Livestock health monitoring method and device, computer equipment and storage medium

Abstract

The embodiment of the invention belongs to the field of artificial intelligence, and relates to a livestock health monitoring method, which comprises the following steps: acquiring livestock sound of a livestock group in a farm; enhancing the livestock sound to obtain a first sound; extracting a voiceprint feature of the first sound, and inputting the voiceprint feature into a health assessment model to obtain a health state score of the livestock group; obtaining a current seasonal influence parameter, and inputting the seasonal influence parameter into the seasonal influence evaluation model to obtain a seasonal correction score of livestock health; calculating the health state score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock group; and generating a health monitoring result of the livestock group according to the health score. The invention further provides a livestock health monitoring device, computer equipment and a storage medium. In addition, the invention also relates to a block chain technology, and the livestock sound can be stored in a block chain. According to the invention, the health state of the livestock can be accurately monitored in real time.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a livestock health monitoring method, device, computer equipment and storage medium. Background Art

[0002] In modern animal husbandry, livestock health management is a key link in ensuring the profitability and safety of farms. Traditional health monitoring methods mainly rely on manual observation and regular physical examinations, which have significant limitations. Manual observation is limited by the experience and energy of the observer, and it is difficult to detect early symptoms of the disease in time, which may cause the disease to spread in the group, causing large-scale infection and huge economic losses. Even through regular physical examinations, it is difficult to monitor the health status of livestock in real time, and often only obvious symptoms can be detected, which makes disease prevention and control lagging.

[0003] As the scale of livestock farming expands, the health status of individual livestock becomes more difficult to manage effectively, making it difficult to take effective preventive measures in a timely manner when a disease breaks out. This not only increases the economic burden on farmers, but also increases the risk of insurance companies in the claims process, making it impossible to effectively intervene before the disease spreads and reduce potential losses. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a livestock health monitoring method, apparatus, computer equipment and storage medium to monitor the health status of livestock in real time and accurately.

[0005] In order to solve the above technical problems, the present application embodiment provides a livestock health monitoring method, which adopts the following technical solution:

[0006] Acquire livestock sounds from livestock herds in breeding farms;

[0007] Performing enhancement processing on the livestock sound to obtain a first sound;

[0008] Extracting voiceprint features of the first sound, and inputting the voiceprint features into a health assessment model to obtain a health status score of the livestock group;

[0009] Obtaining current seasonal impact parameters, and inputting the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health;

[0010] Calculating the health status score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock herd;

[0011] A health monitoring result of the livestock group is generated according to the health score.

[0012] In order to solve the above technical problems, the present application also provides a livestock health monitoring device, which adopts the following technical solution:

[0013] A sound acquisition module, used for acquiring livestock sounds of livestock groups in a breeding farm;

[0014] A sound enhancement module, used for enhancing the livestock sound to obtain a first sound;

[0015] A voiceprint assessment module, configured to extract voiceprint features of the first sound and input the voiceprint features into a health assessment model to obtain a health status score of the livestock group;

[0016] A seasonal assessment module, used to obtain current seasonal impact parameters and input the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health;

[0017] A health calculation module, used for calculating the health status score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock group;

[0018] The result generating module is used to generate the health monitoring result of the livestock group according to the health score.

[0019] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0020] Acquire livestock sounds from livestock herds in breeding farms;

[0021] Performing enhancement processing on the livestock sound to obtain a first sound;

[0022] Extracting voiceprint features of the first sound, and inputting the voiceprint features into a health assessment model to obtain a health status score of the livestock group;

[0023] Obtaining current seasonal impact parameters, and inputting the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health;

[0024] Calculating the health status score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock herd;

[0025] A health monitoring result of the livestock group is generated according to the health score.

[0026] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0027] Acquire livestock sounds from livestock herds in breeding farms;

[0028] Performing enhancement processing on the livestock sound to obtain a first sound;

[0029] Extracting voiceprint features of the first sound, and inputting the voiceprint features into a health assessment model to obtain a health status score of the livestock group;

[0030] Obtaining current seasonal impact parameters, and inputting the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health;

[0031] Calculating the health status score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock herd;

[0032] A health monitoring result of the livestock group is generated according to the health score.

