Method and system for assessing animal welfare level of a pasture and storage medium

Abstract

The application provides a method and system for evaluating animal welfare level of a farm and a storage medium. The method comprises: collecting to-be-processed data, wherein the to-be-processed data comprises environmental feature data, feeding feature data and behavior feature data of an animal to be evaluated; inputting the to-be-processed data into a back propagation neural network model for processing to obtain to-be-processed level data; and inputting the to-be-processed level data into a fuzzy logic controller model for processing to obtain animal welfare level data. According to the embodiment of the application, the environmental feature data, the feeding feature data and the behavior feature data of the animal to be evaluated are organically classified and fused, so that the redundancy and mutual exclusivity among the data can be eliminated, the difficulty of weight distribution of the data can be effectively avoided, the various types of data can be mutually complementary and cooperative, and the reliability of the animal welfare evaluation and monitoring data can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of animal monitoring technology, and in particular to a method, system and storage medium for assessing the animal welfare level of a ranch. Background Technology

[0002] With the development of animal welfare research in pastures, animal welfare studies have evolved from focusing on single behaviors to multi-behavioral studies, and now primarily focus on structured data such as feed intake, milk production, and metabolic rate. Currently, a set of internationally recognized animal welfare evaluation criteria, the "5F" principle, has been established: physiological welfare, environmental welfare, health welfare, behavioral welfare, and psychological welfare. Correspondingly, a multi-source data fusion approach based on multiple pasture monitoring sensors has been proposed, utilizing the "5F" evaluation dimensions. In traditional techniques, multiple monitoring data can be obtained through contact instruments such as triaxial accelerometers and magnetometer collar systems to monitor animal behavior. Monitoring pastures using multiple sensors yields a large amount of heterogeneous data, which exhibits certain correlations, randomness, and ambiguity. Furthermore, due to the varying accuracies of different types of sensors, the different physical meanings of the information obtained, and the correlation between monitoring data from different sensors, the allocation of weights for each monitoring data point during data fusion becomes a challenge, leading to a lack of reasonable interpretation of the fused results. Summary of the Invention

[0003] This application is made in view of at least one of the aforementioned technical problems existing in the prior art. According to one aspect of this application, a method for assessing the animal welfare level of a ranch is provided, the method comprising:

[0004] Collect data to be processed, wherein the data to be processed includes environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animal to be evaluated;

[0005] The data to be processed is input into a backpropagation neural network model for processing to obtain the level data to be processed.

[0006] The data to be processed is input into a fuzzy logic controller model for processing to obtain animal welfare level data.

[0007] In some embodiments, after acquiring the data to be processed, the method further includes:

[0008] When the data to be processed is the behavioral feature data, the behavioral feature data is preprocessed to obtain the preprocessed behavioral feature data.

[0009] The preprocessed behavioral feature data is input into a preset perception model to obtain the structured behavioral feature data corresponding to the behavioral feature data.

[0010] In some embodiments, the behavioral feature data is preprocessed, including:

[0011] The behavioral feature data is processed using a preset algorithm to obtain a preset number of image frames;

[0012] The step of inputting the preprocessed behavioral feature data into a preset perception model includes:

[0013] The preset number of image frames are input into the preset perception model to obtain the behavioral feature structured data corresponding to the behavioral feature data.

[0014] In some embodiments, the data to be processed is input into a backpropagation neural network model for processing to obtain the level data to be processed, including:

[0015] The environmental feature data, the feeding feature data, and the behavioral feature structured data are input into the backpropagation neural network model for processing to obtain environmental level data, feeding level data, and behavioral feature level data.

[0016] In some embodiments, the to-be-processed level data is input into a fuzzy logic controller model for processing to obtain animal welfare level data, including:

[0017] The fuzzy logic controller model is used to process the level data to be processed, and the processed level data is obtained.

[0018] Animal welfare level data is obtained by fusing the processed level data based on preset fuzzy rules.

[0019] In some embodiments, after obtaining the animal welfare rating, the method further includes:

[0020] The animal welfare level data has been revised.

