Method, device and equipment for monitoring the number of head movements of grazing cattle and medium

Abstract

The application provides a grazing cattle head action frequency monitoring method, device, equipment and medium, and belongs to the technical field of animal husbandry. The method comprises the following steps: acquiring multiple initial accelerations of the head of the grazing cattle within a first time sequence; performing vertical acceleration extraction in the multiple initial accelerations according to a second time sequence to obtain multiple vertical accelerations; performing data feature extraction based on the multiple vertical accelerations to obtain multiple incremental data and multiple decremental data; and performing grazing cattle head action frequency conversion based on the multiple incremental data and the multiple decremental data to obtain the grazing cattle head action frequency. The grazing cattle head action frequency monitoring method provided by the application can realize automatic monitoring of the number of times of lifting and lowering the head of the grazing cattle when grazing, solves the difficulty in calculating the behavior quantitative index of the grazing cattle, and thus improves the monitoring efficiency of the number of times of lifting and lowering the head of the grazing cattle when grazing.

Description

Technical Field

[0001] This invention relates to the field of animal husbandry technology, and in particular to a method, device, equipment and medium for monitoring the number of head movements of grazing cattle. Background Technology

[0002] In agricultural production, the health and feeding conditions of cattle have a crucial impact on meat quality and breeding efficiency. Among them, the number of times the cattle raise and lower their heads is an important indicator for assessing grazing behavior and feed intake under grazing conditions, thereby improving the precision management of grassland grazing.

[0003] Currently, the most common method for monitoring the number of head raises and lowers in grazing cattle is manual observation and counting. Existing intelligent devices for grazing behavior mainly use accelerometers, which identify different behaviors such as walking, lying down, and grazing by exploring the changes in acceleration values ​​under different grazing behaviors. This can achieve automatic monitoring of some behaviors, but because the applicable behavior recognition targets are different, it cannot yet monitor the number of head raises and lowers. Therefore, the existing methods for monitoring the number of head raises and lowers in grazing cattle are not efficient; in other words, the existing methods for monitoring the number of head movements in grazing cattle are not very efficient. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and medium for monitoring the number of head movements of grazing cattle, in order to solve the problem of low efficiency in monitoring the number of head movements of grazing cattle in the prior art.

[0005] In a first aspect, the present invention provides a method for monitoring the number of head movements of grazing cattle, comprising:

[0006] Obtain multiple initial accelerations of the grazing cattle's head within the first time series;

[0007] According to the second time series, vertical acceleration is extracted from the plurality of initial accelerations to obtain a plurality of vertical accelerations; the second time series is the time series of the grazing behavior of the grazing cattle within the first time series.

[0008] Based on the multiple vertical accelerations, data features are extracted to obtain multiple increasing data and multiple decreasing data.

[0009] Based on the multiple increasing data and the multiple decreasing data, the number of head movements of grazing cattle is converted to obtain the number of head movements of grazing cattle; the number of head movements of grazing cattle includes the number of times the grazing cattle raise their heads and the number of times the grazing cattle lower their heads.

[0010] In one embodiment, the step of extracting data features based on the plurality of vertical accelerations to obtain a plurality of increasing data and a plurality of decreasing data includes:

[0011] The multiple vertical accelerations are calculated using first-order difference calculations to obtain multiple difference values.

[0012] The multiple differences are multiplied sequentially by adjacent terms to obtain multiple product values;

[0013] The product value that is greater than zero among the plurality of product values ​​is determined as the target product value;

[0014] Based on the product of each objective, multiple increasing data points and multiple decreasing data points are determined.

[0015] In one embodiment, when determining multiple increasing data points and multiple decreasing data points based on each target product value, the following steps are performed for each target product value:

[0016] If both differences used to calculate the target product value are greater than zero, then the two differences are determined to be increasing data.

[0017] If both differences used to calculate the target product value are less than zero, then the two differences are determined to be decreasing data.

[0018] In one embodiment, the step of converting the number of head movements of grazing cattle based on the plurality of increasing data and the plurality of decreasing data to obtain the number of head movements of grazing cattle includes:

[0019] Divide the increasing data that satisfy the condition of increasing in a continuous time sequence into a group, and determine the number of increasing data groups;

[0020] Divide the decreasing data that satisfy the condition of decreasing in a continuous time sequence into a group, and determine the number of decreasing data groups;

[0021] The number of head movements of grazing cattle is converted based on the increasing number of data groups and the decreasing number of data groups to obtain the number of head movements of grazing cattle.

[0022] In one embodiment, the step of converting the number of head movements of grazing cattle based on the increasing number of data sets and the decreasing number of data sets to obtain the number of head movements of grazing cattle includes:

[0023] Multiply the number of incremental data sets by a preset multiple to obtain the number of times the grazing cattle raise their heads in the second time series;

[0024] Multiply the decreasing number of data sets by a preset multiple to obtain the number of times the grazing cattle lowered their heads in the second time series.

