and 130 in which exercise apps for health, fitness, yoga, crossfit, and sports exercises that provide exercise management healthcare services for posture correction are installed.

During exercise, the user attaches a shiny tag or sensor to "measurement points of the human body" and is photographed with a camera to provide a user image,

In an embodiment, when the sensor is attached to measurement points of the human body, the sensor uses a 3D motion detection sensor.

The measurement points of the human body are the head 100, the left shoulder 110, the left elbow joint 111, the left forearm joint 112, the right shoulder 120, the right elbow joint 121, the right forearm joint ( 122), navel 130, left hip 140, left knee joint 141, left ankle joint 142, right hip 150, right knee joint 151, and right ankle joint 152 includes

The user terminals 100, 110, and 130 include a computer 100, a smartphone or tablet PC 110 on which an exercise app for health, fitness, yoga, crossfit, and sports exercise using exercise equipment is installed, or The STB/TV display 130 is used, and in the embodiment, “fitness” of a smartphone with an exercise app installed thereon will be described as an example.

The exercise management server 200 includes a web server 210, a web application server (WAS) 220, and a database (DB) 230.

A camera 274 that captures an exercise image of an expert trainer; an A/V encoder 272 encoding the expert trainer's exercise video and audio data; and a streaming server 270 providing the user terminals 100, 110, and 130 with the exercise video and audio data of the expert trainer for each exercise in a streaming manner.

It is connected to the controller of the exercise management server 200, transmits and receives chat data through a user terminal performing health, fitness, yoga, crossfit, and sports exercises, and an expert trainer terminal and the exercise management server 200, and is an expert in the exercise. It further includes a chat server providing trainer feedback data.

Exercises performed in gyms, fitness, yoga, crossfit, and sports are exercised through wired/wireless communication networks through the web of computers, smart phones/tablet PCs, and exercise apps on STB/TV display terminals in terms of sports science and exercise mechanics. It is connected to the management server, and the exercise management server 200 provides the expert trainer's image, exercise information, and exercise contents to the user terminal, attaches a shiny tag or sensor attached to each part of the body of the person exercising, After selecting the exercise/trainer, the user's exercise video is taken with the camera of the user terminal along with the video of the expert trainer, and a shiny tag or sensor attached to each part of the human body is used for the user's exercise video captured by the camera. It detects motion using alpha blending to detect the movement of the spine and joints of the human body, detects the position of the posture of the exercise, calculates the amount of exercise (p = mv), and calculates the exercise image of the trainer and the exercise image of the user. Connect the joint point nodes of the image and compare each other through motion image analysis. When motion fitting is clicked, it takes the same motion as the opaque fitting silhouette shown on the camera screen, and the corresponding motion displayed according to the degree of matching with the silhouette. It outputs the motion match score of the user image with the expert trainer video and a graph in which the motion match score is scored, and provides exercise management healthcare service for posture correction, such as health, PT (Personal Training), fitness, yoga, crossfit, sports used for exercise

According to the motion model of the exercise motion, the motion consistency of the user image based on the trainer image is compared with the angle of the solid line connected between the vertices of each joint of the measurement points of the human body, and the motion consistency of the user image based on the trainer image is are scored

Additionally, the exercise application (App) of the user terminal (smartphone) is connected to the exercise management server 200 through a wired or wireless communication network, and the user and the height and weight before and after exercise are measured and measured by the body composition detection device (In Body Scale) Measured values such as the user's weight, muscle mass, and body fat mass are stored in the exercise management server 200, and accumulated and stored in the database (DB) of the exercise management server 200 for each time slot/day/week/month/year. Individual exercise statistical data including weight, exercise amount, muscle mass, body fat mass, and BMI value are provided to the user terminal.

The exercise app of the user terminal is linked with the user's height stored in the exercise management server and the weight, muscle mass, body fat mass, and BMI measurement values of the weight measurement and body composition detection device (In Body Scale),

Weight measurement and body composition detection device (In Body scale) test results, skeletal muscle fat analysis (weight, muscle mass, body fat mass), obesity analysis (BMI, body fat percentage), and visceral fat analysis (visceral fat level) chart) displayed on the exercise application (App) of the user terminal (smartphone);

The body balance change process over time of body size (height), weight, muscle mass, body fat mass, body fat percentage, and BMI measurements is expressed in the form of a line chart, and the standard normal distribution curve is used to measure body weight and body composition detectors ( In Body scale) scores are displayed, and statistics having a normal distribution by the average of weight measurement and body composition detector (In Body scale) scores of a sample of a pre-measured population with a standard body shape of a certain number of people The top percentage of the analysis data is displayed on the exercise application (App) of the user terminal (smartphone).