[0033] Compared with the prior art, the embodiments of the present application mainly have the following beneficial effects: livestock sounds of a livestock herd in a farm are obtained, and enhanced processing is performed to obtain a first sound, which can eliminate or reduce background noise and enhance the clarity of the effective signal; the voiceprint features of the first sound are extracted and input into a health assessment model, which can quickly and accurately assess the health status of the livestock herd and generate a preliminary health status score; the current seasonal impact parameters are obtained and input into the seasonal impact assessment model to obtain a seasonal correction score for livestock health, which takes into account the potential impact of seasonal changes on livestock health, making the health monitoring results more comprehensive and accurate; the health status score and the seasonal correction score are weighted and calculated according to a preset dynamic coefficient to obtain a health score for the livestock herd, which reflects the current health status of the livestock herd, and then the health monitoring results of the livestock herd are generated according to the health score; the present application realizes dynamic monitoring of the health status of the livestock herd, and improves the real-time and accuracy of livestock health status monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0036] Figure 2 is a flow chart of an embodiment of a livestock health monitoring method according to the present application;

[0037] Figure 3is a schematic structural diagram of an embodiment of a livestock health monitoring device according to the present application;

[0038] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0040] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0041] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0042] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0043] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0044] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0045] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0046] It should be noted that the livestock health monitoring method provided in the embodiment of the present application is generally executed by a server, and accordingly, the livestock health monitoring device is generally arranged in the server.

[0047] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0048] Continue to refer Figure 2 , shows a flow chart of an embodiment of a livestock health monitoring method according to the present application. The livestock health monitoring method comprises the following steps:

[0049] Step S201, obtaining livestock sounds of a livestock group in a breeding farm.

[0050] In this embodiment, the livestock health monitoring method is executed on an electronic device (eg Figure 1 The server shown in the figure) can communicate with the terminal device through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0051] Specifically, the sound signals emitted by livestock are collected in real time from the farm environment. The sound information of livestock can reflect their health status, and these sounds can be captured by devices such as acoustic sensors or microphones. The acquisition of sound signals needs to be carried out without disturbing the livestock to obtain the most natural and accurate sound data.

[0052] Step S202, enhancing the livestock sound to obtain a first sound.

[0053] Specifically, the original livestock sounds collected may contain noise or other interfering signals, which will affect the accuracy of subsequent analysis. The livestock sounds are enhanced to eliminate or reduce background noise, enhance the clarity of effective signals, and obtain a cleaner first sound. Enhancement processing usually includes filtering, noise reduction and other processing to ensure that the first sound obtained is more in line with the needs of health assessment.

[0054] Furthermore, the above step S202 may include: preprocessing the livestock sounds to obtain multiple sound segments; performing multi-layer convolution operations on each sound segment to obtain high-dimensional features of the livestock sounds; performing dimensionality reduction processing on the high-dimensional features to obtain core features of the livestock sounds; inputting the core features into a recurrent neural network for time series feature analysis to obtain a first sound after sound enhancement.

[0055] Specifically, the livestock sounds are preprocessed, such as denoising, normalization and framing. Framing is to divide the livestock sounds into multiple shorter sound segments. Each sound segment represents the sound information within a time window.

[0056] Each sound clip is input into a convolutional neural network (CNN) for feature extraction. The convolution layer of CNN first extracts low-level features of the sound clip (such as local frequency components in the spectrogram); through multiple layers of convolution operations, CNN gradually extracts high-dimensional features in the sound clip, including information such as the frequency, amplitude, and duration of the sound, which is the result of the gradual abstraction of the original sound data in the convolution network. High-dimensional features can distinguish livestock sounds from background noise. For example, CNN can identify specific frequency and tone patterns through convolution and pooling operations, thereby filtering out obvious background noise.