[0025] In one embodiment, the second time series is determined based on the following method:

[0026] The multiple initial accelerations are preprocessed to obtain multiple target accelerations;

[0027] The multiple target accelerations are input into the head action recognition model to obtain the head action recognition results output by the head action recognition model; the head action recognition results include head-raising and head-lowering recognition results when grazing cattle are performing grazing behavior; the head action recognition model is trained based on acceleration samples and the head behavior labels corresponding to the acceleration samples; the head behavior labels include head-raising behavior labels or head-lowering behavior labels;

[0028] The consecutive moments corresponding to the head movement recognition results are determined as the second time series.

[0029] In one embodiment, the head motion recognition model is trained in the following manner:

[0030] Obtain acceleration samples and frame image samples of the head of grazing cattle in the third time series;

[0031] Based on the frame image samples, the acceleration samples at the same time are labeled with behavior to obtain head behavior labels; the head behavior labels include head-raising behavior labels and head-lowering behavior labels of the grazing cattle samples when they are grazing.

[0032] The head action recognition model is obtained by training the model based on the acceleration samples and the head behavior labels.

[0033] Secondly, the present invention also provides a device for monitoring the number of head movements of grazing cattle, comprising:

[0034] The acquisition module is used to acquire multiple initial accelerations of the head of grazing cattle in the first time series;

[0035] An extraction module is used to extract vertical acceleration from the plurality of initial accelerations according to a second time sequence to obtain a plurality of vertical accelerations; the second time sequence is a time sequence of the grazing behavior of the grazing cattle within a first time sequence.

[0036] The data feature extraction module is used to extract data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data.

[0037] The head movement count conversion module is used to convert the number of herding cattle head movements based on the multiple increasing data and the multiple decreasing data to obtain the number of herding cattle head movements; the number of herding cattle head movements includes the number of times the herding cattle raise their heads and the number of times the herding cattle lower their heads.

[0038] Thirdly, the present invention provides an apparatus comprising an electronic device, the electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any of the above-described methods for monitoring the number of head movements of grazing cattle.

[0039] Fourthly, the present invention also provides a medium comprising a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described methods for monitoring the number of head movements of grazing cattle.

[0040] The present invention provides a method, apparatus, device, and medium for monitoring the number of head movements of grazing cattle. It acquires multiple initial accelerations of the grazing cattle's head in a first time series, further extracts multiple vertical accelerations in a second time series during grazing behavior, and uses these vertical accelerations to extract data features, obtaining multiple increasing and decreasing data points that reflect the trends of head-raising and head-lowering movements. Furthermore, based on these increasing and decreasing data points, it converts the number of head movements of the grazing cattle into the number of head-raising and head-lowering movements, thus achieving automated monitoring of the number of head-raising and head-lowering movements of grazing cattle. This solves the difficulty in calculating quantitative indicators of grazing cattle behavior and improves the monitoring efficiency of head-raising and head-lowering movements during grazing. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0042] Figure 1 This is one of the flowcharts illustrating the method for monitoring the number of head movements of grazing cattle provided by the present invention;

[0043] Figure 2 This is the second flowchart of the method for monitoring the number of head movements of grazing cattle provided by the present invention;

[0044] Figure 3 This is a schematic diagram of the structure of the head movement monitoring device for grazing cattle provided by the present invention;

[0045] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0047] It should be noted that, in the description of this invention, the term "comprising" or any other variations thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0048] The terms "first," "second," etc., used in this invention are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more.

[0049] The following is combined Figures 1-4 This invention describes the method, apparatus, equipment, and medium for monitoring the number of head movements of grazing cattle.

[0050] Figure 1 This is one of the flowcharts illustrating the method for monitoring the number of head movements in grazing cattle provided by the present invention. Figure 2 This is the second flowchart of the method for monitoring the number of head movements of grazing cattle provided by the present invention.

[0051] like Figure 1 As shown, the method for monitoring the number of head movements of grazing cattle provided by the present invention includes, but is not limited to, the following steps:

[0052] Step 100: Obtain multiple initial accelerations of the grazing cattle's head in the first time series;

[0053] Step 200: According to the second time series, extract the vertical acceleration from the multiple initial accelerations to obtain multiple vertical accelerations;

[0054] Step 300: Extract data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data;

[0055] Step 400: Based on the multiple increasing data and the multiple decreasing data, convert the number of head movements of the grazing cattle to obtain the number of head movements of the grazing cattle.

[0056] It should be noted that the method for monitoring the number of head movements of grazing cattle provided in this embodiment of the invention is based on a device for monitoring the number of head movements of grazing cattle. This method identifies the behavior of grazing cattle, specifically head-raising and head-lowering behaviors. It calculates the corresponding behavioral signal features using the first-order difference method, and then obtains an automatic calculation method for the number of head-raising and head-lowering movements by analyzing the relationship between the behavioral signal features and the actual behavior. Behavior recognition is fundamental, while counting the behavioral signal features is crucial, as this depends on the amplitude and frequency of the cattle's movements. This method can meet the needs of monitoring and identifying cattle behavior under grazing conditions, and also meets the needs of calculating quantitative indicators of cattle behavior under grazing conditions. This helps farmers accurately assess cattle feed intake, feed digestibility, and gain a more accurate understanding of cattle activity levels and behavioral habits. Therefore, this embodiment of the invention uses a device for monitoring the number of head movements of grazing cattle as the executing entity to describe the method for monitoring the number of head movements of grazing cattle.