4 is a block diagram of an exercise management healthcare service system using a tag according to the present invention.

The exercise app (App) of the user terminal connects to the exercise management server and manages exercise for health, fitness, yoga, crossfit, and sports exercise. After registering member information, user authentication, exercise selection/trainer selection, the human body Shiny tags or sensors are attached to each part of the human body (measurement points of the human body), and the glitter attached to each part of the human body for the user's exercise image captured by the camera watching the video of the expert trainer and exercising along the way recognizes tags or sensors, detects motions using alpha blending, detects the movement of the spine and joints of the human body, detects the position of the exercise posture, calculates the amount of exercise (p = mv), and trains the exercise image of the trainer. Connecting the joint point nodes of the image of the user's exercise video, and through the motion image analysis, whether the motion of each part matches is compared with the solid line angle between the apex and the vertex of each joint of the "measurement points of the human body" to obtain a trainer image Based on the score, the degree of motion matching of user images is scored and compared with each other, and when motion fitting is clicked, it takes the same motion as the opaque fitting silhouette shown on the camera screen, and is displayed according to the degree of matching with the silhouette It outputs the motion match score and the motion match graph with the expert trainer video of the exercise, and provides exercise management healthcare service for posture correction in terms of exercise dynamics of physical education science.

According to the motion model of the exercise motion, the motion consistency of the user image based on the trainer image is compared with the angle of the solid line connected between the vertices of each joint of the measurement points of the human body, and the motion consistency of the user image based on the trainer image is are scored

During exercise, the exercise application (App) of the user terminal recognizes the motion behavior of the user image exercising along with the trainer image, detects the pose and motion action, and uses the action recognition posenet algorithm Measurement points of the human body may be displayed by connecting them with solid lines. The Posenet algorithm detects the position of the joint feature points (measurement points of the human body) of the main parts of the human body with respect to the position of the head, body, arms and legs of the human body, and the pose of a person moving from an image or video. posture) and used as a model for action recognition.

)를 측정하여 운동 어플(App)로 표시하고 운동 관리 서버에 저장하는 기능 더 포함한다. The exercise app of the user terminal determines the amount of exercise (p = mv) and kinetic energy (

) is measured, displayed as an exercise app, and stored in an exercise management server.

In addition, the exercise management server 200 includes a web server 201 connected to a user terminal through a wired or wireless communication network; A control unit 202 that controls exercise management healthcare services for health, fitness, yoga, crossfit, and sports exercises; A member management unit 203 connected to the control unit 202 and receiving and managing exercise management health care service member information registered; A user authentication unit 204 connected to the control unit 202 and authenticating a user by ID/Passwd or email/password method; an exercise content provider 205 connected to the controller 202 and providing expert trainer images, exercise information, and exercise contents for each exercise; It is connected to the control unit 202 and receives measurement values including the user's height, weight, muscle mass, and body fat mass measured by the body weight measurement and body composition detection device (In Body Scale) from the exercise application (App) of the user terminal. weight and body composition information management unit 206 to store; a user exercise image storage unit 207 connected to the control unit 202 and storing a user exercise image received from an exercise application (App) of a user terminal; It is connected to the controller 202, and the trainer image/user image comparison and analysis function of the exercise application (App) of the user terminal is applied to the server-side in the same way, so that the user's exercise image captured by the camera is applied to the user's exercise image. It recognizes the shiny tag attached to each part of the human body, detects motion using alpha blending, connects the joint point nodes of the trainer's exercise video and the user's exercise video image, and checks whether the motion matches through motion image analysis. A trainer image/user image comparison and analysis unit 208 that compares and analyzes the expert trainer image and the user image by scoring (match score: motion match 3 points, motion partial match - 2 points, motion mismatch 0 points); Measured values including the user's height, weight, muscle mass, body fat mass, and BMI values measured by weight measurement before and after exercise and body composition detection device (In Body scale) are stored in the exercise management server 200, and the exercise management server ( 200) in the database (DB) 212 and accumulated and stored to provide individual exercise statistical data including exercise amount, muscle mass, body fat mass, and BMI value by time period/day/week/month/year Statistical information providing unit 211 ; A database (DB) 212 that stores expert trainer images, exercise information, and exercise contents for each exercise, and stores user exercise images and measurement values including the user's height, weight, muscle mass, body fat mass, and BMI value, and each exercise field It includes a bulletin board 213 for registering writings of expert trainers and users.

)를 측정 저장하고 사용자 단말의 운동 어플(App)로 표시하는 운동량 검출부를 더 포함한다. The exercise management server 200 is connected to the control unit 202, and during machine learning running exercise at the gym, when a user with m weight runs at v speed, the exercise amount (p = mv) and exercise energy (

) and further includes an exercise amount detection unit for measuring and storing and displaying the exercise application (App) of the user terminal.