[0021] The correction of the animal welfare level data includes:

[0022] Collect new behavioral characteristic data;

[0023] The environmental feature data, the feeding feature data, and the new behavioral feature data are used as data to be processed. The process continues by inputting the data to be processed into the backpropagation neural network model to obtain new data to be processed, and inputting the new data to be processed into the fuzzy logic controller model to obtain a new animal welfare level.

[0024] In some embodiments, the data to be collected and processed includes at least one of the following:

[0025] The environmental characteristic data is collected by environmental sensors installed in the pasture and stored in an environmental characteristic database.

[0026] Collect feeding index data and store the feeding index data in a feeding characteristic database;

[0027] Video data of the animal to be evaluated is collected by monitoring equipment, and behavioral feature data is extracted from the video data and stored in a behavioral feature database.

[0028] In some embodiments, the behavioral characteristic data includes coat color characteristic data and / or ear characteristic data; the animal to be evaluated includes dairy cows.

[0029] Another embodiment of this application provides an animal welfare level assessment system for ranches, the system comprising a basic data layer, a network layer, a feature data layer, an algorithm model layer, and a decision application layer:

[0030] The basic data layer is used to collect data to be processed, which includes environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animal to be evaluated.

[0031] The network layer is used to transmit the data to be processed to the feature data layer for storage;

[0032] The feature data layer is used to store the data to be processed;

[0033] The algorithm model layer is used to process the data to be processed using a backpropagation neural network model to obtain the data to be processed level, and to input the data to be processed level into a fuzzy logic controller model for processing to obtain animal welfare level data.

[0034] The decision application layer is used to output the animal welfare level data.

[0035] In another aspect, this application provides a storage medium storing a computer program that, when executed by a processor, causes the processor to perform the animal welfare level assessment method as described above.

[0036] The method, system, and storage medium for assessing animal welfare levels in ranches according to embodiments of this application organically classify and integrate environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animals to be assessed, forming a convergence of similar data. This helps improve the management and utilization rate of ranch data and lays the foundation for the accuracy of animal welfare detection. During data fusion, a backpropagation neural network is used to classify environmental characteristic data, feeding characteristic data, and behavioral characteristic data into different levels for comprehensive analysis and rational utilization, eliminating redundancy and mutual exclusion between data. This effectively avoids the difficulty of weighting data, allowing various types of data to complement and cooperate with each other, thereby effectively improving the reliability of animal welfare assessment and monitoring data. Attached Figure Description

[0037] Figure 1 A schematic flowchart illustrating a method for assessing the animal welfare level of a ranch according to an embodiment of this application is shown.

[0038] Figure 2 A hierarchical classification correspondence table of cow behavioral characteristic data according to an embodiment of this application is shown;

[0039] Figure 3 A schematic block diagram illustrating a data processing procedure based on a fuzzy logic controller according to an embodiment of this application is shown.

[0040] Figure 4(a) shows a schematic diagram of the preset membership functions of each input and output variable in the fuzzy logic controller model according to an embodiment of the present application;

[0041] Figure 4(b) shows a schematic diagram of the preset membership functions of each input and output variable in the fuzzy logic controller model according to an embodiment of the present application;

[0042] Figure 5 A table showing the correspondence between environmental level data, feeding level data, and behavioral characteristic level data and animal welfare level data according to embodiments of this application is provided.

[0043] Figure 6 A schematic diagram of an animal welfare assessment model according to an embodiment of this application is shown;

[0044] Figure 7 A schematic block diagram of an animal welfare rating system for a ranch according to an embodiment of this application is shown.