[0057] Specifically, such as Figure 2 As shown, the following description is the process of obtaining the initial acceleration of the grazing cattle's head. Therefore, it can be understood that the grazing cattle head movement number monitoring device obtains multiple initial accelerations of the grazing cattle's head in the first time sequence.

[0058] It should be noted that the grazing cattle are equipped with wearable behavioral monitoring devices on their heads. These devices operate at a frequency of approximately 5Hz and include built-in accelerometers to collect the acceleration of the cattle's heads as they walk and graze. Accelerometers are devices that measure the acceleration of an object along three axes (x, y, and z), thus determining the magnitude and direction of the object's acceleration.

[0059] Furthermore, such as Figure 2 As shown, the following description is the head movement recognition process of grazing cattle. Therefore, it can be understood that the grazing cattle head movement number monitoring device preprocesses multiple initial accelerations in the first time series to obtain multiple target accelerations in the first time series.

[0060] Furthermore, the head movement frequency monitoring device for grazing cattle inputs multiple target accelerations into the head movement recognition model to obtain the head movement recognition results output by the head movement recognition model. The head movement recognition results include head-raising recognition results and head-lowering recognition results when grazing cattle are performing grazing behavior.

[0061] Therefore, it can be understood that the head movement monitoring device for grazing cattle identifies the head-raising and head-lowering recognition results of grazing cattle when they are grazing, among multiple target accelerations, through a head movement recognition model.

[0062] Furthermore, the head movement frequency monitoring device for grazing cattle determines the continuous moments corresponding to the head movement recognition results as the second time series.

[0063] It should be noted that when grazing cattle are grazing, they will lower their heads to eat grass and raise their heads to stop grazing. This is a continuous process. Therefore, after identifying the head-raising and head-lowering results of grazing cattle, the corresponding continuous moments are determined as the second time series. Furthermore, multiple vertical accelerations within the second time series are extracted from the initial acceleration to be used for subsequent quantitative calculations of indicators such as the number of times grazing cattle raise their heads and lower their heads.

[0064] Furthermore, such as Figure 2 As shown, the following description is the process of extracting vertical acceleration during grazing behavior. Therefore, it can be understood that the device for monitoring the number of head movements of grazing cattle extracts vertical acceleration from multiple initial accelerations according to the second time sequence, and obtains multiple vertical accelerations.

[0065] It should be noted that when extracting vertical acceleration, information such as the time, amplitude, and frequency of each head-up and head-down movement should be recorded. The accuracy and completeness of the data must be ensured during vertical acceleration extraction. For each head-up and head-down movement, all relevant sensor data should be recorded as much as possible to provide more accurate analysis and evaluation.

[0066] Furthermore, the head movement frequency monitoring device for grazing cattle extracts data features based on multiple vertical accelerations, obtaining multiple increasing data points and multiple decreasing data points.

[0067] Furthermore, the grazing cattle head movement frequency monitoring device converts the grazing cattle head movement frequency based on multiple increasing data and multiple decreasing data to obtain the grazing cattle head movement frequency, which includes the number of times the grazing cattle raise their heads and the number of times the grazing cattle lower their heads.

[0068] The present invention provides a method for monitoring the number of head movements of grazing cattle. It acquires multiple initial accelerations of the grazing cattle's head in a first time series, further extracts multiple vertical accelerations in a second time series during grazing behavior, and uses these vertical accelerations to extract data features, obtaining multiple increasing and decreasing data points that reflect the trends of head-raising and head-lowering movements. Furthermore, based on these increasing and decreasing data points, the method converts the number of head movements of the grazing cattle into the number of head-raising and head-lowering movements, thus achieving automated monitoring of the number of head-raising and head-lowering movements of grazing cattle. This solves the difficulty in calculating quantitative indicators of grazing cattle behavior and improves the monitoring efficiency of head-raising and head-lowering movements during grazing.

[0069] Furthermore, the second time series was determined based on the following method:

[0070] The multiple initial accelerations are preprocessed to obtain multiple target accelerations;

[0071] The multiple target accelerations are input into the head action recognition model to obtain the head action recognition results output by the head action recognition model; the head action recognition results include head-raising and head-lowering recognition results when grazing cattle are performing grazing behavior; the head action recognition model is trained based on acceleration samples and the head behavior labels corresponding to the acceleration samples; the head behavior labels include head-raising behavior labels or head-lowering behavior labels;

[0072] The consecutive moments corresponding to the head movement recognition results are determined as the second time series.

[0073] Specifically, the head movement monitoring device for grazing cattle preprocesses multiple initial accelerations to obtain multiple target accelerations.

[0074] It should be noted that the initial acceleration includes acceleration data along three axes, namely the x-axis, y-axis, and z-axis. The target acceleration is obtained by merging the acceleration data along these three axes and inputting it into the model.

[0075] In one embodiment, the acceleration data of the three axes are merged. The average acceleration value can be obtained by averaging the acceleration data of the three axes and then used as the target acceleration.