[0057] High-dimensional features contain a lot of information, which will lead to excessive computational overhead and difficulty in extracting key patterns. Dimensionality reduction technology is used to compress these high-dimensional features, retain the most representative features, reduce redundant information, and integrate them into a fixed-length feature vector to form a core feature. The core feature is the essence of the high-dimensional feature and can represent the original sound signal in a more compact form.

[0058] The core features are input into a recurrent neural network (RNN) for temporal feature analysis. RNN is good at processing sequence data and can capture the temporal dependencies in sound signals. Through its recursive structure, RNN processes the temporal associations between sound segments, further filters out background noise, and enhances the target sound signal. RNN can identify sound patterns that remain consistent over a continuous period of time and exclude those short-lived and irrelevant noise signals. Through RNN processing, the key information in the original sound signal is extracted and enhanced, and the enhanced first sound is finally generated. The first sound has a higher signal-to-noise ratio and clearer features.

[0059] In this embodiment, the livestock sounds are divided into multiple sound segments, which is conducive to analyzing the sound features in each time window; the multi-layer convolution operation extracts the high-dimensional features in the sound segments, and then through dimensionality reduction processing, the high-dimensional features are compressed into more representative core features, reducing information redundancy and retaining the most important feature parts; the recurrent neural network performs time series feature analysis on the core features, further enhancing the key information of the sound, removing noise and interference, and the generated first sound signal has higher quality and clarity, which improves the reliability and availability of the sound signal and ensures the accuracy of subsequent data processing.

[0060] Step S203: extract the voiceprint features of the first sound, and input the voiceprint features into a health assessment model to obtain a health status score of the livestock herd.

[0061] Specifically, the first sound is used to extract voiceprint features. Voiceprint features are the extraction of unique information from livestock sounds, which can reflect the physiological and psychological status of livestock. The extracted voiceprint features are input into the health assessment model, which calculates the health status score of the livestock herd based on historical data and voiceprint features, and the score reflects the overall health status of the current livestock herd.

[0062] Furthermore, the voiceprint features are voiceprint features of a preset number of sound clips. The above step of inputting the voiceprint features into the health assessment model to obtain the health status score of the livestock herd may include: for the voiceprint features of each sound clip, matching the voiceprint features of the sound clip with each inventory voiceprint of known health status in the voiceprint feature library to obtain the health index score of the sound clip; and calculating the health status score of the livestock herd based on the preset number, the health index score of each sound clip, the preset target health score and the maximum health index difference.

[0063] Specifically, when extracting the voiceprint feature of the first sound, the first sound is divided into a preset number of sound segments, and then the voiceprint feature of each sound segment is extracted. The voiceprint feature is a set of parameters that reflect the characteristics of livestock sounds, including frequency, pitch, time duration, etc.

[0064] The voiceprint features of each sound clip are matched with the stock voiceprints of known health status in the voiceprint feature library. The voiceprint feature library is a pre-established database that contains a large number of labeled livestock voiceprint samples, including voiceprint features of health, sub-health and different disease states. The specific calculation method can be to calculate the similarity between the voiceprint features of the sound clip and the stock voiceprints of different health status types in the voiceprint feature library (such as through cosine similarity, Euclidean distance, etc.), and convert the similarity into a health index score to reflect the degree of matching between the voiceprint features.

[0065] Target health score H target It is a preset value, representing the average health index score of a healthy livestock herd under ideal circumstances. It can be determined by the health data set during model training and represents the standard of healthy livestock voiceprints.

[0066] H max The maximum health index difference, also called the maximum possible health index difference, is also a preset value, which represents the maximum possible difference between voiceprints in different health states defined in the model. This value is set based on historical data during the design phase.

[0067] The overall health status score is calculated based on the health indicator score of each sound clip, the target health score and the maximum health indicator difference.

[0068] Furthermore, the health status score is expressed as:

[0069]

[0070] Wherein, N is a preset number, that is, the first sound is divided into N sound segments, H i is the health index score of the ith sound clip, H target is the target health score, H max is the maximum health index difference, S health The health status score.