[0045] Figure 8 A schematic block diagram illustrating the basic data layer principle according to an embodiment of this application is shown;

[0046] Figure 9 A schematic block diagram of an animal welfare rating assessment system according to another embodiment of this application is shown;

[0047] Figure 10 A schematic block diagram of an electronic device according to an embodiment of this application is shown. Detailed Implementation

[0048] To enable those skilled in the art to better understand the technical solutions of the embodiments of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] With the continuous improvement of people's living standards, dietary habits have changed, leading to an increasing demand for meats such as beef and mutton, as well as higher requirements for meat and milk quality. This has propelled my country's dairy farming from extensive, free-range farming to a new stage of large-scale, intensive, and standardized farming, changing industry demands. However, intensive and efficient production activities under human control, along with predatory "profit maximization" management practices, have resulted in a decline in the average welfare of dairy cows. Good animal welfare is essential to comprehensively improve the production efficiency and quality of fresh milk on farms, ensure product safety, enhance product competitiveness, and meet consumer demand. The current pressing issue is how to satisfy consumer demand while ensuring product quality. Therefore, smart farms, combining artificial intelligence, computer vision, and big data, are an inevitable development trend.

[0050] Currently, most scholars monitor dairy cow behavior using contact instruments such as triaxial accelerometers and magnetometer collar systems, and have proposed multi-source data fusion based on multiple monitoring pasture sensors using the "5F" evaluation dimensions. While using multiple sensors to monitor pastures can yield a large amount of heterogeneous data, this data exhibits certain correlations, randomness, and ambiguity. Assigning weights to each monitoring data point during data fusion becomes a challenging problem.

[0051] Based on at least one of the aforementioned technical problems, this application provides a method for assessing animal welfare levels in ranches. The method includes: collecting data to be processed, wherein the data to be processed includes environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animals to be assessed; inputting the data to be processed into a backpropagation neural network model for processing to obtain unprocessed level data; and inputting the unprocessed level data into a fuzzy logic controller model for processing to obtain animal welfare level data. This application's embodiments organically classify and integrate environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animals to be assessed, forming a convergence of similar data. This helps improve the management and utilization rate of ranch data and lays the foundation for the accuracy of animal welfare detection. During data fusion, a backpropagation neural network is used to classify environmental characteristic data, feeding characteristic data, and behavioral characteristic data into different levels for comprehensive analysis and rational utilization, eliminating redundancy and mutual exclusion between data. This effectively avoids the difficulty of weighting data, allowing various types of data to complement and cooperate with each other, thereby effectively improving the reliability of animal welfare assessment and monitoring data.

[0052] The animal welfare level assessment method proposed in this invention employs a multi-source heterogeneous data fusion model based on pasture monitoring data. This model combines a backpropagation (BP) neural network and a fuzzy logic inference model. Three parallel BP neural networks are used to obtain classification levels for environment, feeding, and bovine physical behavior, which are then input into the fuzzy logic inference model. Expert experience is then used to output the welfare level, effectively avoiding weight allocation issues and ensuring the most effective monitoring of dairy cow welfare on pastures. Embedding this multi-source heterogeneous data fusion model into a welfare monitoring system enables automated monitoring of animal welfare in smart pastures, reducing reliance on human labor and meeting the system's requirements for intelligence, automation, and controllability.

[0053] Figure 1 A schematic flowchart illustrating a method for assessing animal welfare levels in a ranch according to an embodiment of this application is shown; Figure 1 As shown, the animal welfare level assessment method 100 for ranches according to an embodiment of this application may include the following steps S101, S102 and S103:

[0054] In step S101, the data to be processed is collected.

[0055] The data to be processed includes environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animals to be evaluated.

[0056] In one embodiment of this application, the behavioral characteristic data includes coat color characteristic data and / or ear characteristic data of the animal to be evaluated.

[0057] Since most ranches currently raise a large number of dairy cows, the animals to be evaluated in this application embodiment may include dairy cows. This application will take dairy cows as an example to collect behavioral characteristic data of dairy cows, namely, coat color characteristic data and / or ear characteristic data.

[0058] In one embodiment of this application, the collected data to be processed includes at least one of the following:

[0059] A1, the environmental feature data is collected by environmental sensors installed in the pasture and the environmental feature data is stored in the environmental feature database;

[0060] A2, collect feeding index data and store the feeding index data in the feeding feature database;

[0061] A3. Video data of the animal to be evaluated is collected through monitoring equipment, and behavioral feature data is extracted from the video data and stored in a behavioral feature database.