[0076] Furthermore, the grazing cattle head movement monitoring device inputs multiple target accelerations into the head movement recognition model to obtain the head movement recognition results output by the head movement recognition model. In other words, it identifies the head-raising and head-lowering recognition results of grazing cattle when they are grazing, among multiple target accelerations.

[0077] Furthermore, the head movement frequency monitoring device for grazing cattle determines the continuous moments corresponding to the head movement recognition results as the second time series.

[0078] This invention preprocesses multiple initial accelerations to obtain multiple target accelerations, which facilitates data processing by a head action recognition model. The head action recognition model accurately identifies the head actions of grazing cattle during grazing. Based on the head action recognition results, a second time series is determined in continuous time intervals. Multiple vertical accelerations within the second time series can then be extracted from the initial accelerations for subsequent quantitative calculations of indicators such as the number of times grazing cattle raise their heads and lower their heads. This enables automated monitoring of the number of times grazing cattle raise and lower their heads during grazing.

[0079] Furthermore, the head motion recognition model is trained in the following manner:

[0080] Obtain acceleration samples and frame image samples of the head of grazing cattle in the third time series;

[0081] Based on the frame image samples, the acceleration samples at the same time are labeled with behavior to obtain head behavior labels; the head behavior labels include head-raising behavior labels and head-lowering behavior labels of the grazing cattle samples when they are grazing.

[0082] The head action recognition model is obtained by training the model based on the acceleration samples and the head behavior labels.

[0083] Specifically, the head movement monitoring device for grazing cattle acquires acceleration samples and frame image samples of the grazing cattle's head in a third time series.

[0084] It should be noted that acceleration samples can be collected by wearable behavior monitoring devices, while frame image samples can be obtained by extracting frame images from videos of grazing cattle behavior taken by cameras. The selected cameras need to ensure that the head-raising and head-lowering movements of the grazing cattle are clearly visible in the videos they are taken.

[0085] Furthermore, the grazing cattle head movement frequency monitoring device determines the behavior of the grazing cattle samples based on frame image samples, and performs behavior labeling on the acceleration samples at the same time according to the behavior of the grazing cattle samples to obtain head behavior labels. The head behavior labels include head-raising behavior labels and head-lowering behavior labels of the grazing cattle samples when they are grazing.

[0086] Furthermore, the head movement frequency monitoring device for grazing cattle trains a model based on acceleration samples and head behavior labels to obtain a head movement recognition model.

[0087] This invention, based on frame image samples, performs behavioral annotation on acceleration samples at the same time to obtain head-raising and head-lowering behavior labels for grazing cattle samples. Further, based on the acceleration samples, head-raising behavior labels, and head-lowering behavior labels, a model is trained to obtain a head action recognition model. This model can be used to achieve accurate head action recognition, meeting the needs for monitoring and recognizing the head behavior of grazing cattle. Furthermore, after knowing the head action recognition results, quantitative indicators such as the number of times grazing cattle raise their heads and lower their heads can be calculated. This enables automated monitoring of the number of times grazing cattle raise and lower their heads, solving the difficulty in calculating quantitative indicators of grazing cattle behavior and thus improving the efficiency of monitoring the number of times grazing cattle raise and lower their heads.

[0088] Further, based on step 300, the data feature extraction based on the multiple vertical accelerations yields multiple increasing data points and multiple decreasing data points, including:

[0089] The multiple vertical accelerations are calculated using first-order difference calculations to obtain multiple difference values.

[0090] The multiple differences are multiplied sequentially by adjacent terms to obtain multiple product values;

[0091] The product value that is greater than zero among the plurality of product values ​​is determined as the target product value;

[0092] Based on the product of each objective, multiple increasing data points and multiple decreasing data points are determined.

[0093] It's important to note that by monitoring the grazing behavior of pasture cattle, when they raise or lower their heads, they generate a continuous and unchanging acceleration in the vertical direction (i.e., the z-axis acceleration direction). Furthermore, when walking and grazing, the cattle's heads move in a rhythmic up-and-down motion, resulting in fluctuating accelerations rather than constantly changing ones. Therefore, based on the head movement patterns of pasture cattle and the variation patterns of the accelerometer, the vertical acceleration can be imported into a programming language (such as Python), and the difference function in the programming language's database can be used to calculate the first-order difference of multiple vertical accelerations.

[0094] Specifically, such as Figure 2 As shown, the following description is the process of detecting increasing and decreasing data in vertical acceleration. Therefore, it can be understood that the grazing cattle head movement monitoring device performs first-order difference calculations on multiple vertical accelerations to obtain multiple difference values.

[0095] It should be noted that the first difference refers to the difference between two adjacent observations in time series data. For a time series {x1, x2, x3, ..., x...}, ... nIts first-order difference can be expressed as {Δx1, Δx2, Δx3, ..., Δx}. n-1}, where Δx i =x i+1 -x i 1≤i≤n-1. First-order differences are commonly used in time series analysis to help extract trends and changes in data. By calculating the first-order differences, the rate of change or increase / decrease of the original data at adjacent time points can be obtained, which helps in analyzing the trend characteristics of the data.

[0096] Furthermore, the head movement monitoring device for grazing cattle multiplies multiple differences by adjacent terms in sequence to obtain multiple product values.