[0071] In this embodiment, the accuracy and objectivity of livestock health status assessment are improved by calculating and standardizing the health indicator scores of multiple sound clips; the standardization process reduces the errors caused by differences in conditions in different sound clips, ensuring the uniformity and reliability of the calculation results; the weighted average of multiple sound clips can comprehensively reflect the overall health status of the livestock group, avoiding assessment bias caused by abnormalities in individual clips, and is suitable for the health monitoring needs of farms; it can capture subtle changes in health status in a timely manner, and trigger an early warning mechanism when the health status score is lower than a preset threshold, so as to take measures early to prevent the spread of the disease and reduce losses.

[0072] In this embodiment, voiceprint features are extracted for a preset number of sound clips, and the voiceprint features are matched with the stock voiceprints of known health status, so that the health index score of each sound clip can be accurately obtained, the health status of the livestock can be analyzed in detail, and the evaluation accuracy is improved; the health index score of each sound clip is standardized, and the calculated health status score not only takes into account each sound clip, but also ensures the balance and representativeness of the evaluation through the preset number, so that the health status score can accurately reflect the overall health status of the livestock herd, reduces the risk of misjudgment caused by abnormalities of individual sound clips, and enhances the sensitivity to early disease states.

[0073] Step S204, obtaining the current seasonal impact parameters, and inputting the seasonal impact parameters into the seasonal impact assessment model to obtain the seasonal correction score of livestock health.

[0074] Specifically, the health status of livestock is also affected by seasonal factors, such as temperature and humidity. The current seasonal impact parameters are obtained and input into the seasonal impact assessment model to calculate the potential correction score for livestock health. The seasonal correction score can be used to adjust the health status score to reflect the actual impact of seasonal changes on livestock health.

[0075] Furthermore, the seasonal impact parameters include a comprehensive impact coefficient, a temperature parameter, a humidity parameter and disease occurrence data. The above step of inputting the seasonal impact parameters into the seasonal impact assessment model to obtain the seasonal correction score for livestock health may include: calculating the temperature impact score based on the temperature parameters; calculating the humidity impact score based on the humidity parameters; calculating the disease risk score based on the disease occurrence data; and calculating the seasonal correction score for livestock health based on the comprehensive impact coefficient, the temperature impact score, the humidity impact score and the disease risk score.

[0076] Specifically, the current seasonal impact parameters are obtained and analyzed, including comprehensive impact coefficient, temperature parameters, humidity parameters and disease occurrence data.

[0077] Among them, the comprehensive impact coefficient (β) is preset, which represents the overall impact weight of seasonal factors on livestock health. Temperature parameters and humidity parameters can be environmental data collected in real time, reflecting the temperature and humidity conditions of the current farm. Disease occurrence data is based on historical data and epidemiological models, reflecting the risk of disease outbreaks in the current season.

[0078] The temperature impact score is calculated based on the temperature parameters, which can quantify the impact of temperature on livestock health by comparing the current temperature to the ideal temperature range. Generally, if the current temperature deviates from the ideal temperature range, the temperature impact score will be lower, indicating an adverse impact of temperature on the health of the livestock herd.

[0079] The humidity impact score can reflect the impact of current humidity conditions on the health of the livestock herd by comparing the degree of deviation of the current humidity from the ideal humidity range. Generally, the greater the deviation of the current humidity from the ideal humidity range, the lower the humidity impact score.

[0080] The disease risk score is obtained by analyzing the disease outbreak risk data of the current season, and is used to reflect the possibility and severity of disease transmission. In one embodiment, the disease risk score is equal to the ratio of the disease outbreak rate of the current season to the average disease outbreak rate of the historical season; if the disease outbreak rate of the current season is higher than the average level, the disease risk score increases, indicating a higher disease risk.