[0062] In one example, environmental sensors may include temperature and humidity sensors, carbon dioxide concentration sensors, and light intensity sensors. Accordingly, environmental characteristic data may include the temperature and humidity, carbon dioxide concentration, and light intensity of the pasture environment, as well as the disinfection time of the cattle shed, the disinfection time of the cattle, and the disinfection time of the feed troughs and waterers at each stage.

[0063] In one example, feeding management data may include the nutritional composition of the feed at each stage of feeding, such as feed dry weight, starch, crude protein, etc.

[0064] Generally, based on the internationally recognized "5F" principle, animal welfare is divided into five dimensions: physiological welfare, environmental welfare, psychological welfare, behavioral welfare, and emotional welfare. Relevant psychological, environmental, physiological, behavioral, and emotional data can be collected using multi-source sensors on the ranch as the basis for evaluating these five welfare dimensions. In this embodiment, environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animal to be evaluated are collected, and this data is organically classified and fused using a multi-source data aggregation model, serving as the basis for evaluating the five animal welfare dimensions. This aggregation of similar data helps improve the management and utilization rate of ranch data and lays the foundation for animal welfare assessment.

[0065] In one embodiment of this application, after collecting the data to be processed, the method further includes:

[0066] B1, when the data to be processed is the behavioral feature data, the behavioral feature data is preprocessed to obtain the preprocessed behavioral feature data;

[0067] B2, input the preprocessed behavioral feature data into a preset perception model to obtain the behavioral feature structured data corresponding to the behavioral feature data.

[0068] In one example, preprocessing the behavioral feature data includes: processing the behavioral feature data using a preset algorithm to obtain a preset number of image frames.

[0069] In one example, inputting the preprocessed behavioral feature data into a preset perception model includes: inputting the preset number of image frames into the preset perception model to obtain the behavioral feature structured data corresponding to the behavioral feature data.

[0070] In this embodiment, behavioral feature data is collected by capturing video data of dairy cows using a camera and analyzing the video data to obtain coat color and / or ear feature data. Since video data is unstructured, it needs to be converted into structured behavioral feature data before further processing. In a specific example, the behavioral feature data can be processed using a k-means framing algorithm to obtain n 128*128 image frames. These n 128*128 image frames include ear image frames and coat color image frames. The ear image frames are input into a pre-built ear behavior perception model to classify the ear feature data. Simultaneously, the coat color image frames are input into a pre-built coat color classification perception model to classify the coat color feature data. This transforms the ear feature data and coat color feature data into structured behavioral feature data. Figure 2 The table shown is a classification table of behavioral characteristics data of dairy cows.

[0071] In step S102, the data to be processed is input into a backpropagation neural network model for processing to obtain the level data to be processed.

[0072] In one embodiment of this application, the data to be processed is input into a backpropagation neural network model for processing to obtain the level data to be processed, including: inputting the environmental feature data, the feeding feature data, and the behavioral feature structured data into the backpropagation neural network model for processing to obtain environmental level data, feeding level data, and behavioral feature level data.

[0073] In this embodiment, environmental characteristic data, feeding characteristic data, and behavioral characteristic data are processed using three corresponding BP neural network models. By employing three parallel BP neural network models to comprehensively analyze and rationally utilize the environmental characteristic data, feeding characteristic data, and behavioral characteristic data, redundancy and mutual exclusion between data are eliminated. This also effectively avoids the technical difficulty of weighting data in traditional technologies. This embodiment enables various types of data to complement and cooperate with each other, thereby effectively improving the reliability of animal welfare monitoring data and increasing its utilization rate.

[0074] The embodiments of this application use BP neural networks to train environmental characteristic data, feeding characteristic data, and behavioral characteristic data to determine their respective levels. Through the technical solution of this application, the management efficiency of the ranch is improved, the degree of integrated ranch management is enhanced, and the management of ranch data is facilitated.

[0075] In step S103, the data to be processed is input into the fuzzy logic controller model for processing to obtain animal welfare level data.

