A method and system for detecting harmful factors in animal hides for the prevention and control of animal pathogens

Abstract

The present invention belongs to the technical field of disease source prevention and control, and particularly relates to a method and system for detecting harmful factors of animal hides for animal disease source prevention and control. The system includes: a multi-modal sensor detection module capable of detecting microorganisms, chemical residues, and other harmful factors on animal hides respectively; an intelligent analysis and machine learning module that fuses the multi-modal data collected by the sensors and classifies, identifies, and analyzes various data through machine learning algorithms; continuously optimizes the detection model through machine learning, and gradually improves the recognition accuracy of pathogens and harmful factors; an Internet of Things real-time monitoring and automatic alarm module that, through Internet of Things technology, uploads the detection data of each sensor to the cloud platform in real time for managers and technicians to monitor the status of the hides at any time. The present invention realizes the real-time and automated detection of harmful factors of animal hides, helps to improve the overall effect of animal disease source prevention and control, and has a wide application prospect.

Description

Technical Field

[0001] The present invention belongs to, but is not limited to, the technical field of disease source prevention and control, and particularly relates to a method and system for detecting harmful factors on animal hides for animal disease source prevention and control. Background Art

[0002] Currently, in the detection of harmful factors on animal hides, manual detection methods or basic technologies such as spectral analysis and microscopic analysis are usually used. These methods can identify harmful substances (such as microorganisms, parasites, or chemical pollutants) on the surface of animal hides, but there are certain limitations:

[0003] Low detection efficiency: Traditional detection methods require manual operation, and the detection process for complex pollutants is cumbersome, time-consuming, and inefficient.

[0004] Limited detection accuracy: Existing detection equipment has insufficient detection sensitivity for minute pathogens (such as bacteria, viruses, etc.), and cannot efficiently and accurately perform comprehensive detection of multiple harmful factors.

[0005] Poor real-time performance: Existing technologies usually cannot achieve real-time monitoring of the surface of hides, and it is difficult to detect and prevent harmful factors on animal hides in the early stage.

[0006] To improve the detection efficiency and accuracy of harmful factors on animal hides, the following technical problems need to be solved:

[0007] Comprehensive detection of multiple pathogens or harmful factors: Existing technologies are difficult to efficiently detect different types of harmful factors (such as viruses, bacteria, chemicals, etc.) in the same system, and lack a unified solution.

[0008] Real-time and rapid detection: Traditional detection technologies cannot analyze harmful factors on hides in real time, especially in the early stage of pathogen infection, and it is difficult to achieve the effect of timely prevention.

[0009] Low degree of automation: There are many manual detection steps, and it depends on the experience of technicians, which affects the standardization and consistency of detection results. Summary of the Invention

[0010] In view of the problems existing in the prior art, the present invention provides a method and system for detecting harmful factors on animal hides for animal disease source prevention and control.

[0011] The present invention is implemented as follows. An animal hide harmful factor detection system for animal disease source prevention and control, the system includes:

[0012] A multimodal sensor detection module, which integrates various detection means such as spectral analysis, fluorescence sensing, and micro-electrochemical sensing, and can respectively detect microorganisms, chemical residues, and other harmful factors on animal hides;

[0013] Intelligent analysis and machine learning module. The system fuses the multi-modal data collected by sensors and classifies, identifies, and analyzes various data through machine learning algorithms such as random forest and support vector machine (SVM), so as to achieve rapid detection and classification of different types of harmful factors in hides; continuously optimizes the detection model through machine learning to gradually improve the recognition accuracy of pathogens and harmful factors.

[0014] Internet of Things real-time monitoring and automatic alarm module. The system uploads the detection data of each sensor to the cloud platform in real time through Internet of Things technology for managers and technicians to monitor the status of hides at any time; an automatic alarm function is set. When harmful factors exceeding the standard, such as the number of microorganisms or the concentration of chemical substances, are detected, the system will send alarm notifications in the form of text messages, emails, etc., to remind relevant personnel to take prevention and control measures in time; through the cloud management platform, users can remotely access the detection data, generate reports, and adjust the detection strategy as needed.

[0015] Furthermore, the specific solution of the multi-modal sensor detection module includes:

[0016] Spectral sensing: Use near-infrared or mid-infrared spectroscopy to scan the surface of the hide to detect whether there are harmful chemical residues or pollutants on the hide.

[0017] Fluorescence sensing: Combine specific fluorescent labeling substances with harmful microorganisms such as bacteria and viruses, and use fluorescence signals for detection to improve the sensitivity of pathogen detection.

