CN113536552A - Human body posture visual tracking system - Google Patents
Human body posture visual tracking system Download PDFInfo
- Publication number
- CN113536552A CN113536552A CN202110736567.5A CN202110736567A CN113536552A CN 113536552 A CN113536552 A CN 113536552A CN 202110736567 A CN202110736567 A CN 202110736567A CN 113536552 A CN113536552 A CN 113536552A
- Authority
- CN
- China
- Prior art keywords
- human body
- opensim
- data
- body posture
- visual
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Abstract
The invention relates to a human body posture visual tracking system, which comprises a human body posture detection module, a wireless transmission module and an OpenSim visual processing module, wherein the human body posture detection module wirelessly transmits posture angle data calculated by a plurality of paths of human body posture detection modules to a PC (personal computer) end, and the tracking process of human body movement posture visual tracking is realized after Matlab and OpenSim script processing; the invention utilizes the nine-axis inertial sensor to detect the attitude angle, greatly improves the detection precision, has wide application range, overcomes the problem that most of the human body attitudes are calculated by adopting a video processing mode in the market and cannot adapt to complex scenes, and simultaneously uses Matlab and OpenSim to carry out visual tracking, thereby being convenient for test development and being widely applied to the fields of medical treatment, entertainment and the like.
Description
Technical Field
The invention belongs to the technical field of human body posture detection, and particularly relates to a human body posture visual tracking system.
Background
With the continuous development of the gesture detection technology and the continuous widening of the gesture detection technology in various fields, the human gesture detection technology is more and more widely applied in the fields of medical treatment, entertainment and the like, and accordingly, the requirements on the measurement accuracy, the stability and the visual function of gesture detection are higher and higher.
The invention establishes a human body tracking model in a video shooting mode, trains the human body tracking model to establish a linear motion model to track the human body posture, has higher efficiency and accuracy, but has narrow applicable environment, can only be applied to fixed scenes, and has complex modeling flow and larger application limitation. The invention also discloses an invention patent with an authorization publication number CN112784786A, which is named as a human body posture identification method and a device, and the invention adopts a shooting mode to obtain a multi-frame two-dimensional image containing a target object, determines a human body motion track of the target object based on image characteristic information of at least two frames of two-dimensional images in the multi-frame two-dimensional image, and identifies and obtains the current human body posture of the target object based on the human body motion track, first image structure information and second image structure information. A human motion attitude detection method based on an inertial sensor is disclosed in an authorization publication number CN112890807A and is named as a human motion attitude detector based on an ARM and a nine-axis sensor, the human motion attitude detection method combines the data such as the number of steps of a human body, the bending degree of a spine and the like calculated by the nine-axis inertial sensor, and transmits the data to a mobile phone end in a Bluetooth mode.
Therefore, a human posture visual tracking system with wide application range, high measurement precision and convenient visual operation is needed.
Disclosure of Invention
The invention aims to provide a human body posture visual tracking system, which aims to solve the problems of small application scene range and complex modeling, improve the measurement precision and the applicable environment range of the human body posture visual tracking technology and reduce the visual development difficulty.
A human body posture visual tracking system based on an inertial sensor and OpenSim comprises a human body posture detection module, a wireless transmission module and an OpenSim visual processing module.
The human body posture detection module comprises an inertial sensor and an MCU signal processing circuit; the wireless transmission module is detachably arranged on each human body posture detection module; the OpenSim visualization processing module performs data fusion on attitude data of at least 10 nodes of a human body, and calculates and displays a human body attitude motion process. The inertial sensor transmits data to the MCU signal processing circuit, and the MCU signal processing circuit performs attitude calculation on the data of the inertial sensor to obtain a high-precision quaternion and transmits the quaternion to the wireless transmission module; the wireless transmission module transmits the high-precision quaternion to the PC, and the PC processes the data into OpenSim readable data; and importing the OpenSim readable data into an OpenSim visual processing module for OpenSim visual processing, and finally realizing visual tracking of the human body posture through Matlab script processing.
The inertial sensor is a nine-axis inertial sensor, comprises a three-axis gyroscope, a three-axis accelerometer and a three-axis magnetometer, and is used for acquiring angular velocity, acceleration and magnetic field intensity data of a human body during movement, transmitting the data to the MCU signal processing circuit, and forming rectangular grooves around the MCU signal processing circuit board for penetrating through the elastic band to fix the module at a human body measurement node.
The MCU signal processing circuit comprises a main control chip and a basic peripheral circuit thereof.
The wireless transmission module adopts a ZigBee protocol, and can be mutually networked when transmitting data of multiple nodes, so that the number of coordinators at a receiving end is greatly reduced.