[0076] In one embodiment of this application, the to-be-processed level data is input into a fuzzy logic controller model for processing to obtain animal welfare level data, including:

[0077] C1, the preset membership function in the fuzzy logic controller model is used to process the level data to be processed to obtain the processed level data;

[0078] C2, based on preset fuzzy rules, fuses the processed level data to obtain animal welfare level data.

[0079] like Figure 3 The diagram shown is a schematic block diagram of the data processing process based on a fuzzy logic controller. Figure 3 The fuzzy controller model receives environmental level data (as shown in the figure), feeding level data (as shown in the figure), and behavioral characteristic level data (as shown in the figure), and then performs fuzzification, fuzzy inference, and declarative processing on the environmental level data, feeding level data, and behavioral characteristic level data. The animal welfare level data is then output through the controlled object module. The fuzzification, fuzzy inference, and declarative processing can share a knowledge base, and the output data can be fuzzified, fuzzy inference, and declarative processing again.

[0080] The preset membership functions of each input and output variable in the fuzzy logic controller model in this embodiment are shown in Figures 4(a) and 4(b). This paper performs fuzzy partitioning on the input and output variables; the number of fuzzy partitions determines the degree of fuzzy logic refinement. The specific fuzzy subset partitioning is as follows:

[0081] A fuzzy subset of the environment rating data: {low, medium, high}; the universe of discourse is set to [0,10].

[0082] A fuzzy subset of the feeding level data (Fodder): {low, medium, high}; the universe of discourse is set to [0,10];

[0083] A fuzzy subset of behavioral characteristic level data: {low, medium, high}; the universe of discourse is set to [0,10];

[0084] A fuzzy subset of animal welfare rating data: {low, medium, high}; the universe of discourse is set to [0, 25].

[0085] In one embodiment of this application, fuzzy rules derived from expert experience can also be incorporated.

[0086] like Figure 5 The table shown is a correspondence table between environmental level data, feeding level data, behavioral characteristic level data, and animal welfare level data.

[0087] In this embodiment, the three levels obtained are aggregated into a fuzzy controller to generate animal welfare level data for the dairy farm. Monitoring the welfare status of dairy cows in this way and making adaptive adjustments based on their welfare status can not only improve the health of dairy cows and extend their lifespan, but also significantly increase milk production and improve the economic benefits of dairy farming.

[0088] In one embodiment of this application, after obtaining the animal welfare level, the method further includes: correcting the animal welfare level data.

[0089] In one example, the correction of the animal welfare level data includes:

[0090] D1, Collect new behavioral characteristic data;

[0091] D2, taking the environmental feature data, the feeding feature data, and the new behavioral feature data as data to be processed, and continuing to execute the steps of inputting the data to be processed into the backpropagation neural network model for processing to obtain new data to be processed, and inputting the new data to be processed into the fuzzy logic controller model for processing to obtain a new animal welfare level.

[0092] like Figure 6 The diagram illustrates an animal welfare assessment model according to an embodiment of this application. This embodiment monitors three aspects: environment, feeding, and behavior, and obtains feedback. Environmental and feeding level data obtained from the monitoring are used for direct evaluation, while behavioral level data is used for indirect evaluation. Environmental levels can be categorized as excellent, average, and poor. Feeding levels can also be categorized as excellent, average, and poor. Behavioral levels can also be categorized as excellent, average, and poor. Environmental, feeding, and behavioral level data can be directly integrated as influencing factors to obtain the animal welfare level of the ranch, for example, excellent, average, and poor. Since animal behavior changes continuously during growth due to adjustments in environment and feeding, the animal's behavioral level undergoes phased changes. As animal behavior changes, new behavioral level data can be obtained at any time to adjust and correct the animal welfare level data.

[0093] Figure 7 This diagram illustrates a schematic block diagram of an animal welfare rating system for a ranch according to an embodiment of this application; as shown. Figure 7 As shown, the animal welfare level assessment system 700 according to an embodiment of this application may include a basic data layer 701, a network layer 702, a feature data layer 703, an algorithm model layer 704, and a decision application layer 705.