[0081] Finally, the comprehensive impact coefficient, temperature impact score, humidity impact score and disease risk score are combined to calculate the seasonal correction score of livestock health. The seasonal correction score is used to adjust and correct the initially calculated health status score to more comprehensively and accurately reflect the actual impact of seasonal changes on livestock health and improve the accuracy of health assessment.

[0082] Furthermore, the seasonal correction score is expressed as:

[0083] S seasonal =β×(S temp +S humidity +S disease_risk )

[0084] Among them, β is the comprehensive influence coefficient, S temp is the temperature influence fraction, S humidity is the humidity impact fraction, S disease_risk is the disease risk score, S seasonal is the seasonal correction score.

[0085] In this embodiment, the seasonal correction score of livestock health can be accurately calculated based on the comprehensive influence coefficient, temperature influence score, humidity influence score and disease risk score, thereby ensuring the accuracy of livestock health monitoring.

[0086] In this embodiment, through real-time analysis of temperature and humidity parameters, the impact of environmental conditions on livestock health is quantified, providing a more detailed health status assessment; based on historical disease occurrence data and disease risk scores of current seasonal conditions, potential disease outbreak risks can be predicted and identified in advance; the comprehensive impact coefficient provides flexibility and adaptability for adjusting health status under different seasonal conditions, ensuring the accuracy and consistency of health status assessment results in different seasons.

[0087] Step S205, calculating the health status score and the seasonal correction score according to the preset dynamic coefficient to obtain the health score of the livestock herd.

[0088] Specifically, a preset dynamic coefficient is obtained, and the obtained health status score and seasonal correction score are weighted and calculated. The setting of the dynamic coefficient can be adjusted according to the specific conditions of different farms, so as to ensure that the calculation of the health score is more flexible and accurate. The calculated health score is the final quantitative evaluation index of the current health status of the livestock herd.

[0089] Furthermore, the health score is expressed as:

[0090] W=α×S health +(1-α)×S seasonal

[0091] Among them, W is the health score, α is the dynamic coefficient, S health is the health status score, S seasonal is the seasonal correction score.

[0092] The dynamic coefficient is expressed as:

[0093]

[0094] Among them, T is the current temperature, T0 is the reference temperature, γ is the temperature sensitivity coefficient, and e is the natural constant.

[0095] In this embodiment, the livestock health status score and seasonal correction score are weighted by the dynamic coefficient to calculate the health score of the livestock group. The dynamic coefficient can flexibly adjust the health status assessment in different seasons and environments, improving the accuracy and adaptability of health monitoring; the dynamic coefficient is adjusted by real-time temperature data to ensure that the health status assessment can be appropriately corrected under extreme temperature conditions to reflect the actual impact of temperature on livestock health. The adjustment of the dynamic coefficient is also based on the temperature sensitivity coefficient and the reference temperature. When the temperature deviates significantly from the reference value, the weight of the seasonal correction score can be automatically increased to more accurately reflect the impact of seasonal changes on health status; the health score comprehensively considers the current health status and seasonal factors, and can also dynamically adapt to environmental changes, improving the reliability and real-time performance of the overall assessment.

[0096] Step S206, generating health monitoring results of the livestock herd according to the health scores.

[0097] Specifically, a specific health monitoring result is generated based on the calculated health score. The risk level of the livestock herd can be determined based on the score interval of the health score. For example, if the health score W is in the score interval [0,1], the risk level is low risk; if the health score W is in the score interval (1,2], the risk level is general risk; if the health score W is in the score interval (2,3], the risk level is high risk; if the health score W is in the score interval (3,7], the risk level is major risk. Livestock herds can also be divided into healthy, sub-healthy, diseased and other health states based on the health score.

[0098] Based on the risk level or health status, warning signals, health advice or automated management instructions can also be generated to help farm managers take timely measures to ensure the health of the livestock herd and breeding benefits.

[0099] It will be appreciated that the present application may also be used to conduct health assessments on individual livestock.

