CN108280423A - Strong appearance system based on human body attitude identification - Google Patents
Strong appearance system based on human body attitude identification Download PDFInfo
- Publication number
- CN108280423A CN108280423A CN201810060071.9A CN201810060071A CN108280423A CN 108280423 A CN108280423 A CN 108280423A CN 201810060071 A CN201810060071 A CN 201810060071A CN 108280423 A CN108280423 A CN 108280423A
- Authority
- CN
- China
- Prior art keywords
- human body
- data
- manikin
- action
- posture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
Based on the strong appearance system of human body attitude identification, belong to athletic posture adjustment technology field.The present invention be in order to solve the problem of existing body motion information identification technology it is affected by environment it is big can not be that the posture recognized is instructed as specification.It includes:Posture acquisition unit:Corresponding site for Posture acquisition sensor to be fixed on to human body obtains exercise data;Gathered data processing unit:For handling exercise data, the identifiable action command data of host computer are converted to;Data transmission unit:For the action command data to be transferred to host computer;Host computer:For prebuild manikin, and action command data are converted to the action of manikin;The action of manikin is compareed with standard operation simultaneously, obtains the remedial teaching of posture.Identification and strong appearance of the present invention for human body attitude.
Description
Technical field
The present invention relates to the strong appearance systems identified based on human body attitude, belong to athletic posture adjustment technology field.
Background technology
The system that human action capture and then identify, oneself is through being widely used in virtual reality, biomethanics, trip
Other field of human-computer interaction such as play, medical health, which includes based on the principles such as machinery, optical, acoustics, electromagnetism
Human action capture system, it using digital information technology come measure human body three dimensions movable information.
Mechanical human action capture device is easy to make human body limitation of movement;System is captured based on optical human action
System cost is high, it is high to environmental requirement, install it is comparatively laborious;Human action based on video analysis method captures system and exists
It is more complicated in terms of Algorithm Analysis and realization:Current human's posture captures identification technology and is mostly based on video processing technique, borrows
High-precision video camera is helped to realize human body dynamic imaging picture;But there are expensive, easily influenced by light condition, it is dead that there are visions
The shortcomings of angle;Human action capture system real time based on acoustics is poor, is easily disturbed.
Invention content
It is affected by environment big the invention aims to solve existing body motion information identification technology, can not be identification
To posture instructed as specification the problem of, provide a kind of strong appearance system identified based on human body attitude.
Strong appearance system of the present invention based on human body attitude identification, it includes:
Posture acquisition unit:Corresponding site for Posture acquisition sensor to be fixed on to human body obtains exercise data;
Gathered data processing unit:For handling exercise data, the identifiable action command of host computer is converted to
Data;
Data transmission unit:For the action command data to be transferred to host computer;
Host computer:For prebuild manikin, and action command data are converted to the action of manikin;Simultaneously will
The action of manikin is compareed with standard operation, obtains the remedial teaching of posture.
In one embodiment of the invention, the Posture acquisition sensor includes two, and the upper arm of human body right arm is under
Arm respectively installation one.
In another embodiment of the present invention, the Posture acquisition sensor include eight, the upper arm of human body both arms and
Underarm respectively installation one, thigh and the shank respectively installation one of both legs.
In yet another embodiment of the present invention, the Posture acquisition sensor is for acquiring 15 working nodes of human body
Action, 15 working nodes include the upper and lower arms of both arms and the thigh of both legs and 8 nodes of shank, 1, head node, on
2 nodes of body, both hands and double-legged 4 nodes.
In yet another embodiment of the present invention, the model MPU9250 of Posture acquisition sensor.
In yet another embodiment of the present invention, the Posture acquisition unit is passed by serial ports to gathered data processing unit
Send data.
In yet another embodiment of the present invention, the action that action command data are converted to manikin includes:
Manikin is initialized, each joint original state of manikin is preserved;Then action command number is received by serial ports
According to the n-th joint of acquisition current human quaternary number takes preceding 30 quaternion algebras according to buffer area is put into, weighted average is as human body n-th
Joint current pose data, and make manikin the n-th joint posture and current pose Data Matching;Continue to n-th of joint
Action command data, the quaternary number increment of rotation of calculating action director data and an adjacent preceding action command data, to rotation
Turn increment and carry out coordinate matching, n-th of joint of manikin is made accordingly to rotate, completes of manikin and user's action
Match, n is the sequence number of human synovial.