[0018] Electrochemical sensor: Detect specific harmful factors on the surface of the hide, such as heavy metal and chemical agent concentration, through electrochemical analysis technology.

[0019] Furthermore, the specific technical solution of the intelligent analysis and machine learning module includes:

[0020] Training data set: The system uses a known data set of harmful factors in animal hides for model training so that it can identify and classify different types of harmful factors.

[0021] Adaptive learning: As data accumulates during actual use, the system automatically optimizes the algorithm to gradually improve the detection accuracy and efficiency.

[0022] Furthermore, the implementation method of the intelligent analysis and machine learning module specifically includes:

[0023] (1) Multi-source data fusion

[0024] Data acquisition: Collect and uniformly transmit the multi-modal data generated by spectral, fluorescence, and electrochemical sensors to the data processing unit.

[0025] Feature extraction: Extract features from the output data of each sensor, such as absorption peaks in spectral data, fluorescence signal intensity, potential changes in electrochemical signals, etc., to form feature vectors;

[0026] Data fusion algorithm: Use the weighted fusion algorithm to perform weighted fusion on data from different sensors. Through reasonable weight allocation, ensure that the detection results of different sensors are comprehensively analyzed;

[0027] (2) Implementation with machine learning

[0028] Model training: Use a pre-prepared database of harmful factors in animal skins, which contains the spectral, fluorescence, and electrochemical characteristics of various known harmful factors; Use this data for training and build a classification model using algorithms such as random forest RF and support vector machine SVM;

[0029] Model deployment: The trained model is deployed into the system. By processing the data collected in real time by the sensors, quickly determine whether there are harmful factors on the skin and classify them;

[0030] Adaptive learning: The system continuously accumulates new detection data and uses online learning algorithms such as incremental SVM to update the model in real time, improving the classification accuracy and response speed of the model;

[0031] Furthermore, the implementation method of the Internet of Things real-time monitoring and automatic alarm module specifically includes:

[0032] (1) Real-time data collection and transmission

[0033] Sensor data collection: The system uploads the real-time data detected by various sensors through wireless communication such as Wi-Fi, 4G / 5G, LoRa, etc. to the cloud platform for unified processing through the integrated Internet of Things sensing module;

[0034] Data transmission protocol: Adopt standard Internet of Things data transmission protocols such as MQTT or CoAP to ensure low-latency and highly reliable data transmission;

[0035] (2) Implementation of automatic alarm function

[0036] Threshold setting: The system automatically triggers an alarm when the threshold, such as the concentration of microorganisms or chemical pollutants, is exceeded according to the pre-set threshold;

[0037] Alarm notification: When a harmful factor exceeding the standard is detected, the system automatically generates an alarm message and notifies the manager through forms such as text messages, emails, or APP push; The alarm message includes detailed data such as the detection location, the type and concentration of the exceeded standard;

[0038] (3) Cloud platform and remote management

[0039] Cloud data storage: Through Internet of Things technology, all detected data is uploaded to the cloud platform in real time for long-term storage and management, facilitating the retrieval of historical data for analysis at any time;

[0040] Remote access: Users can remotely access cloud data through a Web interface or a mobile application, monitor the status of animal skins, and generate detection reports;

[0041] Dynamic strategy adjustment: Based on real-time data and detection results, users can remotely adjust detection strategies through the cloud platform, such as changing the detection frequency of sensors, threshold settings, etc., to improve prevention and control efficiency.

[0042] Furthermore, the integration and implementation method of the animal skin harmful factor detection system for animal disease source prevention and control is as follows:

[0043] S1: System integration, integrating multi-modal sensors, a data processing unit, a machine learning module, and an Internet of Things monitoring module into an embedded system; the hardware part adopts a modular design for easy expansion and maintenance; the software part adopts a layered architecture, and the data acquisition, analysis, storage, and monitoring modules are relatively independent to ensure the scalability and stability of the system;

[0044] S2: On-site deployment and debugging, On-site deployment: The system can be portably installed at detection points or on factory production lines for skin detection through mobile devices or fixed detection stations; System debugging: Before the actual application of the system, the system needs to be debugged according to specific scenarios, including sensor calibration, model parameter adjustment, and network connection of Internet of Things devices;

[0045] S3: Long-term maintenance and upgrade, Sensor maintenance: Regularly maintain and calibrate spectral, fluorescence, and electrochemical sensors to ensure detection accuracy; regularly replace sensor probes as needed; Software upgrade: The cloud platform regularly pushes software updates to improve system performance and add new functions, such as introducing new machine learning models or improving alarm strategies.