[0097] In one embodiment, the multiple differences are represented as {Δx1, Δx2, Δx3, ..., Δx}. n-1}, multiplying multiple differences by adjacent terms in sequence, can be expressed as y i =Δx i ×Δx i+1 , where 1≤i≤n-2.

[0098] Furthermore, the head movement monitoring device for grazing cattle discards product values ​​that are less than zero among multiple product values.

[0099] Furthermore, the grazing cattle head movement frequency monitoring device determines the product value greater than zero among multiple product values ​​as the target product value. Further, based on each target product value, the grazing cattle head movement frequency monitoring device determines multiple increasing data and multiple decreasing data.

[0100] This invention employs a first-order difference calculation method to reflect the trend of vertical acceleration changes at adjacent time points. It further uses a multiplication method of adjacent terms to reflect the trends of head-raising and head-lowering movements of grazing cattle. Then, based on multiple increasing and decreasing data points, it converts the number of head movements of grazing cattle to obtain the number of times grazing cattle raise their heads and lower their heads. This achieves automated monitoring of the number of head-raising and head-lowering movements of grazing cattle, solving the difficulty in calculating quantitative indicators of grazing cattle behavior and thus improving the monitoring efficiency of head-raising and head-lowering movements of grazing cattle.

[0101] Furthermore, when determining multiple increasing and multiple decreasing data points based on each target product value, the following steps are performed for each target product value:

[0102] If both differences used to calculate the target product value are greater than zero, then the two differences are determined to be increasing data.

[0103] If both differences used to calculate the target product value are less than zero, then the two differences are determined to be decreasing data.

[0104] Specifically, if both differences used to calculate the target product value are greater than zero, the grazing cattle head movement monitoring device will determine the two differences as increasing data.

[0105] Furthermore, if both differences used to calculate the target product value are less than zero, the grazing cattle head movement monitoring device will determine the two differences as decreasing data.

[0106] In one embodiment, the above can be represented as follows:

[0107]

[0108] Where 1≤i≤n-1, and n is the number of all product values.

[0109] This invention extracts data features based on multiple vertical accelerations to obtain multiple increasing and decreasing data, which can reflect the trends of head-raising and head-lowering movements of grazing cattle. Furthermore, based on the multiple increasing and decreasing data, the number of head movements of grazing cattle is converted to obtain the number of times grazing cattle raise their heads and lower their heads, realizing automated monitoring of the number of head-raising and head-lowering movements of grazing cattle. This solves the difficulty in calculating quantitative indicators of grazing cattle behavior and improves the monitoring efficiency of the number of head-raising and head-lowering movements of grazing cattle.

[0110] Further, based on step 400, the conversion of the number of head movements of grazing cattle based on the plurality of increasing data and the plurality of decreasing data to obtain the number of head movements of grazing cattle includes:

[0111] Divide the increasing data that satisfy the condition of increasing in a continuous time sequence into a group, and determine the number of increasing data groups;

[0112] Divide the decreasing data that satisfy the condition of decreasing in a continuous time sequence into a group, and determine the number of decreasing data groups;

[0113] The number of head movements of grazing cattle is converted based on the increasing number of data groups and the decreasing number of data groups to obtain the number of head movements of grazing cattle.

[0114] It should be noted that, in one embodiment, in order to construct the conversion relationship between increasing and decreasing data and the number of times the head is raised and lowered, the time is first divided into groups of 1 minute. The number of increasing and decreasing groups and the number of times the head is raised and lowered are compared within 2 hours. After analysis, the results show that the number of times the head is raised per minute is a multiple of the number of groups of increasing data, and the number of times the head is lowered per minute is a multiple of the number of groups of decreasing data.

[0115] Specifically, such as Figure 2As shown, the following description is the process of determining the conversion relationship between increasing data and the number of head raises, and the conversion relationship between decreasing data and the number of head lowers. Therefore, it can be understood that the grazing cattle head movement monitoring device divides the increasing data that satisfy the condition of increasing in a continuous time sequence into a group, obtains multiple increasing data groups, and determines the number of increasing data groups.

[0116] Furthermore, the grazing cattle head movement frequency monitoring device divides the decreasing data that satisfy the condition of decreasing within a continuous time sequence into a group, obtains multiple decreasing data groups, and determines the number of decreasing data groups.

[0117] This invention, through grouping multiple increasing data points and multiple decreasing data points, obtains the number of increasing data groups and the number of decreasing data groups. Then, it determines the conversion relationship between the number of increasing data groups and the number of head-raising times, and the conversion relationship between the number of decreasing data groups and the number of head-lowering times. Based on these two conversion relationships, it can calculate the number of head-raising and head-lowering times of grazing cattle, and simultaneously achieve automated monitoring of these times. This solves the difficulty in calculating quantitative indicators of grazing cattle behavior, thereby improving the monitoring efficiency of head-raising and head-lowering times of grazing cattle.

[0118] Further, the conversion of the number of head movements of grazing cattle based on the increasing number of data groups and the decreasing number of data groups to obtain the number of head movements of grazing cattle includes:

[0119] Multiply the number of incremental data sets by a preset multiple to obtain the number of times the grazing cattle raise their heads in the second time series;

[0120] Multiply the decreasing number of data sets by a preset multiple to obtain the number of times the grazing cattle lowered their heads in the second time series.