Advantages of the present invention:The interactive learning of the present invention person's of may be used in movement posture, people is captured by sensor
It after body attitude data, is compareed with standard posture, completes the study of standard movement posture.
The problems such as movement posture catching mode of the present invention need not take light, privacy and visual dead angle into account, it passes through wireless
Sensor network transfers data to control centre, can realize low-power consumption, low cost, high real-time human body attitude identify and rectify
Just.
The present invention can be used for posture of the specification such as Yoga, swimming, body-building.
Description of the drawings
Fig. 1 is the connection block diagram of the strong appearance system of the present invention identified based on human body attitude;
Fig. 2 is the flow chart of data transmission unit sending action director data;
Fig. 3 is the calibration flow chart of Posture acquisition sensor in Posture acquisition unit;
Fig. 4 is the flow chart that data transmission unit receives action command data;
Fig. 5 is the flow chart for the action that action command data are converted to manikin in host computer.
Specific implementation mode
The embodiments of the present invention will be described in detail below with reference to the drawings:
Based on the strong appearance system of human body attitude identification, it includes:
Posture acquisition unit 1:Corresponding site for Posture acquisition sensor to be fixed on to human body obtains exercise data;
Gathered data processing unit 2:For handling exercise data, the identifiable action command of host computer is converted to
Data;
Data transmission unit 3:For the action command data to be transferred to host computer;
Host computer 4:For prebuild manikin, and action command data are converted to the action of manikin;Simultaneously
The action of manikin is compareed with standard operation, obtains the remedial teaching of posture.
Since Posture acquisition unit 1 needs to be fixed on human body, thus should meet as possible have miniaturization and it is portable
The characteristics of, patch encapsulation can be used in all devices.
In conjunction with shown in Fig. 1, Posture acquisition unit 1, gathered data processing unit 2 and the data of data transmission unit 3 are sent
Part can further include being made of rechargeable battery and voltage stabilizing chip xc6206p302 as the transmitting terminal of data, the transmitting terminal
Power supply system.
Since Posture acquisition unit 1 needs to be fixed on human body, so power cord cannot be used to power, therefore power supply can be used
Rechargeable lithium battary can be charged to lithium battery by TP4056 power management chips, the electricity of lithium battery can be prompted by LED light
Situation is measured, such as red LED lights when charging, the LED light of blue is bright when fully charged;Voltage stabilizing chip is that xc6206p302 can be micro-
Processor and each module provide stable input voltage.
Usb circuit can be used, on the one hand as the charging interface of lithium battery, be on the other hand used as host computer 4 with
The serial communication of hardware circuit.
The nine axis attitude transducers of model MPU9250 may be used in Posture acquisition unit 1, and the DMP in MPU9250 can be defeated
Go out the information such as quaternary number, Eulerian angles and acceleration, angular speed.MPU9250 is the compound chip (MCM) of QFN encapsulation, it
It is made of 2 parts.One group is that 3 axle accelerations also have 3 axis gyroscopes, the AK8963-3 axis magnetometers of another Zu Zeshi AKM companies.
So MPU9250 is a 9 axis motion tracer, he has merged 3 axle accelerations, 3 axis in the encapsulation of slight 3X3X1mm
Gyroscope and digital moving processor (DMP).Motion digital processing engine (DMP) is located inside MPU9250, can directly locate
Data are managed, the task of main control chip is reduced.It only needs obtained acceleration, gyroscope, the even external biography of magnetic field value
The value of sensor is directly to it, the value then directly calculated from register read, or directly good value is stored in
Fifo buffer.DMP has a interrupt pin, can do to wake up main control chip.It can be obtained by its digital moving processor DMP
Required quaternary number and Eulerian angles.
Eulerian angles are the triplets independence angle parameters for uniquely determining Fixed-point Motion of A phaneroplasm position, by nutational angle θ,
Angle of precession ψ and angle of rotation φ compositions.Eulerian angles can express the relative deviation angle between two coordinate systems, can be by three times
Different rotations reaches target-based coordinate system, can choose any one axis for the first time, selects in other two axis one for the second time,
Third time selects another except second, and there are many representation methods.If rigid body is around a certain axis by pinpointing O with angle
Speed omega rotates, such as each components of ω of known any instant t, and ψ, θ, j and time t can be found out according to Euler kinematical equations
Relationship, thus also just determine the movement of rigid body.