A flashing tag or sensor attached to each part of the human body for the image of a user photographed with a camera of a person exercising and exercising according to the trainer image by the exercise scientific exercise analysis program of the exercise app User who watches and follows the trainer's exercise video to detect the exercise motion over time using alpha blending, detect the motion of the spine and joints of the human body, and compare and analyze the trainer image and user image. Connect the joint point nodes of the image of the exercise image, score whether the motion picture of each part motion model matches through motion image analysis, compare each other, and compare the motion match score based on the trainer image of the exercise and a graph in which the motion consistency score is scored Provides, additionally calculates and displays the amount of exercise (p = mv) of the user's exercise image, and measures data such as body weight and muscle mass and exercise data of the body weight measurement and body composition detection device (In Body scale) during exercise and exercise data of the user terminal It is stored in the exercise management server 200 through the app to accumulate and store big data analysis and exercise data in the database of the exercise management server to provide individual exercise statistical data by day/week/month/year, and data visualization It is visually displayed as a bar chart, pie chart, or line chart.

When the mass is m and the velocity is v, the momentum p is calculated as p = mv.

로 계산된다. When the mass is m and the velocity is v, the kinetic energy is

is calculated as

)를 측정하여 이를 머신러닝 장비와 BLE 또는 Wi-Fi 근거리 무선 통신을 통해 스마트폰의 운동 어플(App)에 표시할 수 있다. For example, during machine learning running exercise at the gym, the amount of exercise (p = mv) and kinetic energy (

) can be measured and displayed on the exercise app of the smartphone through machine learning equipment and BLE or Wi-Fi short-range wireless communication.

The sensor may use a plurality of three-dimensional motion detection sensors provided for each part of the human body, or a wrist-worn wearable band having a three-axis acceleration sensor, a control unit, and a communication unit, and Bluetooth (BLE) as a user terminal synchronized with settings during exercise ) or 3D motion detection data or 3-axis acceleration data is transmitted through Wi-Fi communication.

Alpha blending is an effect that makes an image appear transparent when another image is overlaid on top of it.

[How to use alpha blending]

(1) Run the exercise management application (App) on the user terminal (smart phone).

(2) After selecting health, fitness, yoga, crossfit, sports exercise/trainer, click Motion Fitting.

(3) Take the same action as the opaque fitting silhouette shown on the camera screen.

(4) Scores are shown according to the degree of agreement with the silhouette during motion.

Through this, difference data on the motion of the model of the user image and the trainer image is obtained, and the posture shown in the professional trainer image and the joint point nodes of the user image are connected, and each movement motion of the expert trainer image and the motion model of the user image is connected. The accuracy of the user's movement is judged by scoring the motion matching degree to determine whether the motion pictures of each partial motion model match through image comparison analysis for each part.

The data that can be applied to the exercise management app of smart devices such as smartphones and tablet PCs is used as follows.

1. Data that can be applied to exercise apps on smart devices

[Items required for data analysis]

ㆍ Body size: height, weight, muscle mass, bone mass, body water, body fat, basal metabolic rate, BMI, visceral fat peristalsis)

ㆍ Height: Assuming that there is no change even if exercise is based on an adult standard -> Fixed value, input item

ㆍ If you enter your height and age, you can find out the standard weight and obesity level for your body type according to your physical condition.

ㆍ Cortisol level: survey results, mainly used for stress level analysis, muscle fatigue analysis after exercise

ㆍ Exercise video data: Comparative analysis with the posture shown in the professional trainer's video

[Features required by the app]

2.1 Implementation of the investigated body size graph

- Compare your body figure measured by weight measurement and body composition detection device (In Body scale) with the average through a graph

- The user displays the change process of his or her body balance in a graph in units of weeks/months/years

- Weight measurement and body composition detector (In Body scale) test results, skeletal muscle fat analysis (weight, muscle mass, body fat mass), obesity analysis (BMI, body fat percentage), and visceral fat analysis (visceral fat level) are displayed in a bar chart expressed in the form

- Expresses the body balance change process of weight, muscle mass, body fat mass, and body fat percentage in the form of a line chart, displays weight measurement and body composition detector (In Body scale) scores on the standard normal distribution curve, and displays the top percentage

(Add-ons)

- If above or below the average for each item -> Provide the necessary diagnosis to the user (theory, exercise method, etc.)

Example) If body fat level is above average: notification and exercise method and diet management to reduce fat are provided

Example) Less muscle mass: muscle loss warning!