[0094] The basic data layer 701 is used to collect data to be processed, which includes environmental characteristic data, feeding characteristic data, and behavioral characteristic data of the animal to be evaluated.

[0095] Combination Figure 7 The basic data layer 701 consists of pasture environmental monitoring data, pasture feeding management data, and pasture cattle behavior data. Pasture environmental monitoring data includes temperature and humidity sensors, carbon dioxide (CO2) concentration sensors, light intensity sensors, and the disinfection duration of the cowshed, cattle, and feed troughs / waterers at each stage. The feed given to dairy cows at each stage includes concentrates, minerals, and vitamins. These feeds can be analyzed using a nutrient composition analyzer to determine feeding management data, which includes nutritional information for each stage of feeding, such as feed dry weight, starch, acid detergent fiber, neutral detergent fiber, crude protein, and ash content. Pasture cattle behavior data includes monitoring videos of the pasture at each stage, specifically divided into ear videos and coat color videos. All of this data can be uploaded to the network layer via the data transmission module.

[0096] like Figure 8The diagram illustrates a schematic block diagram of the basic data layer principle according to an embodiment of this application. The perception layer for collecting data to be processed in this embodiment can be composed of multiple sensor nodes (first sensor node, second sensor node, ..., nth sensor node) and a gateway node. Environmental data collected by each sensor node is stored in an environmental data queue. Data preprocessing (e.g., data cleaning) can be performed on the environmental data, followed by similar data fusion processing. Feeding data collected by each sensor node is stored in a feeding data queue, and data preprocessing (e.g., data cleaning) can be performed on the feeding data, followed by similar data fusion processing. For behavioral feature data, since it is video data collected by video surveillance equipment, frame segmentation and classification are required to obtain fur color feature data and ear feature data from the behavioral feature data. Then, similar data fusion processing is performed on the fur color feature data and ear feature data.

[0097] The network layer 702 is used to transmit the data to be processed to the feature data layer for storage.

[0098] The network layer 702 in this embodiment of the application realizes data transmission through wireless or wired methods such as WIFI, General Packet Radio Service (GPRS), and Long Range Radio (LoRa). In a specific example, the sensing layer used to collect data to be processed can be composed of multiple sensor nodes (first sensor node, second sensor node, ... nth sensor node) and a gateway node. It mainly collects information such as pasture temperature, humidity, light intensity, and feeding management data. The nodes form a wireless sensor network (WSN) through wireless communication in the 433MHz band. The gateway node is directly connected to the wide area network through narrowband Internet of Things (NB-IoT). The network layer consists of an NB-IoT base station, an NB-IoT core network, and an NB-IoT platform.

[0099] The feature data layer 703 is used to store the data to be processed.

[0100] The feature data layer 703 in this embodiment mainly includes an environmental feature database, a feeding feature database, and a cattle body feature database. The environmental feature database mainly stores environmental data measured by environmental sensors; the feeding feature database mainly stores pasture feeding management data at various stages; and the cattle body feature database mainly stores cattle ear videos and coat color videos at each stage of the pasture.

[0101] The algorithm model layer 704 is used to process the data to be processed using a backpropagation neural network model to obtain the level data to be processed, and to input the level data to be processed into a fuzzy logic controller model for processing to obtain animal welfare level data.

[0102] The algorithm model layer 704 in this embodiment mainly includes a BP neural network model, an ear behavior perception model, and a coat color grade classification perception model. This is divided into direct and indirect evaluation methods. External influencing factors (such as environmental grade and feeding grade) are used as direct evaluation indicators for multi-source data fusion; the behavioral characteristics of the cattle themselves (coat color characteristic data and ear characteristic data) are used as indirect evaluation indicators to assist in the assessment of animal welfare grades.

[0103] The decision application layer 705 is used to output the animal welfare level data. The safety and efficacy of beef or dairy products can be determined from the animal welfare level data.

