Animal action detection method and system and animal wearable device

Abstract

The invention provides an animal action detection method and system and animal wearable equipment. The animal action detection method comprises the following steps: acquiring animal attitude data through an attitude detection module of the animal wearable equipment; spectrum analysis is conducted on the posture data, behavior action recognition of the animal is conducted according to a spectrum analysis result, so that action data are obtained, and the action data comprise walking action data, running action data, grass eating action data, rumination action data and still lying action data; and reporting the action data to a server according to a preset reporting rule. By means of the animal motion detection method and device, the calculation complexity and power consumption of animal motion detection are reduced, and animal motion detection can be directly completed in the control unit of the animal wearable device.

Description

Technical Field

[0001] The present application relates to the field of behavior recognition technology, and in particular to an animal motion detection method, system and animal wearable device. Background Art

[0002] In some areas, herders have large grasslands and mountains suitable for free-range cattle. Depending on the breed of cattle and the local environment, some are free-range from Spring Festival to autumn, while others are free-range all year round. The advantage of this method is that it reduces the cost of fodder and labor, but the disadvantage is that cattle may get lost, walk out of their own pastures, get sick or injured, and need to be inspected regularly.

[0003] The detection of cows' walking, ruminating, eating grass and other movements is related to the physiological habits of cows, and these data can reflect the health of cows. At present, animal movement detection is to wear animal wearable devices such as ear tags or animal collars on cows, and collect animal posture data by configuring posture detection modules in animal wearable devices. The animal posture data is transmitted to the server, and the server performs movement detection through neural networks or machine learning models. This animal movement detection method is computationally complex and consumes a lot of power. Based on energy consumption and cost considerations, animal wearable devices usually use single-chip microcomputers as control units, so the movement detection method using neural networks or machine learning models is not suitable for execution in the control unit of animal wearable devices. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the purpose of the present application is to provide an animal motion detection method, system and animal wearable device, which reduce the computational complexity and power consumption of animal motion detection and can be directly executed in the control unit of animal wearable devices such as ear tags or animal collars.

[0005] To achieve the above-mentioned purpose and other related purposes, the present application provides an animal motion detection method, which is applied to a control unit of an animal wearable device, and the animal motion detection method includes:

[0006] Collecting the posture data of the animal through the posture detection module of the animal wearable device;

[0007] Performing spectrum analysis on the posture data, and identifying the behavior of the animal according to the spectrum analysis result to obtain motion data, wherein the motion data includes walking motion data, running motion data, grazing motion data, ruminating motion data, and lying motion data;

[0008] The action data is reported to the server according to the preset reporting rules.

[0009] In an optional embodiment of the present application, spectrum analysis is performed on the posture data, and the behavior and action recognition of the animal is performed according to the spectrum analysis result to obtain the action data, including:

[0010] Performing Fourier transformation on the posture data to convert the posture data into a frequency domain signal;

[0011] The animal's behavior and action are recognized according to the frequency domain signal to obtain action data.

[0012] In an optional embodiment of the present application, the animal's behavior and action recognition is performed according to the frequency domain signal to obtain action data, including:

[0013] Calculate the sum of energy in the entire frequency range according to the frequency domain signal as the total power;

[0014] The behavior and action of the animal are identified according to the total power and a preset threshold range to obtain action data.

[0015] In an optional embodiment of the present application, reporting the action data to the server according to a preset reporting rule includes:

[0016] The action data is reported to the server at a preset time interval.

[0017] In an optional embodiment of the present application, it also includes:

[0018] When the motion data is lying motion data, obtaining the animal's respiratory rate data according to the posture data corresponding to the lying motion data;

[0019] And report the animal's respiratory rate data to the server according to preset reporting rules.

[0020] To achieve the above-mentioned purpose and other related purposes, the present application provides an animal motion detection system, characterized in that it is applied to a control unit of an animal wearable device, and the animal motion detection system includes:

[0021] A data acquisition module, used for collecting the posture data of the animal through the posture detection module of the animal wearable device;

[0022] An action recognition module is used to perform spectrum analysis on the posture data, and perform behavioral action recognition of the animal according to the spectrum analysis result to obtain action data, wherein the action data includes walking action data, running action data, grazing action data, rumination action data and lying action data;

[0023] The data reporting module is used to report the action data to the server according to the preset reporting rules.