[0121] Specifically, the grazing cattle head movement monitoring device multiplies the incremental data set number by a preset multiple to obtain the number of times the grazing cattle raise their heads in the second time series. The preset multiple is set according to the actual situation.

[0122] In one embodiment, the number of times grazing cattle raise their heads in the second time series is calculated as follows:

[0123] Hu = k × Nu

[0124] Where Hu represents the number of times the user looks up, k represents the preset multiplier, and Nu represents the number of incremental data sets.

[0125] Furthermore, the head movement monitoring device for grazing cattle multiplies the decreasing number of data sets by a preset multiple to obtain the number of times the grazing cattle lower their heads in the second time series.

[0126] In one embodiment, the number of times grazing cattle raise their heads in the second time series is calculated as follows:

[0127] Hd=k×Nd

[0128] Where Hd is the number of times the head is lowered, k is the preset multiple, and Nd is the number of decreasing data sets.

[0129] This invention, based on the conversion relationship between the number of increasing data sets and the number of head-raising times, and the conversion relationship between the number of decreasing data sets and the number of head-lowering times, converts the number of increasing data sets into the number of head-raising times and the number of decreasing data sets into the number of head-lowering times. This enables the calculation of the number of head-raising and head-lowering times of grazing cattle, and simultaneously achieves automated monitoring of the number of head-raising and head-lowering times of grazing cattle. This solves the difficulty in calculating quantitative indicators of grazing cattle behavior, thereby improving the monitoring efficiency of the number of head-raising and head-lowering times of grazing cattle.

[0130] Furthermore, the present invention also provides a device for monitoring the number of head movements of grazing cattle.

[0131] Reference Figure 3 , Figure 3 This is a schematic diagram of the structure of the head movement monitoring device for grazing cattle provided by the present invention.

[0132] The device for monitoring the number of head movements of grazing cattle includes:

[0133] The acquisition module 310 is used to acquire multiple initial accelerations of the head of the grazing cattle in the first time series;

[0134] The extraction module 320 is used to extract vertical acceleration from the plurality of initial accelerations according to the second time sequence to obtain a plurality of vertical accelerations; the second time sequence is the time sequence of the grazing behavior of the grazing cattle in the first time sequence.

[0135] The data feature extraction module 330 is used to extract data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data.

[0136] The head movement count conversion module 340 is used to convert the number of herding cattle head movements based on the multiple increasing data and the multiple decreasing data to obtain the number of herding cattle head movements; the number of herding cattle head movements includes the number of times the herding cattle raise their heads and the number of times the herding cattle lower their heads.

[0137] The present invention provides a device for monitoring the number of head movements of grazing cattle. It acquires multiple initial accelerations of the grazing cattle's head in a first time series, further extracts multiple vertical accelerations in a second time series during grazing behavior, and uses these vertical accelerations to extract data features, obtaining multiple increasing and decreasing data points that reflect the trends of the grazing cattle's head-raising and head-lowering movements. Furthermore, based on these increasing and decreasing data points, it converts the number of head movements of the grazing cattle into the number of times the grazing cattle raise their heads and lower their heads, thus achieving automated monitoring of the number of head-raising and head-lowering movements of grazing cattle. This solves the difficulty in calculating quantitative indicators of grazing cattle behavior and improves the monitoring efficiency of the number of head-raising and head-lowering movements of grazing cattle.

[0138] Furthermore, the device for monitoring the number of head movements of grazing cattle is also used for:

[0139] The multiple initial accelerations are preprocessed to obtain multiple target accelerations;

[0140] The multiple target accelerations are input into the head action recognition model to obtain the head action recognition results output by the head action recognition model; the head action recognition results include head-raising and head-lowering recognition results when grazing cattle are performing grazing behavior; the head action recognition model is trained based on acceleration samples and the head behavior labels corresponding to the acceleration samples; the head behavior labels include head-raising behavior labels or head-lowering behavior labels;

[0141] The consecutive moments corresponding to the head movement recognition results are determined as the second time series.

[0142] Furthermore, the device for monitoring the number of head movements of grazing cattle is also used for:

[0143] Obtain acceleration samples and frame image samples of the head of grazing cattle in the third time series;

[0144] Based on the frame image samples, the acceleration samples at the same time are labeled with behavior to obtain head behavior labels; the head behavior labels include head-raising behavior labels and head-lowering behavior labels of the grazing cattle samples when they are grazing.

[0145] The head action recognition model is obtained by training the model based on the acceleration samples and the head behavior labels.

[0146] Furthermore, the data feature extraction module 330 is also used for:

[0147] The multiple vertical accelerations are calculated using first-order difference calculations to obtain multiple difference values.

[0148] The multiple differences are multiplied sequentially by adjacent terms to obtain multiple product values;

[0149] The product value that is greater than zero among the plurality of product values ​​is determined as the target product value;

[0150] Based on the product of each objective, multiple increasing data points and multiple decreasing data points are determined.