[0046] Another object of the present invention is to provide a method for detecting harmful factors of animal skins for animal disease source prevention and control based on the above-mentioned animal skin harmful factor detection system for animal disease source prevention and control. The method specifically includes:

[0047] S21: Using the multi-modal sensor detection module, which integrates various detection means such as spectral analysis, fluorescence sensing, and micro-electrochemical sensing, and can respectively detect microorganisms, chemical residues, and other harmful factors on animal skins;

[0048] S22: Using the intelligent analysis and machine learning module, the system fuses the multi-modal data collected by the sensors. Through machine learning algorithms such as random forest and support vector machine (SVM), it classifies, identifies, and analyzes various data, thereby achieving rapid detection and classification of different types of harmful factors in the animal skins. By continuously optimizing the detection model through machine learning, the recognition accuracy of pathogens and harmful factors is gradually improved.

[0049] S23: Internet of Things real-time monitoring and automatic alarm module. Through the Internet of Things technology, the system uploads the detection data of each sensor to the cloud platform in real time for managers and technicians to monitor the status of the animal skins at any time. An automatic alarm function is set. When detecting excessive harmful factors such as the number of microorganisms or the concentration of chemical substances, the system will send alarm notifications in the form of text messages, emails, etc., to remind relevant personnel to take prevention and control measures in a timely manner. Through the cloud management platform, users can remotely access the detection data, generate reports, and adjust the detection strategy as needed.

[0050] Another object of the present invention is to provide a computer device, which includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor executes the steps of the method for detecting harmful factors in animal skins for animal disease prevention and control.

[0051] Another object of the present invention is to provide a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the method for detecting harmful factors in animal skins for animal disease prevention and control.

[0052] Another object of the present invention is to provide an information data processing terminal for implementing the system for detecting harmful factors in animal skins for animal disease prevention and control.

[0053] Combined with the above technical solutions and the solved technical problems, the advantages and positive effects of the technical solution to be protected by the present invention are as follows:

[0054] First, multi-factor comprehensive detection: By integrating multi-modal sensors, the system can efficiently detect various harmful factors in animal skins, including microorganisms, chemical residues, pathogens, etc., solving the problem of insufficient single detection ability in traditional technologies.

[0055] High detection accuracy: Combining spectral analysis, fluorescence labeling, and electrochemical sensing technologies, the detection accuracy is significantly improved. Especially in detecting fine pathogens (such as bacteria and viruses), the sensitivity is increased by 30%-50% compared with traditional methods.

[0056] Automation and Real-time Performance: Through automated sensing and machine learning algorithms, the system reduces manual operation steps and can monitor the status of harmful factors in animal hides in real time, enabling real-time data analysis and automatic alarm, thus greatly enhancing the timeliness of disease source prevention and control.

[0057] Intelligent Data Analysis: Through machine learning algorithms, the system can dynamically optimize the detection model, continuously improve the recognition accuracy, and ensure the reliability and consistency of the detection results.

[0058] This harmful factor detection system for animal hides effectively solves the problems of low efficiency, insufficient accuracy, and inability to monitor in real time in traditional technologies. The new solution realizes real-time and automated detection of harmful factors in animal hides through multi-modal sensing, intelligent data analysis, and Internet of Things technology, helps improve the overall effect of animal disease source prevention and control, and has broad application prospects.

[0059] Second, the existing technical problems solved by the present invention in industrial applications and the significant technical progress.

[0060] Existing technical problems solved:

[0061] 1) Insufficient multi-modal detection ability

[0062] Traditional animal hide detection systems usually rely on a single type of sensor, such as spectroscopic or electrochemical sensors, and are unable to detect multiple types of harmful factors (such as microorganisms, chemical residues, and heavy metals) simultaneously. This limitation results in a restricted detection range and a relatively high missed detection rate, making it difficult to meet the requirements of modern animal disease source prevention and control.

[0063] 2) Low data processing and recognition accuracy

[0064] Existing systems lack the ability to fuse and analyze multi-modal data and usually only process a single signal source, resulting in low recognition efficiency and insufficient accuracy for multiple harmful factors. In addition, detection models lacking intelligent algorithm support often cannot adapt to complex and changing detection environments and are difficult to dynamically optimize recognition performance.

[0065] 3) Low real-time and automation levels

[0066] Traditional systems have a slow feedback speed for detection results and usually rely on manual data analysis before taking measures, unable to achieve real-time monitoring and automatic alarm. This lagging working mode may lead to harmful factors not being discovered in time, thus delaying the prevention and control opportunity.