[0024] To achieve the above objectives and other related objectives, the present application provides an animal wearable device, including:

[0025] Posture detection module, used to collect animal posture data;

[0026] A control unit connected to the posture detection module, for acquiring the posture data, performing spectrum analysis on the posture data, and performing behavioral motion recognition of the animal according to the spectrum analysis result to acquire motion data, wherein the motion data includes walking motion data, running motion data, grazing motion data, ruminating motion data, and lying motion data;

[0027] A communication module connected to the control unit, the control unit is used to report the action data within a period of time to the server through the communication module according to a preset reporting rule;

[0028] A power supply module is used to supply power to the posture detection module, the control unit and the communication module.

[0029] In an optional embodiment of the present application, the posture detection module includes a three-axis acceleration sensor.

[0030] In an optional embodiment of the present application, the animal wearable device includes an ear tag or a collar.

[0031] In an optional embodiment of the present application, a positioning module is further included, which is connected to the control unit. The control unit obtains the real-time position data of the animal through the positioning module and reports the real-time position data of the animal to the server.

[0032] In an optional embodiment of the present application, the control unit includes a single chip microcomputer.

[0033] In an optional embodiment of the present application, the communication module includes a 4G module, a 5G module or a LoRa module.

[0034] The animal motion detection solution of the present application collects the posture data of the animal through the posture detection module of the animal wearable device; performs spectrum analysis on the posture data, and identifies the animal's behavioral motion according to the spectrum analysis results to obtain motion data, wherein the motion data includes walking motion data, running motion data, grazing motion data, rumination motion data and lying motion data; the motion data is reported to the server according to preset reporting rules, thereby reducing the computational complexity and power consumption of animal motion detection, and can directly complete the animal motion detection in the control unit of the animal wearable device such as ear tag or animal collar. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Shown is a flow chart of an animal action detection method according to an embodiment of the present application;

[0036] Figure 2aDisplayed as frequency domain signal of the cow’s posture data when it turns its ears;

[0037] Figure 2b They are shown as frequency domain signals of the posture data of cattle when eating grass;

[0038] Figure 2c They are shown as frequency domain signals of the posture data of the cow when walking;

[0039] Figure 2d They are shown as frequency domain signals of the posture data of cattle when ruminating;

[0040] Figure 3 Shown is a structural block diagram of an animal motion detection system according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following describes the implementation of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementations, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application.

[0042] See also Figure 1-3 It should be noted that the illustrations provided in this embodiment are only used to illustrate the basic concept of the present application in a schematic manner, and the illustrations only show the components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0043] The present application discloses an animal wearable device capable of performing animal positioning and motion recognition. The animal wearable device may be an ear tag or a collar. The ear tag is worn on the ear of the animal, and the collar is set on the neck of the animal. The animal wearable device includes a posture detection module, a control unit, a communication module, a power supply module, and a positioning module; the control unit is respectively connected to the posture detection module, the communication module, and the positioning module, and the power supply module is respectively powered by the posture detection module, the control unit, the communication module, and the positioning module.

[0044] The posture detection module is used to collect the posture data of the animal. The posture detection module can use a low-power three-axis acceleration sensor, which not only has low energy consumption and an operating current of less than 2μA, but also has low cost and small size, and is easy to integrate into animal wearable devices to further reduce the energy consumption of animal wearable devices. It should be noted that in other embodiments, an energy-consuming inertial measurement unit IMU, a six-axis sensor, a nine-axis sensor, etc. can also be used instead.

[0045] The communication module is used to communicate with the server and report the relevant data of the animal wearable device to the server. The communication module can be a 4G module, a 5G module or a LoRa module. The positioning module is used to obtain the real-time location of the animal. The positioning module is a GPS positioning module.

[0046] The control unit is used to obtain the posture data, perform spectrum analysis on the posture data, and identify the animal's behavior based on the spectrum analysis results to obtain action data, including walking action data, running action data, grazing action data, rumination action data, and lying action data, and report the action data within a period of time to the server through the communication module according to the preset reporting rules; and is also used to obtain the animal's real-time position data through the positioning module, and report the animal's real-time position data to the server. The control unit can use a single-chip microcomputer.

[0047] Figure 1 The flowchart of the animal motion detection method of the embodiment of the present application is shown, which is applied to the control unit of the animal wearable device. Figure 1 , the animal motion detection method comprises the following steps:

[0048] Step S10, collecting the posture data of the animal through the posture detection module of the animal wearable device;

[0049] Step S20, performing spectrum analysis on the posture data, and performing behavioral action recognition of the animal according to the spectrum analysis result to obtain action data, wherein the action data includes walking action data, running action data, grazing action data, rumination action data and lying action data;

[0050] Step S30: reporting the action data to the server according to preset reporting rules.

