Livestock biological asset supervision method and system based on intelligent ear tags

Abstract

The present invention relates to the field of data processing, and the present invention relates to a livestock biological asset supervision method and system based on smart ear tags. The health status data and environmental data of livestock biologicals are collected by smart ear tags, and the initial abnormality degree of the data point is obtained after preprocessing, and the initial abnormality degree is corrected by an adjustment factor to obtain the final abnormality degree of the data point, wherein the adjustment factor is related to the degree of chaos and neighborhood inclusion of the data point. Finally, the abnormal data point is identified by the final abnormality degree, and the staff is notified to complete the supervision. The present invention collects multi-dimensional health status data and environmental data through smart ear tags, and uses an adjustment factor based on the degree of chaos and neighborhood inclusion to correct the abnormality degree, thereby improving the accuracy and reliability of abnormality identification, realizing the intelligent supervision of livestock biological assets, and significantly improving management efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a livestock biological asset supervision method and system based on intelligent ear tags. Background Art

[0002] Smart ear tags are devices that integrate sensors, wireless communication, and data processing technologies, and are widely used in modern animal husbandry. Its core functions include real-time monitoring of livestock physiological data, such as body temperature, heart rate, respiratory rate, etc., and transmitting data to the management system through wireless communication to achieve real-time monitoring and management of livestock health status. In recent years, with the development of the Internet of Things (IoT) and big data technology, the functions of smart ear tags have been continuously enhanced, and their technologies in data collection, transmission, and analysis have also been continuously improved, greatly improving the management efficiency and production safety of animal husbandry. Although smart ear tags are increasingly used in animal husbandry, there are still some technical challenges in actual use.

[0003] For example, in the analysis of data collected by smart ear tags, due to the high dimensionality and large amount of data collected, traditional data analysis methods are difficult to effectively process and mine these data, especially in the detection of abnormal data. The calculation of the abnormal degree of boundary data points is prone to deviation, resulting in normal data points being misjudged as abnormal data points, thus affecting the accuracy and reliability of monitoring. Therefore, a technical solution that can improve the accuracy of abnormal data detection is urgently needed to optimize the monitoring and management of livestock biological health status.

[0004] In the prior art, commonly used anomaly detection algorithms such as the LOF (local outlier factor) algorithm, although they have certain advantages in processing high-dimensional data, still have defects in detecting anomalies at boundary data points. Specifically: when using the LOF algorithm for anomaly detection, the local reachable density calculation of normal boundary data points is relatively low due to the small number of surrounding data points, which can easily lead to an over-calculation of the degree of anomaly, and then misjudge normal data points as abnormal data points. Therefore, there is an urgent need for an improved anomaly detection method that can effectively adjust the degree of abnormality of boundary data points, improve the accuracy and reliability of anomaly detection, and thus achieve efficient and accurate monitoring of the health status of livestock organisms. Summary of the invention

[0005] In response to the problem of misjudgment of the above-mentioned boundary data points, in the first aspect, the present invention proposes a livestock biological asset supervision method based on smart ear tags, including: collecting health status data and environmental data of livestock biologicals at the same time to obtain data points, and obtaining the initial abnormality degree of any data point; correcting the initial abnormality degree based on an adjustment factor to obtain the final abnormality degree of the data point; the adjustment factor is positively correlated with the degree of confusion of the data point, and negatively correlated with the neighborhood inclusion degree of the data point; the degree of confusion is positively correlated with the local reachable distance of the data point, and the local reachable distance refers to the maximum distance between the data point and a set number of nearest neighboring points; the neighborhood inclusion degree refers to the number of times a data point is included in the neighborhood circle determined by the local reachable distance of the nearest neighboring point; identifying abnormal data points based on the final abnormality degree and issuing an abnormal notification; reminding managers to conduct health screening for livestock biologicals with abnormal conditions to complete livestock biological asset supervision.

[0006] The present invention collects the health status data and environmental data of livestock and animal husbandry based on intelligent ear tags, and realizes accurate livestock and animal husbandry asset supervision by preprocessing and detecting anomalies of the data. Compared with the prior art, the present invention significantly improves the accuracy and reliability of anomaly detection and reduces the false alarm rate by using adjustment factors to correct the initial degree of anomaly based on the degree of confusion of data points and the degree of neighborhood inclusion.