- When daily exercise amount or target value (muscle mass, weight, etc.) is achieved, a kind of reward such as a badge is provided

2.2 Implementing a graph that scores motion consistency

- Motion detection using alpha blending, linking the joint nodes of the trainer's exercise video and the user's exercise video, and comparing them through image analysis

- Motion match graph: Displays the motion match score in the form of a line graph in real time as the exercise progresses, and can be reviewed after checking the parts that do not match the trainer's motion by time period

Example) If the trainer's exercise video is 5 minutes in total, and the graph section from 1 minute to 1 minute 20 seconds is 60 points (60% match), after reviewing the exercise again for that time period, the average score is a certain score (ex: 70 points), encourage them to exercise again later

2.3 Exercise part recommendation function

Provides a system that recommends the areas to be exercised with the highest priority and exercise images using the user's body information (height, weight).

Using a hybrid recommendation system -> Recommending new users by analyzing them through content-based filtering, and increasing the accuracy of recommendations through collaborative filtering after the user's exercise data is sufficiently accumulated.

In fact, there are many people who do not know how to set up an exercise routine, and if you set up a routine by yourself, there are cases where the body is overloaded by focusing only on specific muscles, so the trainer's feedback is intensively reflected.

5 is a main screen and a login screen of a fitness exercise application (App) of a smart phone according to an embodiment of the present invention. 6 is a start screen and a challenge screen of a fitness exercise application (App) of a smartphone.

7 is an exercise selection screen and a trainer selection screen of a fitness exercise application (App) of a smartphone.

8 is an exercise image screen of a fitness exercise application (App) of a smart phone and a comparison screen (I) with a trainer image. 9 is an exercise image screen of a fitness exercise application (App) of a smartphone and a comparison screen (II) with a trainer image. 10 is an exercise image screen of a fitness exercise application (App) of a smart phone and a comparison screen (III) with a trainer image.

11 is a test result of a weight measurement and body composition detection device (In Body scale), and the measured values are stored in an exercise management server, and skeletal muscle fat analysis (weight, muscle mass, body fat mass), obesity analysis (BMI, body fat rate), and visceral fat analysis This is a screen displayed in the exercise app of the user terminal in the form of a measured value and a bar chart (visceral fat level).

)으로 나눈 값을 통해 지방의 양을 추정하는 비만측정법에 의해 계산된다. For reference, the weight measurement and body composition detection device (In Body Scale) measures height, weight, muscle mass, skeletal muscle mass, and body fat mass, uses Bioelectrical Impedance Analysis (BIA), and the human body is electrically It is made of moisture that conducts electricity well, and the degree to which electricity passes and the resistance varies depending on the amount of moisture. BIA is a method of measuring body water with an impedance value obtained by sending a fine alternating current through the human body. The human body has different lengths and cross-sectional areas for each part, and body weight measurement and body composition detectors (In Body Scales) measure arms, legs, and torso separately for each part using simultaneous multi-frequency impedance measurement. Arms and legs have a small area and a high impedance value, but a small amount of muscle. On the other hand, the torso has a relatively large area, so the impedance value is low and the muscle mass is large. Therefore, a small impedance change in the trunk has a great effect on the muscle mass, and must be separately measured to accurately measure the total muscle mass. In addition, the body weight measurement and body composition detection device (In Body Scale) can measure BMI by height (height) and measured weight. BMI (body mass index, body mass index) is the weight (kg) divided by the square of the height (

) is calculated by a barometric method that estimates the amount of fat through the value divided by

[Measurement items of weight measurement and body composition detector (In Body Scale)]

ㆍ Body composition analysis: body weight, body water, muscle mass, body fat mass

ㆍ Skeletal muscle, fat analysis (graph): body weight, skeletal muscle mass, body fat mass

ㆍ Obesity diagnosis: BMI, body fat percentage

ㆍ Impedance by part/frequency, muscle analysis by part, body fat analysis by part

ㆍ Body changes: weight, skeletal muscle mass, body fat percentage (results of 8 measurements)

ㆍ Provides weight measurement and body composition detection device (In Body scale) score, body type check, and weight control function

12 is a line chart showing the body balance change process of the measured values of body weight, muscle mass, body fat mass, and body fat percentage as a result of examination of the body weight measurement and body composition detection device (In Body Scale), and the measured values are stored in the exercise management server. ) form, the weight measurement and body composition detection device (In Body scale) scores are displayed on the standard normal distribution curve, and the weight measurement and body composition of a sample of a population with a pre-measured standard body shape of a certain number of people This is the screen displayed on the exercise app of the user's terminal (smartphone) to determine the top percentage in the statistical graph having a normal distribution by the average of the detection device (In Body scale) scores.