Application of the quaternary number in terms of attitude of carrier description is more and more, it can make up with Eulerian angles in attitude description mistake
Deficiency in journey, quaternary number are made of as a kind of technique of expression of description action three imaginary number is, j, k on a real add,
It can be expressed as:
Q=a+bi+cj+dk (1)
It can also be denoted as:
It carries out calculating the movement that can capture a rigid body by information such as quaternary number, Eulerian angles, acceleration and angular speeds,
Such as the motion conditions of human body right arm are captured, only one sensor need to be respectively installed in upper underarm, you can monitor the entirety of arm
Motion conditions.
The processor of model STM32F103 can be used in gathered data processing unit 2.
Data transmission unit 3 can be divided into data sending part point and data receiving portion.Data sending part point may be used
NRF24L01 modules;Power supply is 3V, is consistent with present system, and the distance wirelessly supported is 20m, can meet use of the present invention
Range, it has 16 working nodes to work simultaneously.Can meet the present invention capture double needs, such as with
Lower situation:The Posture acquisition sensor is used to acquire the action of 15 working nodes of human body, and 15 working nodes include both arms
Upper and lower arms and both legs 8 nodes of thigh and shank, 1, head node, the node of upper body 2, both hands and both feet 4 section
Point.Therefore the number of nodes of NRF24L01 modules is met the requirements.
Host computer 4 simulates the athletic posture of user by the Real-time Monitoring Data of acquisition, and compared with standard posture,
Judge whether user's action is correct.Program in host computer 4 can be based on Unity3D, and design is write with C# language.
Unity3D be by Unity Technologies are developed one can easily create such as 3 D video game, building visualization,
The multi-platform comprehensive development of games tool of the types interaction content such as realtime three dimensional animation is a profession trip integrated comprehensively
Play engine.The host computer of this software programming can establish coordinate system in inside, so that it is corresponded to the world coordinate system of reality, pass through string
The attitude data that mouth receives, is calculated, to achieve the purpose that manikin is kept strokes with user.It is dynamic based on its dimension
It draws, user intuitively can judge that its is acted by accurate observation by its physical impacts to the action oneself done by programming
Standard whether.
The Posture acquisition sensor includes eight, the upper and lower arms respectively installation one of human body both arms, the thigh of both legs
One is respectively installed with shank.
The model MPU9250 of Posture acquisition sensor.
The Posture acquisition unit transmits data by serial ports to gathered data processing unit.
The action that action command data are converted to manikin includes:
Manikin is initialized, each joint original state of manikin is preserved;Then action command number is received by serial ports
According to the n-th joint of acquisition current human quaternary number takes preceding 30 quaternion algebras according to buffer area is put into, weighted average is as human body n-th
Joint current pose data, and make manikin the n-th joint posture and current pose Data Matching;Continue to n-th of joint
Action command data, the quaternary number increment of rotation of calculating action director data and an adjacent preceding action command data, to rotation
Turn increment and carry out coordinate matching, n-th of joint of manikin is made accordingly to rotate, completes of manikin and user's action
Match, n is the sequence number of human synovial.
Software flow pattern shown in Fig. 2 first carries out the initialization of IO etc., is read from FLASH and sends ID and channel transmission
Address measures calibration data, the posture of different node monitoring different parts, therefore each sensor is needed through serial ports, and setting is different
ID, then Initialize installation is carried out to disparate modules, the human body attitude data of monitoring are sent later, successfully LED is sent and refers to
Show the color of lamp from red to blue.
As transmitting terminal, to the three-axis gyroscope, three axis accelerometer and three axle magnetometer of MPU9250 need respectively into
Row calibration, only needs remains stationary motionless gyro calibiatio i, is calibrated respectively without three axis;Accelerometer is then needed with magnetometer
The calibration of tri- axis of XYZ is carried out, each axis calibration is divided into positive negative direction and calibrates twice, and calibration procedure is controlled by serial ports, calibration
Flow it is as shown in Figure 3.
Shown in Fig. 4, first the IO etc. in STM32 is initialized, receiving node number, corresponding channel are read from FLASH
And address is received, then Initialize installation is carried out to disparate modules, the number of nodes and corresponding channel that scanning initialization is read will
The attitude data of different nodes sends host computer to by serial ports.