[0100] In this embodiment, livestock sounds of a livestock herd in a farm are obtained, and enhanced processing is performed to obtain a first sound, which can eliminate or reduce background noise and enhance the clarity of effective signals; the voiceprint features of the first sound are extracted and input into a health assessment model, which can quickly and accurately assess the health status of the livestock herd and generate a preliminary health status score; the current seasonal impact parameter is obtained and input into the seasonal impact assessment model to obtain a seasonal correction score for livestock health, which takes into account the potential impact of seasonal changes on livestock health, making the health monitoring results more comprehensive and accurate; the health status score and the seasonal correction score are weighted according to a preset dynamic coefficient to obtain a health score for the livestock herd, which reflects the current health status of the livestock herd, and then the health monitoring results of the livestock herd are generated according to the health score; the present application realizes dynamic monitoring of the health status of the livestock herd, and improves the real-time and accuracy of livestock health status monitoring.

[0101] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned livestock sounds, the above-mentioned livestock sounds can also be stored in a node of a blockchain.

[0102] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0103] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0104] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0105] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0106] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0107] Further references Figure 3 , as a response to the above Figure 2 In order to realize the method shown in the figure, the present application provides an embodiment of a livestock health monitoring device, and the device embodiment is Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0108] like Figure 3As shown, the livestock health monitoring device 300 described in this embodiment includes: a sound acquisition module 301, a sound enhancement module 302, a voiceprint evaluation module 303, a season evaluation module 304, a health calculation module 305 and a result generation module 306, wherein:

[0109] The sound acquisition module 301 is used to acquire the livestock sounds of the livestock in the farm.

[0110] The sound enhancement module 302 is used to enhance the livestock sound to obtain the first sound.

[0111] The voiceprint assessment module 303 is used to extract the voiceprint features of the first sound and input the voiceprint features into the health assessment model to obtain the health status score of the livestock group.

[0112] The seasonal assessment module 304 is used to obtain the current seasonal impact parameters and input the seasonal impact parameters into the seasonal impact assessment model to obtain the seasonal correction score of livestock health.

[0113] The health calculation module 305 is used to calculate the health status score and the seasonal correction score according to the preset dynamic coefficient to obtain the health score of the livestock group.

[0114] The result generation module 306 is used to generate health monitoring results of the livestock group according to the health scores.

[0115] In this embodiment, livestock sounds of a livestock herd in a farm are obtained, and enhanced processing is performed to obtain a first sound, which can eliminate or reduce background noise and enhance the clarity of effective signals; the voiceprint features of the first sound are extracted and input into a health assessment model, which can quickly and accurately assess the health status of the livestock herd and generate a preliminary health status score; the current seasonal impact parameter is obtained and input into the seasonal impact assessment model to obtain a seasonal correction score for livestock health, which takes into account the potential impact of seasonal changes on livestock health, making the health monitoring results more comprehensive and accurate; the health status score and the seasonal correction score are weighted according to a preset dynamic coefficient to obtain a health score for the livestock herd, which reflects the current health status of the livestock herd, and then the health monitoring results of the livestock herd are generated according to the health score; the present application realizes dynamic monitoring of the health status of the livestock herd, and improves the real-time and accuracy of livestock health status monitoring.

[0116] In some optional implementations of this embodiment, the sound enhancement module 302 may include:

[0117] The preprocessing submodule is used to preprocess the livestock sounds to obtain multiple sound clips.

[0118] The convolution operation submodule is used to perform multi-layer convolution operations on each sound segment to obtain high-dimensional features of livestock sounds.

[0119] The dimensionality reduction processing submodule is used to reduce the dimensionality of high-dimensional features to obtain the core features of livestock sounds.

[0120] The time series analysis submodule is used to input the core features into the recurrent neural network for time series feature analysis to obtain the first sound after sound enhancement.

[0121] In some optional implementations of this embodiment, the voiceprint evaluation module 303 may include:

[0122] The voiceprint matching submodule is used to match the voiceprint features of each sound clip with each stored voiceprint of known health status in the voiceprint feature library to obtain a health index score of the sound clip.