By referring to the user's height stored in the database of the exercise management server, the weight before and after exercise is measured and the body composition detection device (In Body Scale) is used together, and the user's weight, muscle mass, body fat mass, and BMI value are measured. Detecting device (In Body scale) test execution results, skeletal muscle fat analysis (weight, muscle mass, body fat mass), obesity analysis (BMI, body fat percentage), and visceral fat analysis (visceral fat level) are user-defined as measured values and bar charts displayed in the exercise application of the terminal; The body balance change process of measured values of weight, muscle mass, body fat mass, and body fat percentage is expressed in the form of a line chart. In statistical analysis data having a normal distribution by the average of weight measurement and body composition detection device (In Body scale) scores of a sample of a population with a certain number of standard body types, display the top percentage in the normal distribution graph User terminal (smartphone) ) is displayed in the exercise app.

13 is a diagram according to neck joints/trunk joints/elbow joints/knee joints of user images attached with shiny tags or sensors attached to each part of the human body based on trainer images during health, fitness, yoga, and sports exercises. It is a diagram explaining the measurement principle of the motion agreement score of the torso and limb motion model with the trainer image.

14 connects the apex of the joint of the human body measurement points in the exercise image of the user who watches and follows the exercise image of the expert trainer during health, fitness, yoga, and sports exercise and recognizes the exercise action. It is a picture.

During health, fitness, yoga, and sports exercises, joint point nodes of the expert trainer's exercise video and the user's exercise video image are connected, and through exercise image analysis, it is determined whether the motion picture of each part motion model matches. The angles of the connected solid lines between the vertices of the joints are compared and scored, and mutual comparisons are made. According to the classified exercise motions, based on the trainer image, the motion consistency of the motion model of the human body of the user image is between the vertices of each joint. The motion consistency is scored by comparing with the angle of the connected solid line.

During health, fitness, yoga, and sports exercises, neck motion, torso and limb motion are performed according to the neck joint/trunk joint/elbow joint/knee joint, and the measurement points of the human body are the head (100) and the left shoulder. (110), left elbow joint (111), left forearm joint (112), right shoulder (120), right elbow joint (121), right forearm joint (122), belly button (130), left hip (140) , including the left knee joint 141, the left ankle joint 142, the right hip 150, the right knee joint 151, and the right ankle joint 152, based on the trainer image, the human body Depending on the neck joint/trunk joint/elbow joint/knee joint of the user image to which shiny tags or sensors attached to measurement points are attached, the body and limb motions are the left elbow joint 111 and the right elbow joint 121. ), based on the left knee joint 141 and the right knee joint 151, the vertex of each joint is compared with the solid line angle between the vertices, and a motion consistency score of the exercise motion of the user image is given based on the trainer image (Criterion 1,2,3,4,5).

* Measurement points

Criterion 1 left arm 110,111,112

Criterion 2 right arm 120,121,122

Criterion 3 left foot 140,141,142

Criterion 4 right foot 150,151,152

Standard 5 body 100, 130

Criterion 6 Upper body area 100,110,120

Criterion 7 lower body area 130,140,150 (hip joint)

Each expert trainer and user have different heights and weights, and upper body parts 100 , 110 , and 120 and lower body parts 130 , 140 , and 150 are different.

During exercise, the exercise application (App) of the user terminal sees the trainer image of the exercise motion, detects the pose and exercise action in real time of the user image exercising along, and uses the action recognition posenet algorithm. Thus, it is possible to connect measurement points of the human body with solid lines and display them visually.

Exercise motions include neck motion, arm motion, and leg motion along with trunk motion. ), an exercise operation is performed according to the right knee joint 151.

For example, during “machine running running” and “push-up” exercises, the arm motion is the left shoulder 110 of the torso, the left elbow 111, and the left arm motion of the left forearm 112 and the right shoulder 120 ), the right elbow 121, and the right arm movement of the right forearm 122.

The leg motion is the left hip part 140, the left knee joint 141, and the left ankle joint 142 of the lower body, and the right hip part 150, the right knee joint 151, and the right ankle joint 152 ) is detected by the movement of the right foot.

During health, fitness, yoga, and sports exercises, the exercise application (App) of the user terminal connects the joint point nodes of the user's exercise image based on the expert trainer's exercise image and analyzes the exercise image to determine the motion model of each part. The left elbow joint 111, the right elbow joint 121, the left knee joint 141, and the right knee joint ( 151), the angles of the solid lines connected between the vertices of each joint of the measurement points of the human body are mutually compared, and the motion consistency of the exercise motions of the user image based on the trainer image is scored and displayed.