The OpenSim visualization processing comprises the following steps:
1) installing a script interface of OpenSim in Matlab;
2) determining the model number of the human body model to be used according to the subject;
3) determining the initial posture and the number of measuring nodes of the human body model;
4) the initial posture of the tested person is kept the same as that of the model, then posture change is made, and data are wirelessly transmitted to a PC (personal computer) end;
5) carrying out format processing on the collected multi-node data to match an OpenSim data format, and importing and applying the data to an OpenSim script in Matlab;
6) setting a central node as a basic calibration node for calibrating the data direction of other nodes;
7) and after the calibration is finished, executing visual script operation to generate a file with motion attitude information, reading the file by using OpenSim to display the motion process of the human body, and simultaneously displaying the angle change value between each measurement node.
The invention achieves the aim of tracking the human body posture by using the nine-axis inertial sensor to measure the human body posture data and combining the application programming interface of OpenSim with Matlab, so that the accuracy of the analyzed human body posture angle data is greatly improved, the posture tracking is not limited to a fixed scene any more, the method can adapt to a more complex environment, and meanwhile, the visual tracking method by means of OpenSim and Matlab is simple to apply, has low cost and can be widely applied to the fields of medical treatment, entertainment and the like.
Drawings
FIG. 1 is a schematic diagram of a human detection node;
FIG. 2 is a block diagram of the overall system flow;
FIG. 3 is a block diagram of the lower level machine;
FIG. 4 is a block diagram of the lower computer work flow;
FIG. 5 shows the human body model used in OpenSim and its world coordinate system orientation.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the drawings used in the embodiments of the present invention. It is obvious that the described embodiments are only some embodiments of the present invention, and all other embodiments obtained by those skilled in the art without inventive efforts fall within the protection scope of the present invention.
The following describes the implementation of the present invention in detail with reference to specific embodiments.
Examples
As shown in fig. 1, the relative positions of the nodes of the human body detected by the invention are tested by selecting 10 nodes, and the human body posture detection module is bound at the positions of the upper and lower limbs, the back and the pelvis of the human body according to the position shown in the figure;
as shown in fig. 2, the MCU of each node reads the values of the three-axis acceleration, the three-axis angular velocity, and the three-axis magnetometer of the inertial sensor, and after the values are read, the attitude angles at the nodes are obtained through attitude calculation, and then the data are sent to the receiving module of the PC in a wireless transmission manner, and after the data are formatted, the data can be processed by the Matlab script of OpenSim, so as to realize the visual tracking of the human body attitude.
As shown in fig. 3, the schematic diagram of the human body posture detection module PCB is shown, the PCB design at each node is the same, 3-1 is an inertial sensor, the installation direction is 3-2, a rectangular groove 3-3 formed in the PCB is used for passing through an elastic band to bind the PCB to a test node, the human body posture detection module comprises a wireless transmission module interface 3-4 and an MCU 3-5, and the whole module is powered by a lithium battery.
As shown in fig. 4, which is a flow chart of each node, after the system is initialized, nine-axis data of the sensor is continuously read and subjected to attitude calculation to obtain an accurate attitude angle, and then the attitude angle is sent to the ZigBee wireless transmission module through a serial port, but for different nodes, data needs to be numbered, so that an upper computer can distinguish data of different nodes, and the data corresponds to the relative position of a human body.
After the attitude angle information at the 10 nodes is sent to the PC, the data is processed into data which can be read by OpenSim. Next, installing an OpenSim script interface in Matlab, and after the installation is finished, firstly, setting a human body model, as shown in FIG. 5, wherein the used human body model is 5-1, and the OpenSim world coordinate system is 5-2; then initializing the attribute of each node of the model, namely setting the human body position corresponding to each node, naming each node by bone names in OpenSim, and writing the bone name of the position of the node into a script; then setting the tracking time and determining to execute calibration; and then reading the received node data, calibrating by taking the first line data of the inertial sensor at the central node as an initial value, and simultaneously rotating the coordinate axis direction of the inertial sensor to be the same as the OpenSim world coordinate system.
After the steps are executed, the human body posture tracking script file is operated, the tracking result is observed in Matlab, an OpenSim executable file is generated, the human body functional exercise posture can be observed in OpenSim, the relative motion angle between the nodes can be output, and subsequent data analysis is facilitated.
The system measures the posture information by using the inertial sensor and realizes the visual tracking of the human posture by means of OpenSim and Matlab, so that the human posture tracking is not limited to fixed places any more, a subject can move freely within a range of dozens of meters, and meanwhile, by means of Matlab and OpenSim, the system can meet the test requirements of most developers.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiment, but is capable of various modifications, equivalents, changes, and the like. And that the present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. However, such variations are within the scope of the invention as long as they do not depart from the spirit of the invention.
