A wireless human body posture marking system and its working method

Abstract

The present invention provides a wireless human body posture marking system and its working method, which relates to the technical field of human body posture recognition. The system includes a data receiver, a handheld remote control and an inertial measurement module that are communicatively connected to each other. There are multiple inertial measurement modules, which are respectively installed on the set parts of the human body and are used to collect the posture data of the set parts. The handheld remote control includes a plurality of buttons, and in response to the user's selection operation on the buttons, it is used to record the action tags corresponding to the buttons. The data receiver is used to receive the posture data and the action tags, and generate sample data with tags. Compared with the existing method of recording the human body posture data after the human body finishes an action and then manually tagging and classifying the data, through this system, accurate and usable sample data can be quickly generated for subsequent training of the human body posture recognition network model.

Description

Technical Field

[0001] The present invention belongs to the technical field of human body posture recognition, and particularly relates to a wireless marking system for human body postures and a working method thereof. Background Art

[0002] The statements in this part only provide background technical information related to the present invention, and do not necessarily constitute the prior art that is already well-known to those of ordinary skill in the art.

[0003] Due to the rapid development of deep learning algorithms in recent years, some scholars have attempted to apply them to motion posture recognition based on inertial devices. In deep learning, the training of neural networks requires a large amount of data sets, and only then can good results be achieved in practical applications. The production of data sets is actually also completed manually, which is a very time-consuming and laborious task. Some researchers run their neural network models on open-source data sets, which are convenient for comparing their own methods with other methods. However, in some specific problems, private data sets can more significantly improve the algorithm effect. Therefore, making one's own data set is of great significance for model application.

[0004] For the recognition of human body postures, the existing data production method is to install an inertial measurement module at the place where the human body needs to be detected. After the human body finishes an action, the human body posture data is recorded, and then the data is manually labeled and classified. This method takes a long time and the produced data set is inaccurate. Summary of the Invention

[0005] In order to solve the above problems, in the first aspect of the present invention, a wireless marking system for human body postures is provided. By means of a set data receiver, a hand-held remote control and an inertial measurement module, human body posture information and corresponding action labels can be obtained in a timely manner, and accurate and usable sample data can be generated.

[0006] To achieve the above object, the present invention mainly includes the following aspects:

[0007] In the first aspect, an embodiment of the present invention provides a wireless marking system for human body postures, including a data receiver, and a hand-held remote control and an inertial measurement module that are communicatively connected to the data receiver. There are multiple inertial measurement modules, which are respectively installed at set parts of the human body and are used for collecting posture data of the set parts; the hand-held remote control includes multiple buttons, and in response to a user's selection operation on the buttons, is used for recording action labels corresponding to the buttons; the data receiver is used for receiving the posture data and action labels, and generating sample data with labels.

[0008] In a possible implementation, the data receiver is further connected to a terminal device, and the terminal device is used to store the sample data generated by the data receiver.

[0009] In a possible implementation, the data receiver and the terminal device are connected through a serial port.

[0010] In a possible implementation, the inertial measurement module includes a microprocessor, and a three-axis acceleration sensor, a three-axis angular velocity sensor, a lithium battery module, and a wireless communication module that are electrically connected to the microprocessor.

[0011] Second, an embodiment of the present invention further provides a working method based on the human body posture wireless marking system described in the first aspect and any possible implementation of the first aspect above, including:

[0012] Establish a data communication network between the data receiver, the handheld remote control, and the inertial measurement module;

[0013] The data receiver broadcasts a data access request at preset intervals;

[0014] After receiving the data access request, the handheld remote control feeds back an action tag to the data receiver; and after receiving the data access request, the inertial measurement module feeds back posture data to the data receiver;

[0015] The data receiver generates tagged sample data according to the posture data and the action tag.

[0016] In a possible implementation, it further includes: transmitting the tagged sample data generated by the data receiver to the terminal device, storing the sample data through the terminal device, and using it for the training of the human body posture recognition network model.

[0017] In a possible implementation, the handheld remote control includes a plurality of buttons, and each button corresponds to an action tag; in response to the user's selection operation on the button, the action tag corresponding to the button is recorded.

[0018] In a possible implementation, the data receiver and the handheld remote control, the inertial measurement module adopt a wireless communication method; wherein, the wireless communication method includes: Bluetooth, WiFi, 2.4G, and 5G.

[0019] In a possible implementation, a private communication protocol is established, and the data receiver and the handheld remote control perform the transmission of relevant data according to the established private communication protocol; and the data receiver and the inertial measurement module perform the transmission of relevant data according to the established private communication protocol in the middle.

[0020] In a possible implementation manner, the human body posture data includes the acceleration, angular velocity, and quaternion of a set part of the human body.