That is, according to the motion model of the motion motion classified by motion motion, the degree of motion coincidence at each motion stage during motion of the user image based on the trainer image is the vertex of each joint of the measurement points of the human body constituting the main point of the human body. The motion agreement is also scored for each motion step by comparing the angle of the connected solid line between the vertices.

The reference angle of the horizontal line of the major joints at each motion stage during exercise obtained from the expert trainer image (A) and the user's statue (B) will be described as an example.

The trainer image and the user image are the angle of the head 100 and the torso belly button 130, the angle of the left shoulder 110 and the right shoulder 112, the left shoulder 110 and The angle of the left elbow joint 111, the angle between the left elbow joint 111 and the left forearm joint 112, the angle between the right shoulder 120 and the right elbow joint 121, the right elbow joint 121 and the right forearm joint The angle of (122), the angle between the belly button 130 and the left hip 140, the angle between the belly button 130 and the right hip 150, the left hip 140 and the right hip 150 angle, the angle between the left hip 140 and the left knee joint 141, the angle between the left knee joint 141 and the left ankle joint 142, the angle between the right hip 150 and the right knee joint 151, and the angles of the right knee joint 151 and the right ankle joint 152 are measured, and during exercise, the user's image to which the sparkling tags are attached to the measuring points of the human body based on the trainer image for each action step is exercised. For each motion step, the solid line angle between the apex and the vertex of each joint of the measurement points of the human body is compared, and a motion coincidence score is given for each motion step, and the motion coincidence score is calculated by cumulative summing.

For example, the trainer image and the user image are an angle 111-112 between the left elbow 111 and the left forearm joint 112, and an angle 121 between the right elbow 121 and the right forearm joint 122 relative to the horizontal. -122), the angle 141-142 between the left knee joint 141 and the left ankle joint 142, and the angle 151-152 between the right knee joint 151 and the ankle joint 152 will be described as examples. do.

trainer Angle of the left elbow (111-112) Angle of right elbow (121-122) Angle of the left knee (141-142) Angle of right knee (151-152) note motion 0 (0,1) 10° (0,2) 10° (0,3) 70° (0,4) 70° action 1 (1,1) 20° (1,2) 20° (1,3) 30° (1,4) 30° action 2 (2,1) 30° (2,2) 20° (2,3) 50° (2,4) 50° action 3 (3,1) 40° (3,2) 10° (3,3) 20° (3,4) 30°

User image (B) user Angle of the left elbow (111-112) Angle of right elbow (121-122) Angle of the left knee (141-142) Angle of right knee (151-152) motion concordance motion 0 (0,1) 11° (0,2) 11° (0,3) 60° (0,4) 60° 3+3+2+2 = 10 points action 1 (1,1) 23° (1,2) 23° (1,3) 42° (1,4) 42° 3+3+1+1=8 points action 2 (2,1) 39° (2,2) 29° (2,3) 63° (2,4) 56° 2+2+1+1=6 points action 3 (3,1) 47° (3,2) 13° (3,3) 23° (3,4) 41° 2+2+1+2 = 7 points Motion Consensus Score
31/48 points

A(0,1) - B(0,1) = x(0,1)

A(0,2) - B(0,2) = x(0,2)

A(0,3) - B(0,3) = x(0,3)

A(0,4) - B(0,4) = x(0,4)

…

(Example) 0°≤ x(0,1) ≤ 5°: motion coincide (3 points), good

5°< x(0,1) ≤10°: approximate motion match (2 points), normal

10° ≤ x(0,1) ≤15°: motion mismatch (1 point), bad

x(0,1) > 15°: complete mismatch of motion (0 points), very bad

…

Humans have different upper body parts (100, 110, and 120) and lower body parts (130, 140, and 150) having a hip joint. The measurement points of the human body are the upper body parts (100, 110, 120) in the front posture form an isosceles triangle with the base (110-120) between 110 and 120, the left side between 100 and 110, and the right side between 100 and 120, and the lower body area (130,140,150) forms an isosceles triangle with a base (140-150) between 140 and 150, a left side between 130 and 140, and a right side between 130 and 150.

Referring to FIG. 13, since the human body and body shape are different, the length and height of the bases of the upper body parts (100, 110, 120)-reference 6 and lower body parts (130, 140, 150)-reference 7 isosceles triangles in the front posture of the trainer image and the user image. is different,

The exercise app of the user terminal uses an image scaling technique to match the user image to the length and height of the base of the isosceles triangle of the expert trainer image, and uses the upper body part (100, 110, 120)-reference 6 or lower body part (130, 140, 150) of the trainer image - Enlarges/reduces the video/photo so that the upper body or lower body of the user's image approximately matches the length and height of the base of the isosceles triangle according to the length and height of the base of the isosceles triangle of criterion 7,