[0007] Furthermore, the adjustment factor is:

[0008]

[0009] in, Indicates The adjustment factor for each data point; represents the neighborhood inclusion threshold; Indicates The neighborhood inclusion degree of each data point; Indicates The degree of confusion of each data point; Represents the minimum function.

[0010] Furthermore, the neighborhood inclusion threshold is: obtain the neighborhood inclusion of all data points, where the maximum neighborhood inclusion and the minimum neighborhood inclusion are recorded as and ; respectively , and Perform K-Means clustering on the neighborhood inclusion of all data points as the initial center; The minimum value of the cluster obtained for the initial center is recorded as the neighborhood inclusion threshold .

[0011] The neighborhood inclusion threshold is obtained through the K-Means clustering method, which effectively classifies the neighborhood inclusion of all data points, ensuring the rationality and adaptability of the inclusion threshold. This method makes anomaly detection more accurate and avoids the limitations brought by a single threshold.

[0012] Furthermore, the degree of confusion is:

[0013]

[0014] in, Indicates The degree of confusion of each data point; Indicates the total number of dimensions; , Respectively represent Data point The maximum and minimum values ​​of the neighborhood data of the dimensional data; Indicates The local reachable distance of a data point; the neighborhood data refers to the nearest neighbor point of a set number of The value of the dimension.

[0015] By defining the calculation formula of the degree of chaos and combining the local reachable distance of the data point with the maximum and minimum values ​​of the neighborhood data, a method for accurately measuring the degree of chaos of the data point is provided. This method improves the accuracy of the degree of abnormality of the data point by quantifying the degree of chaos.

[0016] Furthermore, collecting health status data and environmental data of livestock organisms also includes: obtaining the organism's body temperature, heart rate, respiratory rate and weight data through smart ear tags; and using temperature and humidity sensors to collect environmental temperature and humidity data.

[0017] The present invention uses intelligent ear tags and temperature and humidity sensors to comprehensively collect health status data and environmental data of livestock and provide multi-dimensional data support. Compared with a monitoring system with a single data source, the present invention can provide more comprehensive and accurate monitoring information.

[0018] Furthermore, collecting the health status data of livestock organisms also includes collecting the data at a time interval of one week.

[0019] Furthermore, the collected data is preprocessed, specifically, the collected data is denoised and standardized, wherein the standardization method is as follows:

[0020]

[0021] in, Indicates data points in any dimension Standardized data; Indicates data points in any dimension Initial data of and Represents any dimension Mean and standard deviation of all initial data.

[0022] By denoising and standardizing the data, the quality and consistency of the data are improved, the interference of noise on anomaly detection is reduced, and the accuracy of the detection results is ensured.

[0023] Furthermore, obtaining the initial abnormality degree of any data point also includes: calculating the initial abnormality degree of the data point by using a local outlier factor algorithm.

[0024] Furthermore, identifying abnormal data points based on the final abnormality degree also includes: presetting an abnormality degree threshold; and recording data points whose final abnormality degree is greater than the abnormality degree threshold as abnormal data points.

[0025] In a second aspect, the present invention provides a livestock biological asset supervision system based on smart ear tags, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the livestock biological asset supervision method based on smart ear tags of the present invention is implemented.

[0026] The technical effects of the present invention are:

[0027] The present invention collects multi-dimensional health status data and environmental data through smart ear tags, and uses adjustment factors based on the degree of chaos and neighborhood inclusion to correct the degree of abnormality, thereby improving the accuracy and reliability of abnormality identification, realizing intelligent supervision of livestock biological assets, and significantly improving management efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0029] Figure 1 is a flowchart schematically showing a livestock biological asset supervision method based on smart ear tags according to an embodiment of the present invention;

[0030] Figure 2 is a grayscale image schematically showing a smart ear tag in an embodiment of the present invention;

[0031] Figure 3It is a block diagram schematically showing the structure of a livestock biological asset supervision system based on smart ear tags according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0033] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0034] Example of livestock biological asset supervision method based on smart ear tags:

[0035] like Figure 1 As shown, the livestock biological asset supervision method based on smart ear tags of the present invention includes:

[0036] S1. Collect smart ear tag data and environmental data of livestock and poultry; pre-process the collected data.