[0021] The above one or more technical solutions have the following beneficial effects:

[0022] The present invention provides a wireless human body posture marking system. Through an inertial measurement module installed at a set part of the human body, posture data of the set part is collected, and an action label selected by a user through pressing a key is recorded by a handheld remote control. A data receiver receives the posture data and the action label to generate sample data with labels. Through this system, accurate and available sample data can be quickly generated for subsequent training of a human body posture recognition network model. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The specification drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention.

[0024] Figure 1 is a schematic structural diagram of the wireless human body posture marking system provided in the first embodiment of the present invention;

[0025] Figure 2 is one of the schematic flowcharts of the working method of the wireless human body posture marking system provided in the second embodiment of the present invention;

[0026] Figure 3 is the second of the schematic flowcharts of the working method of the wireless human body posture marking system provided in the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The present invention will be further described below in conjunction with the drawings and embodiments.

[0028] It should be noted that the following detailed descriptions are all exemplary and are intended to provide a further description of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0029] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0030] Embodiment 1

[0031] As shown Figure 1 In the embodiment of the present invention, a wireless human body posture marking system is provided, including a data receiver 5, a hand-held remote controller 1 and an inertial measurement module which are communicatively connected to each other. There are multiple inertial measurement modules, which are respectively installed at set parts of the human body and used to collect posture data of the set parts; the hand-held remote controller 1 includes multiple buttons, and in response to a user's selection operation on the buttons, it is used to record action tags corresponding to the buttons; the data receiver 5 is used to receive the posture data and action tags, and generate sample data with tags.

[0032] In a specific implementation, taking the human arm posture as an example, there are multiple inertial measurement modules, including a first inertial measurement module 2, a second inertial measurement module 3 and a third inertial measurement module 4, which are respectively bound to the forearm, upper arm and shoulder to collect arm postures. Manually press the buttons on the hand-held remote controller 1, and each button corresponds to an action tag to record the action tag currently selected by the user. The data receiver 5 is communicatively connected to the hand-held remote controller 1 and the inertial measurement module respectively. The data receiver 5 is used to receive the posture data collected by each inertial measurement module and the action tags recorded by the hand-held remote controller 1, and generate sample data with tags. Through this system, accurate and available sample data can be quickly generated for subsequent training of the human body posture recognition network model.

[0033] As an optional implementation manner, the data receiver 5 is further connected to a terminal device 6, and the terminal device 6 is used to store the sample data generated by the data receiver. In this way, rich sample data can be provided for subsequent training of the human body posture recognition network model. Compared with the open-source data set, the sample data is for the specific problem of human body posture recognition, which is beneficial to improving the recognition accuracy of human body postures and significantly enhancing the recognition effect.

[0034] As an optional implementation manner, the data receiver 5 and the terminal device 6 are connected through a serial port. Here, the terminal device includes but is not limited to a computer, a laptop, etc.

[0035] As an optional implementation manner, the inertial measurement module includes a microprocessor, a three-axis acceleration sensor, a three-axis angular velocity sensor, a lithium battery module and a wireless communication module which are electrically connected to the microprocessor. The three-axis acceleration sensor is used to collect acceleration information of the set part of the human body, and the three-axis angular velocity sensor is used to collect angular velocity information of the set part of the human body and transmit it to the microprocessor for processing of the acceleration information and angular velocity information; the lithium battery module is used to supply power to the inertial measurement module; the wireless communication module is used to send the posture data to the data receiver.

[0036] The data receiver 5 communicates wirelessly with the handheld remote control 1 and the inertial measurement module; among them, the wireless communication method includes, but is not limited to: Bluetooth, WiFi, 2.4G, and 5G. In this way, the measurement process can be unrestricted by data lines.

[0037] In the system provided by the embodiments of the present invention, the measurement module is simply deployed, powered by its own independent power supply, and transmits data wirelessly, which can be unrestricted by the environment, and has flexible measurement and high reliability.

[0038] Embodiment 2

[0039] Based on the human body posture wireless marking system in Embodiment 1, the embodiments of the present invention provide a working method for the human body posture wireless marking system, as Figure 2 shown, which specifically includes the following steps:

[0040] S201: Establish a data communication network among the data receiver, the handheld remote control, and the inertial measurement module;

[0041] S202: The data receiver broadcasts a data access request at preset intervals;

[0042] S203: After receiving the data access request, the handheld remote control feeds back the action label to the data receiver; and, after receiving the data access request, the inertial measurement module feeds back the posture data to the data receiver;

[0043] S204: The data receiver generates labeled sample data according to the posture data and the action label.

[0044] In a specific implementation, first, a data communication network is established among the data receiver 5, the handheld remote control 1, and the inertial measurement module for data transmission between the data receiver 5 and the handheld remote control 1 and the inertial measurement module. The data receiver 5 broadcasts a data access request at preset intervals (such as 15 minutes, etc.); after receiving the data access request, the handheld remote control 1 feeds back the action label to the data receiver; and, after receiving the data access request, the inertial measurement module feeds back the posture data to the data receiver. The data receiver 5 combines the obtained posture data and the action label as a labeled sample, thereby generating labeled sample data.