In Fig. 5, a manikin is first made in the unity software of host computer, and world coordinate system is set, serial ports is passed through
The node ID and attitude data of transmission control manikin by processing to data and calculating.It initializes first
Each joint coordinates match variable, and each joint initial attitude of preservation model receives the data that serial ports is sent, and obtains current human n-th
Joint quaternary number, takes preceding 30 data to be put into buffer area, and weighted average completes the n-th joint as the n-th joint of human body initial attitude
Calibration calculates the quaternary number increment of rotation in n-th of joint of human body further according to data, carries out coordinate matching to changing increment, is transformed to
The quaternary number increment of n-th of model, executes the rotation in the joint, thus manikin is made to keep strokes with user.
Standard operation program in host computer can be write in advance, such as one simply judges example:Right arm is swum
When action, the position range that underarm should occur can be write according to the different location of upper arm, when underarm goes beyond the scope, system
Show stroke defect.If action is correct, underarm is raised to for the first time from underarm second in surface process, if being
System is not detected underarm and goes beyond the scope, then system display action is correct, completes the judgement to human action.
Claims (7)
1. a kind of strong appearance system based on human body attitude identification, which is characterized in that it includes:
Posture acquisition unit:Corresponding site for Posture acquisition sensor to be fixed on to human body obtains exercise data;
Gathered data processing unit:For handling exercise data, the identifiable action command data of host computer are converted to;
Data transmission unit:For the action command data to be transferred to host computer;
Host computer:For prebuild manikin, and action command data are converted to the action of manikin;Simultaneously by human body
The action of model is compareed with standard operation, obtains the remedial teaching of posture.
2. the strong appearance system according to claim 1 based on human body attitude identification, which is characterized in that
The Posture acquisition sensor includes two, the upper and lower arms respectively installation one of human body right arm.
3. the strong appearance system according to claim 1 based on human body attitude identification, which is characterized in that
The Posture acquisition sensor includes eight, the respectively installation one of the upper and lower arms of human body both arms, the thigh of both legs and small
Leg respectively installation one.
4. the strong appearance system according to claim 1 based on human body attitude identification, which is characterized in that the Posture acquisition passes
Sensor is used to acquire the actions of 15 working nodes of human body, 15 working nodes include both arms upper and lower arms and both legs it is big
8 nodes of leg and shank, 1, head node, the node of upper body 2, both hands and double-legged 4 nodes.
5. the strong appearance system according to any one of claim 1 to 4 based on human body attitude identification, which is characterized in that appearance
State acquires the model MPU9250 of sensor.
6. the strong appearance system according to any one of claim 1 to 4 based on human body attitude identification, which is characterized in that institute
It states Posture acquisition unit and data is transmitted to gathered data processing unit by serial ports.
7. the strong appearance system according to claim 1 based on human body attitude identification, which is characterized in that described by action command
The action that data are converted to manikin includes:
Manikin is initialized, each joint original state of manikin is preserved;Then action command data are received by serial ports, obtained
The n-th joint of current human quaternary number is taken, takes preceding 30 quaternion algebras according to buffer area is put into, weighted average is as the n-th joint of human body
Current pose data, and make manikin the n-th joint posture and current pose Data Matching;Continue to the dynamic of n-th joint
Make director data, the quaternary number increment of rotation of calculating action director data and an adjacent preceding action command data increases rotation
Amount carries out coordinate matching, and n-th of joint of manikin is made accordingly to rotate, and completes the matching of manikin and user's action, and n is
The sequence number of human synovial.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810060071.9A CN108280423A (en) | 2018-01-22 | 2018-01-22 | Strong appearance system based on human body attitude identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810060071.9A CN108280423A (en) | 2018-01-22 | 2018-01-22 | Strong appearance system based on human body attitude identification |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108280423A true CN108280423A (en) | 2018-07-13 |
Family