[0067] 4) Lack of remote monitoring and data management capabilities

[0068] In the prior art, there is usually a lack of an effective connection between the detection system and the management platform. Most data is stored locally and cannot be uploaded in real-time or remotely monitored. Meanwhile, the management and retrieval efficiency of historical data is low, making it difficult to provide a reliable basis for long-term prevention and control strategies.

[0069] Significant technological progress:

[0070] 1) Multimodal fusion detection technology

[0071] By integrating three sensor technologies, namely spectroscopy, fluorescence, and electrochemistry, the present invention significantly expands the detection range and can simultaneously detect harmful factors such as microorganisms, chemical residues, and heavy metals. The multimodal fusion technology makes the detection results more comprehensive and accurate, overcoming the limitations of single detection methods.

[0072] 2) Intelligent analysis and machine learning optimization

[0073] The system uses machine learning algorithms such as random forest (RF) and support vector machine (SVM) to classify and identify multimodal data, and continuously optimizes the detection model through adaptive learning. Compared with traditional methods, machine learning significantly improves the detection accuracy and efficiency, especially showing stronger robustness and adaptability in complex and changing environments.

[0074] 3) Real-time monitoring and automatic alarm function

[0075] Through the Internet of Things technology, the sensor data is uploaded in real-time, and the alarm mechanism is automatically triggered in combination with preset thresholds. When harmful factors are detected to exceed the standard, the system can quickly send out an alarm notification to remind relevant personnel to take prevention and control measures in a timely manner. This function significantly shortens the time from detection to response.

[0076] 4) Cloud data management and remote control

[0077] The detection data is uploaded to the cloud platform through the Internet of Things, supporting long-term storage, instant access, and historical trend analysis. The remote access function of the cloud platform enables users to view the data in real-time and dynamically adjust the detection strategy, thus greatly enhancing the flexibility and management efficiency of the system.

[0078] 5) Overall improvement of prevention and control efficiency and effectiveness

[0079] The present invention combines multimodal detection, intelligent analysis, and real-time monitoring technologies to achieve fast and comprehensive detection and management of harmful factors in animal hides. Compared with the prior art, it significantly reduces the missed detection rate and false alarm rate, while improving the timeliness and accuracy of responding to disease source prevention and control.

[0080] In summary, the present invention has made remarkable technological progress in aspects such as multimodal detection, intelligent data processing, real-time monitoring, and remote management. It has solved problems in the prior art such as insufficient detection range, low recognition accuracy, feedback delay, and inconvenient management, providing an efficient and intelligent solution for the prevention and control of animal disease sources, and having broad promotion value in industrial applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a structural diagram of an animal skin harmful factor detection system for animal disease source prevention and control provided by an embodiment of the present invention;

[0082] Figure 2 It is a flowchart of the integration and implementation method of an animal skin harmful factor detection system for animal disease source prevention and control provided by an embodiment of the present invention;

[0083] Figure 3 It is a flowchart of an animal skin harmful factor detection method for animal disease source prevention and control provided by an embodiment of the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0084] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0085] As Figure 1 shown, an embodiment of the present invention provides an animal skin harmful factor detection system for animal disease source prevention and control, and the system includes:

[0086] A multimodal sensor detection module 1, which integrates various detection means such as spectral analysis, fluorescence sensing, and micro-electrochemical sensing, and can respectively detect microorganisms, chemical residues, and other harmful factors on animal skins;

[0087] An intelligent analysis and machine learning module 2. The system fuses the multimodal data collected by the sensors, and through machine learning algorithms such as random forest and support vector machine (SVM), classifies, identifies, and analyzes various data, so as to realize the rapid detection and classification of different types of harmful factors in the skin; continuously optimizes the detection model through machine learning, and gradually improves the recognition accuracy of pathogens and harmful factors;

[0088] The Internet of Things real-time monitoring and automatic alarm module 3. Through Internet of Things technology, the system uploads the detection data of each sensor to the cloud platform in real time for managers and technicians to monitor the status of hides at any time. An automatic alarm function is set. When harmful factors exceeding the standard, such as the number of microorganisms or the concentration of chemical substances, are detected, the system will send alarm notifications in the form of text messages, emails, etc., to remind relevant personnel to take prevention and control measures in time. Through the cloud management platform, users can remotely access the detection data, generate reports, and adjust the detection strategy as needed.