At each motion step during exercise, the motion using alpha blending is detected, the position of the exercise posture is detected, the amount of exercise (p = mv) is calculated, and the image of the user's exercise video watching and following the expert trainer's exercise video By connecting the joint point nodes of the trainer image and the motion image analysis of the user image, the measured points of the human body are compared with the solid line angle between the apex and the vertex of each joint, and the motion of the user's motion image is based on the trainer's motion image. The degree of agreement is scored (ex. motion agreement 3 points, approximate motion agreement - 2 points, motion mismatch - 1 point, motion complete mismatch - 0 points) and compared with each other, and the exercise of the user's video with the expert trainer video of the exercise It outputs the motion congruency score and the graph in which the motion congruency is scored, and provides exercise management healthcare service for posture correction in terms of exercise dynamics of physical education science.

Embodiments according to the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer readable recording medium. The computer readable recording medium may include program instructions, data files, and data structures alone or in combination. Computer-readable recording media include storage, hard disks, magnetic media such as floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - A hardware device configured to store and execute program instructions in storage media such as magneto-optical media, ROM, RAM, flash memory, etc. may be included. Examples of program instructions may include those produced by compilers and machine language codes as well as high-level language codes that can be executed by a computer using an interpreter. The hardware device may be configured to operate as one or more software modules to perform the operations of the present invention.

As described above, the method of the present invention is implemented as a program and can be read using computer software on a recording medium (CD-ROM, RAM, ROM, memory card, hard disk, magneto-optical disk, storage device, etc.) ) can be stored in

Although described with reference to specific embodiments of the present invention, the present invention is not limited to the same configuration and operation as the specific embodiments to illustrate the technical idea as described above, and within the limit that does not deviate from the technical spirit and scope of the present invention It can be implemented with various modifications, and the scope of the present invention should be determined by the claims described later.

100: computer 110: smartphone, tablet PC
130: TV display 200: exercise management server
210: web server 220: web application server (WAS)
230: DB 270: streaming server
272 video encoder 274 camera
201: web server 202: control unit
203: member management unit 204: user authentication unit
205: exercise content providing unit 206: weight and body composition measurement information input unit
207: user exercise image storage unit
208: Trainer video/user video comparison analysis unit
209: momentum detection unit 211: statistical information providing unit
212: database (DB) 213: bulletin board

Claims (7)