[0123] The score calculation submodule is used to calculate the health status score of the livestock group based on the preset number, the health indicator score of each sound clip, the preset target health score and the maximum health indicator difference.

[0124] In some optional implementations of this embodiment, the health status score is expressed as:

[0125]

[0126] Among them, N is the preset number, H i is the health index score of the ith sound clip, H target is the target health score, H max is the maximum health index difference, S health The health status score.

[0127] In some optional implementations of this embodiment, the seasonal impact parameters include a comprehensive impact coefficient, a temperature parameter, a humidity parameter, and disease occurrence data, and the seasonal assessment module 304 may include:

[0128] The temperature calculation submodule is used to calculate the temperature impact score based on the temperature parameters.

[0129] The humidity calculation submodule is used to calculate the humidity impact score based on the humidity parameters.

[0130] The disease calculation submodule is used to calculate the disease risk score based on the disease occurrence data.

[0131] The seasonal calculation submodule is used to calculate the seasonal correction score of livestock health based on the comprehensive impact coefficient, temperature impact score, humidity impact score and disease risk score.

[0132] In some optional implementations of this embodiment, the seasonal correction score is expressed as:

[0133] S seasonal =β×(S temp +S humidity +S disease_risk )

[0134] Among them, β is the comprehensive influence coefficient, S temp is the temperature influence fraction, S humidity is the humidity impact fraction, S disease_risk is the disease risk score, S seasonal is the seasonal correction score.

[0135] In some optional implementations of this embodiment, the health score is expressed as:

[0136] W=α×S health +(1-α)×S seasonal

[0137] Among them, W is the health score, α is the dynamic coefficient, S health is the health status score, S seasonal is the seasonal correction score;

[0138] The dynamic coefficient is expressed as:

[0139]

[0140] Among them, T is the current temperature, T0 is the reference temperature, γ is the temperature sensitivity coefficient, and e is the natural constant.

[0141] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0142] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 4 with a memory 41, a processor 42, and a network interface 43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0143] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0144] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the livestock health monitoring method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0145] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the livestock health monitoring method.

[0146] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0147] The computer device provided in this embodiment can execute the livestock health monitoring method described above. The livestock health monitoring method here can be the livestock health monitoring method of each of the above embodiments.

[0148] In this embodiment, livestock sounds of a livestock herd in a farm are obtained, and enhanced processing is performed to obtain a first sound, which can eliminate or reduce background noise and enhance the clarity of effective signals; the voiceprint features of the first sound are extracted and input into a health assessment model, which can quickly and accurately assess the health status of the livestock herd and generate a preliminary health status score; the current seasonal impact parameter is obtained and input into the seasonal impact assessment model to obtain a seasonal correction score for livestock health, which takes into account the potential impact of seasonal changes on livestock health, making the health monitoring results more comprehensive and accurate; the health status score and the seasonal correction score are weighted according to a preset dynamic coefficient to obtain a health score for the livestock herd, which reflects the current health status of the livestock herd, and then the health monitoring results of the livestock herd are generated according to the health score; the present application realizes dynamic monitoring of the health status of the livestock herd, and improves the real-time and accuracy of livestock health status monitoring.

[0149] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the livestock health monitoring method as described above.

[0150] In this embodiment, livestock sounds of a livestock herd in a farm are obtained, and enhanced processing is performed to obtain a first sound, which can eliminate or reduce background noise and enhance the clarity of effective signals; the voiceprint features of the first sound are extracted and input into a health assessment model, which can quickly and accurately assess the health status of the livestock herd and generate a preliminary health status score; the current seasonal impact parameter is obtained and input into the seasonal impact assessment model to obtain a seasonal correction score for livestock health, which takes into account the potential impact of seasonal changes on livestock health, making the health monitoring results more comprehensive and accurate; the health status score and the seasonal correction score are weighted according to a preset dynamic coefficient to obtain a health score for the livestock herd, which reflects the current health status of the livestock herd, and then the health monitoring results of the livestock herd are generated according to the health score; the present application realizes dynamic monitoring of the health status of the livestock herd, and improves the real-time and accuracy of livestock health status monitoring.