[0037] The body temperature, heart rate, respiratory rate and weight change data of livestock can directly reflect their health status; the temperature and humidity of the environment in which the organisms live often also have an impact on their health status. Based on these data, the health status of livestock can be evaluated and monitored in real time, so that managers can better understand the real-time status of livestock, optimize management strategies and improve management efficiency.

[0038] Specifically, the temperature sensor in the smart ear tag collects body temperature data, the heart rate sensor collects heart rate data, the pressure sensor collects breathing rate data, and the temperature and humidity sensor collects ambient temperature and humidity data. Figure 2 shown.

[0039] The present invention uses intelligent ear tags and temperature and humidity sensors to comprehensively collect health status data and environmental data of livestock and provide multi-dimensional data support. Compared with a monitoring system with a single data source, the present invention can provide more comprehensive and accurate monitoring information.

[0040] First, a data collection cycle is preset. In this embodiment, the data collection cycle takes an empirical value of one week. The implementer can adjust it according to the actual situation. That is, for a livestock creature, the parameters of its smart ear tag and environmental parameters are obtained once a week. The multi-dimensional data obtained at the same collection time constitutes a data point. For example, the first data point includes the body temperature data, heart rate data, respiratory rate data, and environmental temperature and humidity data collected in the first week. After the collection is completed, the collected data is denoised using data cleaning. The data cleaning is a well-known technology and will not be described here.

[0041] It should be noted that for the collected multi-dimensional data, due to the different sizes of the data, the impact on the Euclidean distance in the process of LOF algorithm anomaly detection is also different. Therefore, the data can be preprocessed first to unify the data dimensions of different dimensions.

[0042] Specifically, for any dimension of data, obtain its mean value and standard deviation, and convert any dimension The average value of the data is recorded as , any dimension The standard deviation of the data is recorded as , standardize the single data in the dimension according to the mean and standard deviation of all data in the dimension:

[0043]

[0044] in, Indicates data points in any dimension Standardized data; Indicates data points in any dimension Initial data of and Represents any dimension Mean and standard deviation of all initial data.

[0045] By standardizing the data, we can ensure the quality and consistency of the data, reduce the interference of noise on anomaly detection, and ensure that the influence of various data on the Euclidean distance is the same when calculating the Euclidean distance in the subsequent LOF algorithm.

[0046] S2. Obtain the initial abnormality degree of the data point.

[0047] In the present invention, the LOF algorithm is used to detect abnormalities on the standardized data points. The initial abnormality degree of the data point, that is, the local outlier factor, is recorded as ; If the value is greater than 1, it means that the density of the data point is lower than the density of its neighborhood, indicating that the point is more likely to be an outlier.

[0048] S3, obtaining the neighborhood inclusion degree of the data point; obtaining the degree of disorder of the data point; and obtaining the adjustment factor of the data point based on the neighborhood inclusion degree and disorder degree of the data point.

[0049] All data points collected from a livestock organism constitute a data set. For normal data points, even if they are at the boundary of the data set, they will still be within the neighborhood circle determined by the local reachable distance of other data points as the radius, where the local reachable distance refers to the maximum distance between the data point and a set number of nearest neighboring points. In this embodiment, the set number is an empirical value of 5, and the implementer can select it according to the actual situation. However, since abnormal data points are local isolated points, they will not or rarely be within the neighborhood circle determined by the local reachable distance of other data points as the radius.

[0050] In addition, since the boundary data points are at the boundary of the data set, the distribution of the data points around them is uneven, which leads to the miscalculation of the abnormality of the boundary data points. In summary, the adjustment factor is obtained according to the number of times the data point is included in the neighborhood circle determined by the local reachable distance of other data points as the radius and the degree of chaos of other surrounding data points to adjust the abnormality of the data point.

[0051] For the neighborhood inclusion of a data point, we first obtain the other data points contained in the local reachable distance of each data point, and then count the number of times the data point appears in the neighborhood circle determined by the local reachable distance of other data points as the radius, and record the first The neighborhood inclusion degree of a data point is It should be noted that for different data points, in the process of abnormality determination based on the neighborhood inclusion threshold, due to the different densities of data points in different regions, the neighborhood inclusion threshold will also be different, so it is necessary to set the neighborhood inclusion threshold based on the neighborhood inclusion level of all data points.