[0051] The animal motion detection of the present application can be applied to the motion detection of cattle, sheep and other animals in a free-range state, and the result of the motion detection can be reported to the server as data for monitoring the health status of the animal. The following will be described using free-range cattle as an example.

[0052] First, execute step S10 to collect the posture data of the animal through the posture detection module of the animal wearable device. When performing action detection of cattle, it is necessary to first obtain the posture data of the cattle. In this embodiment, the posture detection module in the ear tag is used to collect the posture data of the cattle, and the posture data is sent to the control unit, wherein the posture data may include acceleration data of the cattle when turning ears, walking, running, eating grass, ruminating, lying still, etc., and there are differences in the acceleration data corresponding to different behaviors.

[0053] Next, step S20 is executed to perform spectrum analysis on the posture data, and to identify the animal's behavior based on the spectrum analysis results to obtain motion data, which includes walking motion data, running motion data, grazing motion data, rumination motion data and lying motion data.

[0054] Since there are spectral differences in the acceleration data of cattle when they perform different behaviors such as turning their ears, walking, running, eating grass, ruminating, and lying still, the control unit of the present application can first perform a spectrum analysis on the cattle's posture data, and then identify the cattle's behavioral actions based on the spectral analysis results to obtain action data corresponding to the cattle's behavior, and the action data may include walking action data, running action data, eating grass action data, ruminating action data, lying still action data and other data.

[0055] In a specific embodiment of the present application, when performing spectral analysis on the posture data and identifying the behavior of the animal based on the spectral analysis results to obtain the motion data, the posture data can first be subjected to Fourier transformation to convert the posture data into a frequency domain signal; then, the animal's behavior is identified based on the frequency domain signal to obtain the motion data.

[0056] Figure 2a-2d The frequency domain signals of the posture data collected by the ear tags when the cow is turning its ears, eating grass, walking, and ruminating are shown respectively. Figure 2a-2d It can be seen that when the cow turns its ears, the power in the frequency spectrum shows a decreasing trend with the increase of frequency, and the total power is greater than 1000; when the cow is eating grass, the power is equivalent at 1-2HZ, and then decreases significantly, with the total power between 660 and 700. When the cow walks, the power in the frequency spectrum shows a decreasing trend with the increase of frequency, and the total power is between 600 and 660. When the cow ruminates, the power is evenly distributed in the frequency spectrum, and the total power is between 550 and 600. It should be noted that when the collar is used as an animal wearable device, due to the different installation positions of the collar and the ear tag, the collected posture data is different, which leads to differences in the frequency signal.

[0057] It can be seen that the total power of the frequency domain signal of the posture data of the cattle in the entire frequency range is different when they have different behavioral actions, so the difference in total power can be used as the basis for identifying different actions. Specifically, when the animal's behavioral action is identified based on the frequency domain signal to obtain action data, the sum of the energy in the entire frequency range can be calculated based on the frequency domain signal as the total power; then the animal's behavioral action is identified based on the total power and the preset threshold range to obtain action data, wherein different actions of the same free-range animal correspond to different preset threshold ranges, and the same action of different free-range animals corresponds to different preset threshold ranges.

[0058] Finally, step S30 is executed to report the action data to the server according to a preset reporting rule, for example, the action data may be reported to the server at a preset time interval, or the action data may be reported to the server synchronously when the position data is reported.

[0059] The animal detection method of the present application also includes, when the action data is lying action data, the animal's respiratory rate data can be obtained according to the posture data corresponding to the lying action data; and the animal's respiratory rate data is reported to the server according to preset reporting rules to monitor the health status of the cattle.

[0060] For the same cow, its walking motion data, running motion data, grazing motion data, rumination motion data, lying motion data, and breathing rate data are relatively stable data in a period of time. If there are large changes, it means that there is a problem with the cow's health. Therefore, the server can monitor the cow's health status based on the reported motion data.