[0151] Furthermore, the data feature extraction module 330 is also used for:

[0152] If both differences used to calculate the target product value are greater than zero, then the two differences are determined to be increasing data.

[0153] If both differences used to calculate the target product value are less than zero, then the two differences are determined to be decreasing data.

[0154] Furthermore, the head movement count conversion module 340 is also used for:

[0155] Divide the increasing data that satisfy the condition of increasing in a continuous time sequence into a group, and determine the number of increasing data groups;

[0156] Divide the decreasing data that satisfy the condition of decreasing in a continuous time sequence into a group, and determine the number of decreasing data groups;

[0157] The number of head movements of grazing cattle is converted based on the increasing number of data groups and the decreasing number of data groups to obtain the number of head movements of grazing cattle.

[0158] Furthermore, the head movement count conversion module 340 is also used for:

[0159] Multiply the number of incremental data sets by a preset multiple to obtain the number of times the grazing cattle raise their heads in the second time series;

[0160] Multiply the decreasing number of data sets by a preset multiple to obtain the number of times the grazing cattle lowered their heads in the second time series.

[0161] It should be noted that the grazing cattle head movement count monitoring device provided by the present invention can execute the grazing cattle head movement count monitoring method described in any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0162] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 4As shown, the electronic device may include a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a method for monitoring the number of head movements of grazing cattle. This method includes: acquiring multiple initial accelerations of the grazing cattle's head in a first time series; extracting vertical accelerations from the multiple initial accelerations according to a second time series to obtain multiple vertical accelerations; the second time series is a time series of the grazing behavior of the grazing cattle in the first time series; extracting data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data; converting the number of head movements of grazing cattle based on the multiple increasing data and the multiple decreasing data to obtain the number of head movements of grazing cattle; the number of head movements of grazing cattle includes the number of times the grazing cattle raise their heads and the number of times the grazing cattle lower their heads.

[0163] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0164] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute the method for monitoring the number of head movements of grazing cattle provided in the above embodiments, the method comprising: acquiring multiple initial accelerations of the head of the grazing cattle in a first time series; extracting vertical accelerations from the multiple initial accelerations according to a second time series to obtain multiple vertical accelerations; the second time series being a time series of the grazing behavior of the grazing cattle in the first time series; extracting data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data; converting the number of head movements of the grazing cattle based on the multiple increasing data and the multiple decreasing data to obtain the number of head movements of the grazing cattle; the number of head movements of the grazing cattle includes the number of times the grazing cattle raise their heads and the number of times the grazing cattle lower their heads.

[0165] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the method for monitoring the number of head movements of grazing cattle provided in the above embodiments. The method includes: acquiring multiple initial accelerations of the head of the grazing cattle in a first time series; extracting vertical accelerations from the multiple initial accelerations according to a second time series to obtain multiple vertical accelerations; the second time series being a time series of the grazing behavior of the grazing cattle in the first time series; extracting data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data; converting the number of head movements of the grazing cattle based on the multiple increasing data and the multiple decreasing data to obtain the number of head movements of the grazing cattle; the number of head movements of the grazing cattle includes the number of times the grazing cattle raise their heads and the number of times the grazing cattle lower their heads.

[0166] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0167] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring the number of head movements in grazing cattle, characterized in that, include: Obtain multiple initial accelerations of the grazing cattle's head within the first time series; The grazing cattle are equipped with accelerometers on their heads to collect the initial acceleration of their heads when walking and grazing. The initial acceleration includes acceleration in the x-axis, y-axis, and z-axis directions. The first time series is a time series in which the accelerometer continuously collects the initial acceleration. According to the second time series, the acceleration components in the z-axis direction corresponding to the time of the second time series are extracted from the plurality of initial accelerations to obtain a plurality of vertical accelerations; the second time series is the time series of the grazing behavior of the grazing cattle in the first time series; Based on the multiple vertical accelerations, data features are extracted to obtain multiple increasing data and multiple decreasing data. Based on the multiple increasing data and the multiple decreasing data, the number of head movements of grazing cattle is converted to obtain the number of head movements of grazing cattle; the number of head movements of grazing cattle includes the number of times grazing cattle raise their heads and the number of times grazing cattle lower their heads. The second time series was determined based on the following method: Based on the initial accelerations described above, the head movement recognition results of grazing cattle during grazing behavior are determined. The consecutive moments corresponding to the head movement recognition results are determined as the second time series; The data feature extraction based on the multiple vertical accelerations yields multiple increasing data points and multiple decreasing data points, including: The multiple vertical accelerations are calculated using first-order difference calculations to obtain multiple difference values. The multiple differences are multiplied sequentially by adjacent terms to obtain multiple product values; The product value that is greater than zero among the plurality of product values ​​is determined as the target product value; Based on the comparison results of the two differences used to calculate each of the target product values ​​with zero, multiple increasing data and multiple decreasing data are determined; The process of converting the number of head movements of grazing cattle based on the multiple increasing data and the multiple decreasing data to obtain the number of head movements of grazing cattle includes: Divide the increasing data that satisfy the condition of increasing in a continuous time sequence into a group, and determine the number of increasing data groups; Divide the decreasing data that satisfy the condition of decreasing in a continuous time sequence into a group, and determine the number of decreasing data groups; Based on the increasing number of data sets, the number of head movements of the grazing cattle is converted to obtain the number of times the grazing cattle raise their heads in the second time series. Based on the decreasing number of data sets, the number of head movements of the grazing cattle is converted to obtain the number of times the grazing cattle lower their heads in the second time series.