[0052] First, the neighborhood inclusion of all data points is obtained, where the maximum neighborhood inclusion and the minimum neighborhood inclusion are recorded as and ; and then respectively , and Perform K-Means clustering on the neighborhood inclusion of all data points as the initial center; finally, The minimum value of the cluster obtained for the initial center is recorded as the neighborhood inclusion threshold .

[0053] The neighborhood inclusion threshold is obtained through the K-Means clustering method, which effectively classifies the neighborhood inclusion of all data points, ensuring the rationality and adaptability of the inclusion threshold. This method makes anomaly detection more accurate and avoids the limitations brought by a single threshold.

[0054] Regarding the degree of chaos of data points, it should be noted that: for data points at the boundary of the data set among all data points, the distribution of other data points around them is often uneven, and it is precisely because of the uneven distribution of data points around them that the calculation of the abnormality degree of boundary data points is not accurate enough. Therefore, whether a data point is a boundary data point can be judged based on the degree of chaos of the distribution of other data points around the data point.

[0055] Specifically, for the data points, get their The maximum and minimum values ​​of the neighborhood data of the dimensional data are recorded as , ; Get the first The local reachable distance of a data point is recorded as , then The degree of confusion of a data point can be expressed as:

[0056]

[0057] in, Indicates The degree of confusion of each data point; Indicates the total number of dimensions; , Respectively represent Data point The maximum and minimum values ​​of the neighborhood data of the dimensional data; Indicates The local reachable distance of a data point; the neighborhood data refers to the nearest neighbor point of a set number of For neighborhood data, the following example illustrates: data points, and The data of dimension is recorded as ; Then obtain the remaining data points within the neighborhood circle determined by the local reachable distance of the data point as the radius, and put the first The data point The data of dimension is recorded as ; Then for the Data point The neighborhood data of the dimension data is: ,in For the first The local reachable distance of a data point is the total number of other data points within the neighborhood circle determined by the radius.

[0058] It should be noted that the Partially indicates Data point The ratio between the neighborhood data range and the local reachability distance of the dimensional data; for the data point, if the distribution of its neighboring data points is uniform, then The data of the dimension should also be evenly distributed, and because the The neighboring data points of a data point are all data points within the neighborhood circle determined by the local reachable distance as the radius. Therefore, the range of the data in any dimension will not exceed twice the local reachable distance. Data point The neighborhood data range of the data of the dimension is compared with the local reachable distance. The larger the ratio is, the more uniform the distribution of the data of the dimension is. In the formula, 1 is subtracted from the ratio, that is, The larger the The data point The more uneven the distribution of the neighborhood data of the dimensional data is, the more the degree of disorder of the neighborhood data of all dimensional data can be obtained. The degree of chaos of all data points in the neighborhood of a data point.

[0059] By defining the calculation formula of the degree of chaos and combining the local reachable distance of the data point with the maximum and minimum values ​​of the neighborhood data, a method for accurately measuring the degree of chaos of the data point is provided. This method improves the accuracy of the degree of abnormality of the data point by quantifying the degree of chaos.

[0060] Obtain the neighborhood inclusion and disorder degree of all data points. Specifically, obtain the adjustment factor according to the neighborhood inclusion and disorder degree of the data points, and the calculation formula is:

[0061]

[0062] in, Indicates The adjustment factor for each data point; represents the neighborhood inclusion threshold; Indicates The neighborhood inclusion degree of each data point; Indicates The degree of confusion of each data point; Represents the minimum function. So far, the adjustment factors of all data points are obtained.

[0063] S4. Correct the initial abnormality of the data point based on the adjustment factor of the data point.

[0064] For a data point, there is a local outlier factor , and its adjustment factor is calculated in step S3 , then the final abnormality of the data point, that is, the corrected local outlier factor have:

[0065]

[0066] For the final abnormality degree: the present invention first collects the health status and environmental data of livestock and performs standardization processing; then calculates the initial abnormality degree through the local outlier factor (LOF) algorithm; then calculates the degree of chaos and neighborhood inclusion, and calculates the adjustment factor according to the formula; finally, uses the adjustment factor to correct the initial abnormality degree to obtain the final abnormality degree.

[0067] This value combines health status and environmental data to provide comprehensive monitoring. It also improves detection accuracy by adjusting the degree of chaos and neighborhood inclusion. It also uses K-Means clustering to dynamically determine the neighborhood inclusion threshold to improve adaptability.