[0061] Specifically, cattle are ruminants, and the grass they eat goes into their first stomach. In a safe place, the cattle then regurgitate the grass, chew it in their mouths, and swallow it into their second stomach. The time and frequency of cattle rumination vary depending on factors such as the breed, age, gender, breeding environment, and feed type of the cattle. Generally speaking, healthy cattle begin to ruminate 30 minutes to 1 hour after they stop eating, ruminating for about 8 hours a day and ruminating 10 to 15 times. By testing the daily rumination time of cattle, you can find out whether the cattle are healthy.

[0062] The time cattle spend grazing varies depending on factors such as breed, age, gender, feeding environment, and feed type. Generally speaking, cattle need to spend 6 to 8 hours grazing every day, with a larger intake in the morning and evening. Detecting the time cattle spend grazing every day can indicate the health of cattle.

[0063] Cows go into estrus every 20 days, and their walking and running movements increase significantly during estrus. Detecting the amount of walking movements of cows can guide the timing of artificial insemination.

[0064] The respiratory rate of cattle can be affected by many factors, such as breed, age, sex, health status, ambient temperature and activity level. Generally speaking, adult cattle breathe 10 to 30 times per minute, and calves breathe 20 to 50 times per minute. If a cattle's respiratory rate is found to be abnormal, further examination and treatment may be required, such as fever.

[0065] Since cattle often eat grass and ruminate, if a cattle is stolen, the thief will take off the ear tag or collar and throw it on the ground or hang it on a tree. The ear tag or collar will not detect the characteristic action of cattle eating grass and ruminating. If it takes a long time, the alarm information can be prompted to the owner of the cattle, and the owner of the cattle can rush to the scene and call the police in the shortest time. Therefore, the free-range animal positioning method can also send an alarm information to the owner of the free-range animal in time when the grazing action data and rumination action data of the free-range animal are not detected for a long time. So that the owner of the free-range animal can rush to the scene or call the police in the shortest time.

[0066] As shown in FIG. 2 , an embodiment of the present application further discloses an animal motion detection system 10 , which includes a data acquisition module 11 , a motion recognition module 12 and a data reporting module 13 .

[0067] The data acquisition module 11 is used to collect the posture data of the animal through the posture detection module of the animal wearable device;

[0068] The action recognition module 12 is used to perform spectrum analysis on the posture data, and perform behavioral action recognition of the animal according to the spectrum analysis result to obtain action data, wherein the action data includes walking action data, running action data, grazing action data, rumination action data and lying action data;

[0069] The data reporting module 13 is used to report the action data to the server according to a preset reporting rule.

[0070] It should be noted that the animal motion detection system 10 of the present application is a system corresponding to the above-mentioned animal motion detection method, and the functional modules in the animal motion detection system 10 respectively correspond to the corresponding steps in the animal motion detection method. The animal motion detection system 10 of the present application can be implemented in conjunction with the animal motion detection method. The relevant technical details mentioned in the animal motion detection method of the present application are still valid in the animal motion detection system 10, and in order to reduce repetition, they will not be repeated here. Accordingly, the relevant technical details mentioned in the animal motion detection system 10 of the present application can also be applied to the above-mentioned animal motion detection method.

[0071] It should be noted that the above-mentioned functional modules can be fully or partially integrated into one physical entity during actual implementation, or they can be physically separated. And these units can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some units can also be implemented in the form of processing elements calling software, and some units can be implemented in the form of hardware. In addition, these units can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the control unit or instructions in the form of software.

[0072] To summarize, the animal motion detection method, system and animal wearable device of the present application collect the posture data of the animal through the posture detection module of the animal wearable device; perform spectral analysis on the posture data, and identify the behavioral motion of the animal based on the spectral analysis results to obtain motion data, wherein the motion data includes walking motion data, running motion data, grazing motion data, rumination motion data and lying motion data, and the motion data is reported to the server according to preset reporting rules, thereby reducing the computational complexity and power consumption of animal motion detection, and can directly complete the animal motion detection in the control unit of the animal wearable device such as ear tag or animal collar.

[0073] In the description herein, many specific details are provided, such as examples of parts and / or methods, to provide a complete understanding of the embodiments of the present application. However, those skilled in the art will recognize that the embodiments of the present application can be practiced without one or more specific details or by other devices, systems, components, methods, parts, materials, parts, etc. In other cases, known structures, materials or operations are not specifically shown or described in detail to avoid blurring the aspects of the embodiments of the present application.