[0151] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0152] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A livestock health monitoring method, characterized in that: The steps include: Acquire livestock sounds from livestock herds in breeding farms; Performing enhancement processing on the livestock sound to obtain a first sound; Extracting voiceprint features of the first sound, and inputting the voiceprint features into a health assessment model to obtain a health status score of the livestock group; Obtaining current seasonal impact parameters, and inputting the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health; Calculating the health status score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock herd; A health monitoring result of the livestock group is generated according to the health score.

2. The livestock health monitoring method according to claim 1, characterized in that: The step of enhancing the livestock sound to obtain the first sound comprises: Preprocessing the livestock sounds to obtain a plurality of sound clips; Performing multi-layer convolution operations on each sound segment to obtain high-dimensional features of the livestock sound; Performing dimensionality reduction processing on the high-dimensional features to obtain the core features of the livestock sounds; The core features are input into a recurrent neural network for time series feature analysis to obtain a first sound after sound enhancement.

3. The livestock health monitoring method according to claim 1, characterized in that: The voiceprint features are voiceprint features of a preset number of sound segments, and the step of inputting the voiceprint features into a health assessment model to obtain the health status score of the livestock group includes: For each voiceprint feature of a sound clip, the voiceprint feature of the sound clip is matched with each stored voiceprint of a known health status in a voiceprint feature library to obtain a health index score of the sound clip; The health status score of the livestock group is calculated based on the preset number, the health indicator score of each sound clip, the preset target health score and the maximum health indicator difference.

4. The livestock health monitoring method according to claim 3, characterized in that: The health status score is expressed as: Among them, N is the preset number, H i is the health index score of the ith sound clip, H target is the target health score, H max is the maximum health index difference, S health The health status score.

5. The livestock health monitoring method according to claim 1, characterized in that: The seasonal impact parameters include a comprehensive impact coefficient, a temperature parameter, a humidity parameter and disease occurrence data. The step of inputting the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health includes: Calculating a temperature impact score based on the temperature parameter; Calculating a humidity impact score based on the humidity parameter; Calculating a disease risk score based on the disease occurrence data; A seasonal correction score for livestock health is calculated based on the comprehensive impact coefficient, the temperature impact score, the humidity impact score and the disease risk score.

6. The livestock health monitoring method according to claim 5, characterized in that: The seasonal correction score is expressed as: S seasonal =β×(S temp +S humidity +S disease_risk ) Among them, β is the comprehensive influence coefficient, S temp is the temperature influence fraction, S humidity is the humidity impact fraction, S disease_risk is the disease risk score, S seasonal is the seasonal correction score.

7. The livestock health monitoring method according to claim 1, characterized in that: The health score is expressed as: W=α×S health +(1-a)×S seasonal Among them, W is the health score, α is the dynamic coefficient, S health is the health status score, S seasonal is the seasonal correction score; The dynamic coefficient is expressed as: Among them, T is the current temperature, T0 is the reference temperature, γ is the temperature sensitivity coefficient, and e is the natural constant.

8. A livestock health monitoring device, characterized in that: include: A sound acquisition module, used for acquiring livestock sounds of livestock groups in a breeding farm; A sound enhancement module, used for enhancing the livestock sound to obtain a first sound; A voiceprint assessment module, configured to extract voiceprint features of the first sound and input the voiceprint features into a health assessment model to obtain a health status score of the livestock group; A seasonal assessment module, used to obtain current seasonal impact parameters and input the seasonal impact parameters into a seasonal impact assessment model to obtain a seasonal correction score for livestock health; A health calculation module, used for calculating the health status score and the seasonal correction score according to a preset dynamic coefficient to obtain a health score of the livestock group; The result generating module is used to generate the health monitoring result of the livestock group according to the health score.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the livestock health monitoring method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the livestock health monitoring method according to any one of claims 1 to 7 are implemented.

Images