[0104] like Figure 9The diagram shows a schematic block diagram of an animal welfare level assessment system for a ranch, according to another embodiment of this application. In this embodiment, the indicators used to assess animal welfare levels can be divided into two categories based on the evaluation method: direct evaluation indicators and indirect evaluation indicators. Direct evaluation indicators include environmental indicators and feeding indicators, while indirect evaluation indicators include bovine physical characteristics. Environmental indicators include temperature and humidity index, carbon dioxide concentration, light intensity, duration of cattle shed disinfection, duration of cattle body disinfection, and duration of feed trough and waterer disinfection. These indicators are input into a BP neural network model, which outputs environmental level data. Feeding indicators include feed dry weight, starch, acid detergent fiber, neutral detergent fiber, crude protein, and ash content. These indicators are input into a BP neural network model, which outputs feeding level data. Bovine physical characteristics include ear behavior characteristics and coat color level characteristics. These two indicators are input into a BP neural network model, which then outputs bovine physical condition level data (i.e., the behavior level data mentioned above). By inputting the environmental level data, feeding level data, and bovine physical condition level data into the analysis and decision-making module, and through fuzzy logic reasoning, the final animal welfare evaluation level can be obtained. As dairy cows grow, their appearance changes, and their physical characteristics can be updated at any time to obtain new physical characteristics and a new bovine condition level. The animal welfare evaluation level is then adjusted based on this new bovine condition level. This application's embodiment uses technologies such as the Internet of Things, big data, and artificial intelligence to collect and aggregate multi-dimensional and multi-level pasture data. By classifying and normalizing multi-source heterogeneous monitoring data, it inputs them into three parallel BP neural network models to aggregate environmental characteristics, feeding characteristics, and bovine behavioral characteristics, obtaining environmental level data, feeding level data, and bovine condition level data, respectively. These data are then used as input for the next level of fuzzy logic reasoning, ultimately outputting the animal welfare level of the pasture. This reduces reliance on human labor and meets the system's requirements for intelligence, automation, and controllability.

[0105] The following is combined with Figure 10 The electronic device described in this application, wherein, Figure 10 A schematic block diagram of an electronic device 1000 according to an embodiment of this application is shown.

[0106] like Figure 9 As shown, the electronic device 1000 includes: one or more memories 901 and one or more processors 902, wherein the memory 901 stores a computer program that is executed by the processor 902, and the computer program, when executed by the processor 902, causes the processor 902 to perform the animal welfare level assessment method described above.

[0107] Electronic device 900 may be part or all of a computer device that can implement animal welfare level assessment methods through software, hardware, or a combination of software and hardware.

[0108] like Figure 9 As shown, the electronic device 900 includes one or more memories 901, one or more processors 902, a display (not shown), and a communication interface, etc., which are interconnected via a bus system and / or other forms of connection mechanisms (not shown). It should be noted that... Figure 9 The components and structure of the electronic device 900 shown are merely exemplary and not limiting. The electronic device 900 may also have other components and structures as needed.

[0109] The memory 901 is used to store various data and executable program instructions generated during the operation of the animal welfare level assessment method, such as algorithms for storing various applications or implementing various specific functions. It may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0110] The memory 901 includes an internal memory 9011 and an auxiliary memory 9012.

[0111] The processor 902 may be a central processing unit (CPU), a graphics processing unit (GPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other processing units with data processing capabilities and / or instruction execution capabilities, and may be other components in the electronic device 900 to perform the desired functions.

[0112] The processor 902 includes an arithmetic logic unit 9021 and a controller 9022.

[0113] In one example, the electronic device 900 also includes an output device that can output various information (such as images or sounds) to the outside (e.g., a user), and may include one or more of a display device, a speaker, etc.

[0114] The communication interface can be any known communication protocol interface, such as a wired interface or a wireless interface. The communication interface may include one or more serial ports, USB interfaces, Ethernet ports, WiFi, wired networks, DVI interfaces, device integrated interconnect modules, or other suitable ports, interfaces, or connections.

[0115] The electronic device in this application embodiment can be connected to sensing devices via a network, such as temperature sensors, humidity sensors, light sensors, and gas sensors. Data collected by the sensing devices is transmitted through input devices in the electronic device (e.g., keyboard, mouse, and communication interface), and the calculation results are output via output devices.