[0074] The above description of the embodiment shown in the application (including the content described in the abstract of the specification) is not intended to be exhaustive or to limit the application to the precise form disclosed herein. Although the specific embodiment of the application and the example of the application are described herein for the purpose of illustration only, as those skilled in the art will recognize and understand, various equivalent modifications can be within the spirit and scope of the application. As pointed out, these modifications can be made to the application according to the above description of the embodiment described in the application, and these modifications will be within the spirit and scope of the application.

[0075] This paper has generally described systems and methods as details that are helpful for understanding the application. In addition, various specific details have been given to provide an overall understanding of the embodiments of the application. However, those skilled in the relevant art will recognize that the embodiments of the application can be put into practice without one or more specific details, or put into practice using other devices, systems, accessories, methods, assemblies, materials, parts, etc. In other cases, known structures, materials and / or operations are not particularly shown or described in detail to avoid confusion in various aspects of the embodiments of the application.

[0076] Thus, although the present application has been described herein with reference to specific embodiments thereof, freedom of modification, various changes and substitutions are also within the above disclosure, and it should be understood that in some cases, some features of the present application will be adopted without corresponding use of other features without departing from the scope and spirit of the proposed invention. Therefore, many modifications may be made to adapt specific environments or materials to the substantial scope and spirit of the present application. The present application is not intended to be limited to the specific terms used in the claims below and / or the specific embodiments disclosed as the best mode for carrying out the present application, but the present application will include any and all embodiments and equivalents falling within the scope of the appended claims. Thus, the scope of the present application will be determined solely by the appended claims.

Claims

1. A method for detecting animal movements, characterized in that: A control unit applied to an animal wearable device, wherein the animal motion detection method comprises: Collecting the posture data of the animal through the posture detection module of the animal wearable device; Performing spectrum analysis on the posture data, and identifying the behavior of the animal according to the spectrum analysis result to obtain motion data, wherein the motion data includes walking motion data, running motion data, grazing motion data, ruminating motion data, and lying motion data; The action data is reported to the server according to the preset reporting rules.

2. The animal motion detection method according to claim 1, characterized in that: Performing spectrum analysis on the posture data, and identifying the behavior of the animal according to the spectrum analysis result to obtain the action data, including: Performing Fourier transformation on the posture data to convert the posture data into a frequency domain signal; The animal's behavior and action are recognized according to the frequency domain signal to obtain action data.

3. The animal motion detection method according to claim 2, characterized in that: Performing behavioral action recognition of the animal according to the frequency domain signal to obtain action data includes: Calculate the sum of energy in the entire frequency range according to the frequency domain signal as the total power; The behavior and action of the animal are identified according to the total power and a preset threshold range to obtain action data.

4. The animal motion detection method according to claim 1, characterized in that: Reporting the action data to the server according to the preset reporting rules includes: The action data is reported to the server at a preset time interval.

5. The animal motion detection method according to claim 1, characterized in that: Also includes: When the motion data is lying motion data, obtaining the animal's respiratory rate data according to the posture data corresponding to the lying motion data; And report the animal's respiratory rate data to the server according to preset reporting rules.

6. An animal motion detection system, characterized in that: Applied to a control unit of an animal wearable device, the animal motion detection system comprises: A data acquisition module, used for collecting the posture data of the animal through the posture detection module of the animal wearable device; An action recognition module is used to perform spectrum analysis on the posture data, and perform behavioral action recognition of the animal according to the spectrum analysis result to obtain action data, wherein the action data includes walking action data, running action data, grazing action data, rumination action data and lying action data; The data reporting module is used to report the action data to the server according to the preset reporting rules.

7. An animal wearable device, characterized in that: include: Posture detection module, used to collect animal posture data; A control unit connected to the posture detection module, for acquiring the posture data, performing spectrum analysis on the posture data, and performing behavioral motion recognition of the animal according to the spectrum analysis result to acquire motion data, wherein the motion data includes walking motion data, running motion data, grazing motion data, ruminating motion data, and lying motion data; A communication module connected to the control unit, the control unit is used to report the action data within a period of time to the server through the communication module according to a preset reporting rule; A power supply module is used to supply power to the posture detection module, the control unit and the communication module.

8. The animal wearable device according to claim 7, characterized in that: The posture detection module includes a three-axis acceleration sensor.

9. The animal wearable device according to claim 7, characterized in that: The animal wearable device includes an ear tag or a collar.

10. The animal wearable device according to claim 7, characterized in that: It also includes a positioning module connected to the control unit. The control unit obtains the real-time position data of the animal through the positioning module and reports the real-time position data of the animal to the server.

Images