Claims (5)
1. A human body posture visual tracking system comprises a human body posture detection module, a wireless transmission module and an OpenSim visual processing module;
the method is characterized in that: the human body posture detection module comprises an inertial sensor and an MCU signal processing circuit; the wireless transmission module is detachably arranged on each human body posture detection module; the OpenSim visualization processing module performs data fusion on attitude data at least at 10 measurement nodes of a human body, and resolves and displays a human body attitude motion process; the inertial sensor transmits data to the MCU signal processing circuit, and the MCU signal processing circuit performs attitude calculation on the data of the inertial sensor to obtain a high-precision quaternion and transmits the quaternion to the wireless transmission module; the wireless transmission module transmits the high-precision quaternion to the PC, and the PC processes the data into OpenSim readable data; and importing the OpenSim readable data into an OpenSim visual processing module for OpenSim visual processing, and finally realizing visual tracking of the human body posture through Matlab script processing.
2. The human body posture visual tracking system of claim 1, characterized in that: the inertial sensor is a nine-axis inertial sensor and comprises a three-axis gyroscope, a three-axis accelerometer and a three-axis magnetometer, and is used for acquiring angular velocity, acceleration and magnetic field intensity data of a human body during movement, transmitting the data to the MCU, and opening rectangular grooves around the MCU signal processing circuit board for installing elastic bands so as to fix the human body posture detection module at a human body measurement node.
3. The human body posture visual tracking system of claim 1, characterized in that: the MCU signal processing circuit comprises a main control chip and a basic peripheral circuit thereof.
4. The human body posture visual tracking system of claim 1, characterized in that: the wireless transmission modules adopt ZigBee protocol, and can be mutually networked when transmitting data of multiple nodes.
5. The human body posture visual tracking system of claim 1, characterized in that: the OpenSim visualization processing comprises the following steps: 1) installing a script interface of OpenSim in Matlab; 2) determining the model number of the used human body model; 3) determining the initial posture and the number of measuring nodes of the human body model; 4) the initial posture of the tested person is kept the same as that of the model, then posture change is made, and data are wirelessly transmitted to a PC (personal computer) end; 5) carrying out format processing on the collected multi-node data to match an OpenSim data format, and importing and applying the data to an OpenSim script in Matlab; 6) setting a central node as a basic calibration node for calibrating the data direction of other nodes; 7) and after the calibration is finished, executing visual script operation to generate a file with motion attitude information, reading the file by using OpenSim to display the motion process of the human body, and simultaneously displaying the angle change value between each node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110736567.5A CN113536552A (en) | 2021-06-30 | 2021-06-30 | Human body posture visual tracking system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110736567.5A CN113536552A (en) | 2021-06-30 | 2021-06-30 | Human body posture visual tracking system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113536552A true CN113536552A (en) | 2021-10-22 |
Family
ID=78097348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110736567.5A Pending CN113536552A (en) | 2021-06-30 | 2021-06-30 | Human body posture visual tracking system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113536552A (en) |
-
2021
- 2021-06-30 CN CN202110736567.5A patent/CN113536552A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Daponte et al. | State of the art and future developments of measurement applications on smartphones | |
| KR101509472B1 (en) | Motion parameter determination method and device and motion auxiliary equipment | |
| Zhang et al. | Quaternion-based Kalman filter with vector selection for accurate orientation tracking | |
| KR101768958B1 (en) | Hybird motion capture system for manufacturing high quality contents | |
| CN111091587B (en) | Low-cost motion capture method based on visual markers | |
| CN109737941B (en) | Human body motion capture method | |
| CN102245100A (en) | Graphical representations | |
| US11010917B2 (en) | Systems and methods for position and pose determination and tracking | |
| CN110327048B (en) | Human upper limb posture reconstruction system based on wearable inertial sensor | |
| CN103136912A (en) | Moving posture capture system | |
| CN203039726U (en) | Human body three-dimensional posture identifying system | |
| CN108279773B (en) | Data glove based on MARG sensor and magnetic field positioning technology | |
| Lin et al. | Development of an ultra-miniaturized inertial measurement unit WB-3 for human body motion tracking | |
| Wang et al. | Pose-invariant inertial odometry for pedestrian localization | |
| CN114926378B (en) | Method, system, device and computer storage medium for sound source tracking | |
| Choi et al. | Development of a low-cost wearable sensing glove with multiple inertial sensors and a light and fast orientation estimation algorithm | |
| CN108151738A (en) | Codified active light marked ball with attitude algorithm | |
| RU121947U1 (en) | TRAFFIC CAPTURE SYSTEM | |
| CN108280423A (en) | Strong appearance system based on human body attitude identification | |
| CN106970705A (en) | Motion capture method, device and electronic equipment | |
| CN204889937U (en) | Oral cavity endoscope | |
| CN112486331A (en) | IMU-based three-dimensional space handwriting input method and device | |
| US10748320B2 (en) | Animation simulation of biomechanics | |
| CN111382701A (en) | Motion capture method, motion capture device, electronic equipment and computer-readable storage medium | |
| 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 |