An exercise management server that has a web server and a database for health, fitness, yoga, and sports exercises, and provides exercise management healthcare services by providing video and exercise contents of expert trainers to user terminals for each expert trainer; and
It is connected to the exercise management server through a wired/wireless communication network, attaches a sparkling tag attached to each part of the human body, and with respect to the exercise image of the user photographed with a camera that exercises according to the video of the expert trainer, each part of the human body Recognizes the shiny tag attached to and detects the motion using alpha blending, connects the joint point nodes of the expert trainer's exercise image and the user's exercise image image, and analyzes the motion image Comparing the solid line angle between vertices of , and scoring the motion match between the expert trainer exercise image and the user's exercise image, and outputting a graph in which the motion match score and the motion match score with the expert trainer image of the exercise are scored A user terminal in which an exercise application (App) for health, fitness, yoga, crossfit, and sports exercise is installed; contains
The exercise management server may include a web server;
A controller that controls exercise management healthcare services for health, fitness, yoga, crossfit, and sports exercises; and
An exercise content provider connected to the control unit and providing expert trainer images, exercise information, and exercise contents for each exercise;
a weight and body composition information management unit that is connected to the control unit and receives and stores measurement values including a user's height, weight, muscle mass, and body fat mass from an exercise application (App) of a user terminal;
a user exercise image storage unit connected to the control unit and storing a user exercise image received from an exercise application (App) of a user terminal;
It is connected to the control unit, recognizes the shiny tag attached to each part of the body of a person exercising with respect to the user's exercise image captured by the camera, and detects the motion using alpha blending. A trainer image/user image comparison and analysis unit that compares and analyzes the expert trainer image and user image by connecting joint point nodes of the image of the exercise image and scoring whether the motions of each part match through analysis of the exercise image;
It is connected to the control unit, and measures values including the user's height, weight, muscle mass, body fat mass, and BMI before and after exercise are stored in the exercise management server, and accumulated and stored in the database of the exercise management server for each time period/day/week. Statistical information providing unit that provides individual exercise statistical data including weight, exercise amount, muscle mass, body fat mass, and BMI value for each month/year;
A database (DB) connected to the control unit, storing expert trainer images for each exercise, exercise information, and exercise contents, and storing user exercise images and measured values including the user's height, weight, muscle mass, body fat mass, and BMI values; and
a bulletin board connected to the control unit and registering texts of expert trainers and users for each exercise field; contains
The exercise management server 200 is connected to the control unit 202, and during machine learning running exercise at the gym, when a user with m weight runs at v speed, the amount of exercise p is measured as m * v, and the exercise energy is Further comprising an exercise amount detector to measure and store and display it as an exercise application (App) of the user terminal,
The measurement points of the human body are the head 100, left shoulder 110, left elbow joint 111, left forearm joint 112, right shoulder 120, right elbow joint 121, right forearm joint 122 , including navel 130, left hip 140, left knee joint 141, left ankle joint 142, right hip 150, right knee joint 151, and right ankle joint 152 do
In addition, in the front posture of the expert trainer image and the user image, the reference 6 isosceles triangle, which is the head 100, left shoulder 110, and right shoulder 120, which are the upper body parts, and the navel 130, which is the lower body part, and the left hip part ( 140) and the right hip part 150, if the length and height of the base of the reference 7 isosceles triangle are different,
The exercise app of the user terminal uses an image scaling technique to match the user image to the length and height of the base of the isosceles triangle of the expert trainer image
According to the length and height of the base of the reference 6 isosceles triangle, which is the upper body part of the expert trainer image, or the reference 7 isosceles triangle, which is the lower body part
Magnifying or reducing the image so that the length and height of the base of the isosceles triangle of the upper body part or the isosceles lower body part of the user image match,
Comparative analysis of the expert trainer image and the user image
Based on the expert trainer image, tags are attached to the measurement points of the user's body and the angles of solid lines connecting each tag for each action step are compared in the image of the user exercising.
In the expert trainer image and the user image, the angle of the solid line connecting the head 100 and the belly button 130 and the solid line connecting the left shoulder 110 and the right shoulder 112 based on the horizontal line for each action step The angle of , the angle of the solid line connecting the left shoulder 110 and the left elbow joint 111, the angle of the solid line connecting the left elbow joint 111 and the left forearm joint 112, the right shoulder 120 and the right elbow The angle of the solid line connecting the joint 121, the angle of the solid line connecting the right elbow joint 121 and the right forearm joint 122, the angle of the solid line connecting the navel part 130 and the left hip part 140, The angle of the solid line connecting the belly button 130 and the right buttock 150, the angle of the solid line connecting the left buttock 140 and the right buttock 150, the left buttock 140 and the left knee joint ( 141), the angle of the solid line connecting the left knee joint 141 and the left ankle joint 142, the angle of the solid line connecting the right hip 150 and the right knee joint 151, and Measuring the angle of the solid line connecting the right knee joint 151 and the right ankle joint 152
A score is given according to the match between the expert trainer image and the user image, and the motion match score is calculated by accumulating and summing them to obtain a graph in which the motion match score and the motion match score of the user image and the expert trainer image of the corresponding exercise are scored. Exercise management healthcare service system using output tags.
According to claim 1,
During exercise, the user attaches shiny tags or sensors to the measurement points of the human body, takes pictures with a camera, and provides user images.
When the sensor is attached to the measurement points of the human body, the sensor uses a three-dimensional motion detection sensor, exercise management healthcare service system using a tag. delete delete According to claim 1,
The exercise app of the user terminal is connected to the exercise management server through a wired/wireless communication network, and measures the user's height and weight before and after exercise, and the user's weight, muscle mass, and body fat mass measured by a body composition detection device (In Body Scale). , Measured values including BMI values are stored in the exercise management server, and accumulated and stored in the database of the exercise management server to determine the amount of exercise, weight, muscle mass, body fat mass, and BMI values by time slot/day/week/month/year. An exercise management healthcare service system using a tag that provides individual exercise statistical data to the user terminal. According to claim 5,
The exercise app of the user terminal is interlocked with measurement values including weight, muscle mass, body fat mass, and BMI values of the user's height and weight measurement and body composition detection device (In Body scale) stored in the exercise management server. , Exercise management healthcare service system using tags. According to claim 6,
Weight measurement and body composition detection device (In Body scale) test results, skeletal muscle fat analysis (weight, muscle mass, body fat mass), obesity analysis (BMI, body fat percentage), and visceral fat analysis (visceral fat level) ) to display in the exercise application (App) of the user terminal;
The body balance change process over time of the measured values of weight, muscle mass, body fat mass, and body fat percentage is expressed as a line chart, and the weight measurement and body composition detector (In Body scale) scores are displayed on the standard normal distribution curve, The exercise application of the user terminal displays the top percentage of the statistical analysis data having a normal distribution by the average of the weight measurement and body composition detector (In Body scale) scores of a sample of a population with a predetermined number of standard body shapes measured in advance. Exercise management healthcare service system using tags displayed in (App).