[0068] S5. Make an abnormal judgment on the smart ear tag data based on the final abnormality degree of the data point; complete the supervision of livestock biological assets.

[0069] First, a final abnormality threshold is preset, and the empirical value in this embodiment is , implementers can set the final abnormality threshold according to the actual implementation situation; for the data set collected from a livestock creature, if the final abnormality of a data point is greater than the final abnormality threshold, an abnormality notification will be issued to remind managers to pay special attention to the livestock creature with abnormal status, and further health screening and status assessment can be carried out to complete the livestock creature asset supervision.

[0070] Livestock biological asset supervision system based on smart ear tags:

[0071] On the other hand, the present invention also provides a livestock biological asset supervision system based on smart ear tags. Figure 3 As shown, the livestock biological asset supervision system based on smart ear tags includes a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, the livestock biological asset supervision method based on smart ear tags according to the first aspect of the present invention is implemented.

[0072] The livestock biological asset monitoring system based on smart ear tags also includes other components familiar to those skilled in the art, such as communication interfaces, and their settings and functions are known in the art, so they will not be described in detail here.

[0073] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0074] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0075] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A livestock biological asset supervision method based on smart ear tags, characterized in that: Methods include: Collect the health status data and environmental data of livestock organisms at the same time to obtain data points, and obtain the initial abnormality degree of any data point; Correcting the initial abnormality degree based on the adjustment factor to obtain the final abnormality degree of the data point; Adjustment Factor: ； in, Indicates The adjustment factor for each data point; represents the neighborhood inclusion threshold; Indicates The neighborhood inclusion degree of each data point; Indicates The degree of confusion of each data point; represents the minimum function; The adjustment factor is positively correlated with the degree of chaos of the data point and negatively correlated with the neighborhood inclusion of the data point; The degree of confusion is: ； in, Indicates the total number of dimensions; , Respectively represent Data point The maximum and minimum values ​​of the neighborhood data of the dimensional data; Indicates The local reachable distance of each data point; Neighborhood data refers to the nearest neighbor points among the set number of The value of the dimension; The degree of chaos is positively correlated with the local reachable distance of the data point. The local reachable distance refers to the maximum distance between the data point and the set number of nearest neighboring points. The neighborhood inclusion degree refers to the number of times a data point is included in the neighborhood circle determined by the local reachable distance of the nearest neighboring point. The neighborhood inclusion threshold is: Get the neighborhood inclusion of all data points, where the maximum neighborhood inclusion and the minimum neighborhood inclusion are recorded as and ; Respectively , and Perform K-Means clustering on the neighborhood inclusion of all data points for the initial center; Will be The minimum value of the cluster obtained for the initial center is recorded as the neighborhood inclusion threshold ; Identify abnormal data points based on the final abnormality level and issue abnormality notifications; remind managers to conduct health screening of livestock and animal species with abnormal conditions and complete livestock and animal asset supervision; Collect data on the health status of livestock and the environment, including: Obtain the body temperature, heart rate, respiratory rate and weight data of the organism through the smart ear tag; Use temperature and humidity sensors to collect environmental temperature and humidity data.

2. The livestock biological asset supervision method based on smart ear tags according to claim 1 is characterized in that: The collection of livestock health status data also includes a collection interval of one week.

3. The livestock biological asset supervision method based on smart ear tags according to claim 1 is characterized in that: It also includes preprocessing of the collected data, specifically: The collected data is denoised and standardized, and the standardization method is as follows: ； in, Indicates data points in any dimension Standardized data; Indicates data points in any dimension Initial data of and Represents any dimension Mean and standard deviation of all initial data.

4. The livestock biological asset supervision method based on smart ear tags according to claim 1 is characterized in that: Get the initial abnormality level of any data point, including: The initial degree of abnormality of a data point is calculated using the local outlier factor algorithm.

5. The livestock biological asset supervision method based on smart ear tags according to claim 1 is characterized in that: Identifying abnormal data points based on the final abnormality level includes: Preset abnormality threshold; The data points whose final abnormality degree is greater than the abnormality degree threshold are recorded as abnormal data points.

6. Livestock biological asset supervision system based on smart ear tags, characterized by: The method comprises a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the livestock biological asset supervision method based on smart ear tags as described in any one of claims 1 to 5 is implemented.

Images