ID=62804408
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810060071.9A Pending CN108280423A (en) | 2018-01-22 | 2018-01-22 | Strong appearance system based on human body attitude identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108280423A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108939512A (en) * | 2018-07-23 | 2018-12-07 | 大连理工大学 | A kind of swimming attitude measurement method based on wearable sensor |
CN114225343A (en) * | 2021-12-23 | 2022-03-25 | 成都德鲁伊科技有限公司 | Method for swimming by utilizing AR |
CN115120954A (en) * | 2022-07-26 | 2022-09-30 | 江苏智恒文化科技有限公司 | Sports fitness exercise control system |
CN117115925A (en) * | 2023-10-24 | 2023-11-24 | 广州华夏汇海科技有限公司 | Intelligent pool wall for swimming examination room |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103135765A (en) * | 2013-02-20 | 2013-06-05 | 兰州交通大学 | Human motion information capturing system based on micro-mechanical sensor |
US20140267611A1 (en) * | 2013-03-14 | 2014-09-18 | Microsoft Corporation | Runtime engine for analyzing user motion in 3d images |
CN104700452A (en) * | 2015-03-24 | 2015-06-10 | 中国人民解放军国防科学技术大学 | Three-dimensional body posture model matching method for any posture |
CN107330967A (en) * | 2017-05-12 | 2017-11-07 | 武汉商学院 | Knight's athletic posture based on inertia sensing technology is caught and three-dimensional reconstruction system |
-
2018
- 2018-01-22 CN CN201810060071.9A patent/CN108280423A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103135765A (en) * | 2013-02-20 | 2013-06-05 | 兰州交通大学 | Human motion information capturing system based on micro-mechanical sensor |
US20140267611A1 (en) * | 2013-03-14 | 2014-09-18 | Microsoft Corporation | Runtime engine for analyzing user motion in 3d images |
CN104700452A (en) * | 2015-03-24 | 2015-06-10 | 中国人民解放军国防科学技术大学 | Three-dimensional body posture model matching method for any posture |
CN107330967A (en) * | 2017-05-12 | 2017-11-07 | 武汉商学院 | Knight's athletic posture based on inertia sensing technology is caught and three-dimensional reconstruction system |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108939512A (en) * | 2018-07-23 | 2018-12-07 | 大连理工大学 | A kind of swimming attitude measurement method based on wearable sensor |
CN114225343A (en) * | 2021-12-23 | 2022-03-25 | 成都德鲁伊科技有限公司 | Method for swimming by utilizing AR |
CN115120954A (en) * | 2022-07-26 | 2022-09-30 | 江苏智恒文化科技有限公司 | Sports fitness exercise control system |
CN115120954B (en) * | 2022-07-26 | 2024-03-08 | 江苏智恒文化科技有限公司 | Sports body-building motion control system |
CN117115925A (en) * | 2023-10-24 | 2023-11-24 | 广州华夏汇海科技有限公司 | Intelligent pool wall for swimming examination room |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107330967B (en) | Rider motion posture capturing and three-dimensional reconstruction system based on inertial sensing technology | |
CN108280423A (en) | Strong appearance system based on human body attitude identification | |
CN110327048B (en) | Human upper limb posture reconstruction system based on wearable inertial sensor | |
CN101579238B (en) | Human motion capture three dimensional playback system and method thereof | |
US20100194879A1 (en) | Object motion capturing system and method | |
US20140009384A1 (en) | Methods and systems for determining location of handheld device within 3d environment | |
CN108939512A (en) | A kind of swimming attitude measurement method based on wearable sensor | |
CN110457414A (en) | Offline map processing, virtual objects display methods, device, medium and equipment | |
CN104197987A (en) | Combined-type motion capturing system | |
WO2023221524A1 (en) | Human movement intelligent measurement and digital training system | |
CN106843507A (en) | A kind of method and system of virtual reality multi-person interactive | |
CN110456905A (en) | Positioning and tracing method, device, system and electronic equipment | |
Alexandre et al. | Wearable and IoT technologies application for physical rehabilitation | |
CN108109460A (en) | Equipment is visited in a kind of teaching with augmented reality chemical plant | |
US20220026981A1 (en) | Information processing apparatus, method for processing information, and program | |
CN108151738A (en) | Codified active light marked ball with attitude algorithm | |
CN206819290U (en) | A kind of system of virtual reality multi-person interactive | |
Sama et al. | 3dID: a low-power, low-cost hand motion capture device | |
Lin et al. | Using hybrid sensoring method for motion capture in volleyball techniques training | |
WO2019021149A1 (en) | Portable virtual reality system for carrying out a subjective visual vertical test | |
US10748320B2 (en) | Animation simulation of biomechanics | |
CN110757471A (en) | Dance robot system based on computer vision and operation method thereof | |
CN114756130A (en) | Hand virtual-real interaction system | |
WO2021190421A1 (en) | Virtual reality-based controller light ball tracking method on and virtual reality device | |
CN113536552A (en) | Human body posture visual tracking system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180713 |
|
RJ01 | Rejection of invention patent application after publication |