[0116] Furthermore, according to embodiments of this application, a storage medium is also provided, on which program instructions are stored. When executed by a computer or processor, these program instructions are used to perform corresponding steps of the animal welfare level assessment method of this application. The storage medium may, for example, include a memory card of a smartphone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media.

[0117] The electronic device and storage medium of this application embodiment have the same advantages as the aforementioned animal welfare level assessment method because they are able to implement the aforementioned animal welfare level assessment method.

[0118] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of this application. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of this application. All such changes and modifications are intended to be included within the scope of this application as claimed in the appended claims.

[0119] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0121] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0122] Similarly, it should be understood that, in order to streamline this application and aid in understanding one or more of the various inventive aspects, features of this application may sometimes be grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of this application. However, this approach should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with features fewer than all features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.

[0123] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus so disclosed can be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0124] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of this application and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0125] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules according to the embodiments of this application. This application can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0126] It should be noted that the above embodiments are illustrative of this application and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0127] The above description is merely a specific embodiment or illustration of the embodiments of this application. The scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. The scope of protection of this application shall be determined by the scope of the claims.

Claims

1. A method of assessing the welfare level of animals in a pasture, characterized by, The method comprises: The method comprises: The method comprises: The method comprises:

2. The method of claim 1, wherein, The method comprises: The method comprises: The method comprises:

3. The method of claim 2, wherein, The method comprises: The method comprises: The method comprises: The method comprises:

4. The method of claim 2, wherein, The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: inputting the environment feature data, the feeding feature data and the behavior feature structured data into the back propagation neural network model for processing to obtain environment level data, feeding level data and behavior feature level data.

5. The method of claim 1, wherein, inputting the to-be-processed level data into a fuzzy logic controller model for processing to obtain animal welfare level data, including: processing the to-be-processed level data by using a preset membership function in the fuzzy logic controller model to obtain processed to-be-processed level data; fusing the processed to-be-processed level data based on a preset fuzzy rule to obtain animal welfare level data.

6. The method of claim 1, wherein, After the animal welfare level is obtained, the method further includes: correcting the animal welfare level data; wherein the correcting the animal welfare level data includes: collecting new behavior feature data; continuing to execute the steps of inputting the environment feature data, the feeding feature data and the new behavior feature data as to-be-processed data into the back propagation neural network model for processing to obtain new to-be-processed level data, and inputting the new to-be-processed level data into the fuzzy logic controller model for processing to obtain a new animal welfare level.

7. The method of claim 1, wherein, The collecting to-be-processed data includes at least one of: collecting the environment feature data by using an environment sensor arranged in a pasture and storing the environment feature data in an environment feature database; collecting feeding index data and storing the feeding index data in a feeding feature database; collecting video data of the to-be-evaluated animal by using a monitoring device and extracting the behavior feature data from the video data and storing the behavior feature data in a behavior feature database.

8. The method of claim 1, wherein, The behavior feature data includes coat color feature data and / or ear feature data; and the to-be-evaluated animal includes a dairy cow.

9. A system for assessing the welfare level of animals in a pasture, characterized in that, The method of claim 1 is adopted; The system includes a basic data layer, a network layer, a feature data layer, an algorithm model layer and a decision application layer: The basic data layer is configured to collect to-be-processed data, wherein the to-be-processed data includes environment feature data, feeding feature data and behavior feature data of a to-be-evaluated animal; The network layer is configured to transmit the to-be-processed data to the feature data layer for storage; The feature data layer is configured to store the to-be-processed data; The algorithm model layer is configured to process the to-be-processed data by using a back propagation neural network model to obtain to-be-processed level data, and input the to-be-processed level data into a fuzzy logic controller model for processing to obtain animal welfare level data; The decision application layer is configured to output the animal welfare level data.

10. A storage medium, characterized by The storage medium has a computer program stored thereon, and the computer program, when executed by a processor, causes the processor to execute the animal welfare level evaluation method of any one of claims 1 to 8.

Images