[0045] As an optional implementation manner, the labeled sample data generated by the data receiver 5 is transmitted to the terminal device 6, and the terminal device 6 stores the sample data and is used for training the human body posture recognition network model, which can provide rich sample data for the subsequent training of the human body posture recognition network model.

[0046] As an optional implementation manner, the handheld remote controller 1 includes a plurality of buttons, and each button corresponds to an action label; in response to a user's selection operation on a button, the action label corresponding to the button is recorded. In a specific implementation, when the user completes the corresponding action, the user presses the corresponding button to provide an action label for the current action.

[0047] As an optional implementation manner, the data receiver 5 and the handheld remote controller 1 and the inertial measurement module adopt a wireless communication method; wherein, the wireless communication method includes: Bluetooth, WiFi, 2.4G, and 5G. Optionally, a private communication protocol is established, and the data receiver and the handheld remote controller perform transmission of relevant data according to the established private communication protocol; and, the data receiver and the inertial measurement module perform transmission of relevant data according to the established private communication protocol.

[0048] In a specific implementation, a 2.4G private communication protocol is established. All inertial measurement modules, including the handheld remote controller 1, belong to slave devices, and the data receiver 5 is the master device. The master device continuously sends broadcasts to access the slave devices. If the data of a slave device is ready after receiving the broadcast, the ready data is sent to the master device.

[0049] As an optional implementation manner, the human body posture data includes acceleration, angular velocity, quaternion, etc. of a set part of the human body. In a specific implementation, acceleration information of a set part of the human body is collected by a triaxial acceleration sensor, angular velocity information of the set part of the human body is collected by a triaxial angular velocity sensor, and is transmitted to a microprocessor for processing of the acceleration information and the angular velocity information to obtain human body posture data.

[0050] As Figure 3 shown, in a specific application, the working method of the human body posture wireless marking system includes: reading angular velocity, acceleration information, and label data of the remote controller; transmitting inertial measurement module (International Mathematical Union, IMU) data and label data of the remote controller to a data acquisition module; merging the label and the data in the data acquisition module; generating available samples and storing them in a computer. Compared with the existing method of recording human body posture data after a human body finishes an action and then manually classifying the data by tagging, the method provided by the embodiment of the present invention can quickly generate accurate and available sample data for subsequent training of a human body posture recognition network model.

[0051] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A wireless marking system for human body postures, characterized in that, It includes a data receiver, a handheld remote control and an inertial measurement module that are communicatively connected to each other. There are multiple inertial measurement modules, which are respectively installed on the set parts of the human body to collect the attitude data of the set parts. The handheld remote control includes multiple buttons, and in response to the user's selection operation on the buttons, it is used to record the action tags corresponding to the buttons. The data receiver is used to receive the attitude data and action tags, and generate sample data with tags.

2. The wireless body posture marking system according to claim 1, wherein The data receiver is also connected to a terminal device, and the terminal device is used to store the sample data generated by the data receiver.

3. The human body posture wireless marking system according to claim 2, characterized in that The data receiver and the terminal device are connected through a serial port.

4. The wireless human body posture marking system according to claim 1, characterized in that The inertial measurement module includes a microprocessor, and a three-axis acceleration sensor, a three-axis angular velocity sensor, a lithium battery module and a wireless communication module that are electrically connected to the microprocessor.

5. A working method of a human body posture wireless marking system according to any one of claims 1 to 4, characterized in that, It includes: Establish a data communication network among the data receiver, the handheld remote control and the inertial measurement module; The data receiver broadcasts a data access request at preset intervals. After receiving the data access request, the handheld remote control feeds back the action tag to the data receiver; and after receiving the data access request, the inertial measurement module feeds back the attitude data to the data receiver. The data receiver generates sample data with tags according to the attitude data and the action tags.

6. The working method of the human body posture wireless marking system according to claim 5, characterized in that, It also includes: Transmit the sample data with tags generated by the data receiver to the terminal device, store the sample data through the terminal device, and use it for the training of the human body attitude recognition network model.

7. The working method of the human body posture wireless marking system according to claim 5, characterized in that, The handheld remote control includes multiple buttons, and each button corresponds to an action tag; in response to the user's selection operation on the buttons, record the action tags corresponding to the buttons.

8. The working method of the human body posture wireless marking system according to claim 5, characterized in that, The data receiver and the handheld remote control, the inertial measurement module adopt a wireless communication method; among them, the wireless communication method includes: Bluetooth, WiFi, 2.4G and 5G.

9. The working method of the wireless marking system for human body postures according to claim 8, characterized in that, Establish a private communication protocol, and the data receiver and the handheld remote control transmit relevant data according to the established private communication protocol; and the data receiver and the inertial measurement module transmit relevant data according to the established private communication protocol.

10. The working method of the human body posture wireless marking system according to claim 5, characterized in that, The human body attitude data includes the acceleration, angular velocity and quaternion of the set parts of the human body.

Images