2. The method for monitoring the number of head movements of grazing cattle according to claim 1, characterized in that, When determining multiple increasing and multiple decreasing data points based on the comparison results of the two differences used to calculate each target product value with zero, the following steps are performed for each target product value: If both differences used to calculate the target product value are greater than zero, then the two differences are determined to be increasing data. If both differences used to calculate the target product value are less than zero, then the two differences are determined to be decreasing data.

3. The method for monitoring the number of head movements of grazing cattle according to claim 1, characterized in that, The process of converting the number of head movements of grazing cattle based on the increasing number of data sets to obtain the number of times the grazing cattle raise their heads in the second time series, and converting the number of head movements of grazing cattle based on the decreasing number of data sets to obtain the number of times the grazing cattle lower their heads in the second time series, includes: Multiply the number of incremental data sets by a preset multiple to obtain the number of times the grazing cattle raise their heads in the second time series; Multiply the decreasing number of data sets by a preset multiple to obtain the number of times the grazing cattle lowered their heads in the second time series.

4. The method for monitoring the number of head movements of grazing cattle according to claim 1, characterized in that, The second time series was determined based on the following method: The multiple initial accelerations are preprocessed to obtain multiple target accelerations; The multiple target accelerations are input into the head motion recognition model to obtain the head motion recognition result output by the head motion recognition model; The head action recognition results include head-raising and head-lowering recognition results when grazing cattle are performing grazing behavior; the head action recognition model is trained based on acceleration samples and the corresponding head behavior labels of the acceleration samples; the head behavior labels include head-raising behavior labels or head-lowering behavior labels. The consecutive moments corresponding to the head movement recognition results are determined as the second time series.

5. The method for monitoring the number of head movements of grazing cattle according to claim 4, characterized in that, The head motion recognition model was trained in the following manner: Obtain acceleration samples and frame image samples of the head of grazing cattle in the third time series; Based on the frame image samples, the acceleration samples at the same time are labeled with behavior to obtain head behavior labels; the head behavior labels include head-raising behavior labels and head-lowering behavior labels of the grazing cattle samples when they are grazing. The head action recognition model is obtained by training the model based on the acceleration samples and the head behavior labels.

6. A device for monitoring the number of head movements of grazing cattle, characterized in that, include: The acquisition module is used to acquire multiple initial accelerations of the head of grazing cattle in the first time series; The grazing cattle are equipped with accelerometers on their heads to collect the initial acceleration of their heads when walking and grazing. The initial acceleration includes acceleration in the x-axis, y-axis, and z-axis directions. The first time series is a time series in which the accelerometer continuously collects the initial acceleration. An extraction module is used to extract the acceleration components in the z-axis direction at the time corresponding to the second time sequence from the plurality of initial accelerations according to the second time sequence, so as to obtain multiple vertical accelerations; the second time sequence is the time sequence of the grazing behavior of the grazing cattle in the first time sequence; The data feature extraction module is used to extract data features based on the multiple vertical accelerations to obtain multiple increasing data and multiple decreasing data. The head movement count conversion module is used to convert the number of head movements of grazing cattle based on the multiple increasing data and the multiple decreasing data to obtain the number of head movements of grazing cattle; the number of head movements of grazing cattle includes the number of times the grazing cattle raise their heads and the number of times the grazing cattle lower their heads. The second time series was determined based on the following method: Based on the initial accelerations described above, the head movement recognition results of grazing cattle during grazing behavior are determined. The consecutive moments corresponding to the head movement recognition results are determined as the second time series; The data feature extraction based on the multiple vertical accelerations yields multiple increasing data points and multiple decreasing data points, including: The multiple vertical accelerations are calculated using first-order difference calculations to obtain multiple difference values. The multiple differences are multiplied sequentially by adjacent terms to obtain multiple product values; The product value that is greater than zero among the plurality of product values ​​is determined as the target product value; Based on the comparison results of the two differences used to calculate each of the target product values ​​with zero, multiple increasing data and multiple decreasing data are determined; The process of converting the number of head movements of grazing cattle based on the multiple increasing data and the multiple decreasing data to obtain the number of head movements of grazing cattle includes: Divide the increasing data that satisfy the condition of increasing in a continuous time sequence into a group, and determine the number of increasing data groups; Divide the decreasing data that satisfy the condition of decreasing in a continuous time sequence into a group, and determine the number of decreasing data groups; Based on the increasing number of data sets, the number of head movements of the grazing cattle is converted to obtain the number of times the grazing cattle raise their heads in the second time series. Based on the decreasing number of data sets, the number of head movements of the grazing cattle is converted to obtain the number of times the grazing cattle lower their heads in the second time series.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for monitoring the number of head movements of grazing cattle as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for monitoring the number of head movements of grazing cattle as described in any one of claims 1 to 5.

Images