[0089] The specific solution of the multimodal sensor detection module includes:

[0090] Spectral sensing: Using near-infrared or mid-infrared spectra to scan the surface of the hide to detect whether there are harmful chemical residues or pollutants on the hide;

[0091] Fluorescence sensing: By binding specific fluorescent labeling substances to harmful microorganisms such as bacteria and viruses, and using fluorescence signals for detection to improve the sensitivity of pathogen detection;

[0092] Electrochemical sensor: Detect specific harmful factors on the surface of the hide, such as heavy metals and the concentration of chemical agents, through electrochemical analysis technology.

[0093] The specific technical solution of the intelligent analysis and machine learning module includes:

[0094] Training data set: The system uses a known data set of harmful factors of animal hides for model training so that it can identify different types of harmful factors and classify them;

[0095] Adaptive learning: As data accumulates during actual use, the system automatically optimizes the algorithm to gradually improve the detection accuracy and efficiency.

[0096] The implementation method of the intelligent analysis and machine learning module specifically includes:

[0097] (1) Multi-source data fusion

[0098] Data acquisition: Collect and uniformly transmit the multimodal data generated by spectral, fluorescence, and electrochemical sensors to the data processing unit;

[0099] Feature extraction: Extract features from the output data of each sensor, such as absorption peaks in spectral data, fluorescence signal intensity, potential changes in electrochemical signals, etc., to form feature vectors;

[0100] Data fusion algorithm: Use a weighted fusion algorithm to weight and fuse the data from different sensors, and through reasonable weight allocation, ensure that the detection results of different sensors are comprehensively analyzed;

[0101] (2) Machine learning implementation

[0102] Model training: Use the pre-prepared database of harmful factors for animal skins, which contains the spectral, fluorescence, and electrochemical characteristics of various known harmful factors; use this data for training, and adopt algorithms such as random forest RF and support vector machine SVM to build a classification model;

[0103] Model deployment: The trained model is deployed into the system. By processing the data collected in real time by the sensor, it quickly judges whether there are harmful factors on the skin and classifies them;

[0104] Adaptive learning: The system continuously accumulates new detection data and uses online learning algorithms such as incremental SVM to update the model in real time, improving the classification accuracy and response speed of the model;

[0105] The implementation method of the Internet of Things real-time monitoring and automatic alarm module specifically includes:

[0106] (1) Real-time data collection and transmission

[0107] Sensor data collection: The system, through the integrated Internet of Things sensing module, uploads the real-time data detected by various sensors to the cloud platform for unified processing through wireless communication such as Wi-Fi, 4G / 5G, LoRa, etc.;

[0108] Data transmission protocol: Adopt standard Internet of Things data transmission protocols such as MQTT or CoAP to ensure low-latency and high-reliability data transmission;

[0109] (2) Automatic alarm function implementation

[0110] Threshold setting: The system automatically triggers an alarm when the threshold, such as the concentration of microorganisms or chemical pollutants, is exceeded according to the pre-set threshold;

[0111] Alarm notification: When harmful factors exceeding the standard are detected, the system automatically generates an alarm message and notifies the manager through forms such as text messages, emails, or APP push; the alarm message includes detailed data such as the detection location, the type and concentration of the exceeded standard;

[0112] (3) Cloud platform and remote management

[0113] Cloud data storage: Through Internet of Things technology, all detection data is uploaded to the cloud platform in real time for long-term storage and management, facilitating the retrieval of historical data for analysis at any time;

[0114] Remote access: Users can remotely access the cloud data through a Web interface or a mobile application, monitor the status of animal skins, and generate a detection report;

[0115] Dynamic policy adjustment: Based on real-time data and detection results, users can remotely adjust the detection policy through the cloud platform, such as changing the detection frequency of sensors, threshold settings, etc., to improve the prevention and control efficiency.

[0116] As Figure 2 shown, the integration and implementation method of the animal skin harmful factor detection system for animal disease source prevention and control is as follows:

[0117] S1: System integration. Integrate multi-modal sensors, data processing units, machine learning modules, and Internet of Things monitoring modules into an embedded system. The hardware part adopts a modular design for easy expansion and maintenance. The software part adopts a layered architecture, and the data acquisition, analysis, storage, and monitoring modules are relatively independent to ensure the scalability and stability of the system.

[0118] S2: On-site deployment and debugging. On-site deployment: The system can be portably installed at detection points or on factory production lines for skin detection through mobile devices or fixed detection stations. System debugging: Before the actual application of the system, it is necessary to debug the system according to specific scenarios, including sensor calibration, model parameter adjustment, and network connection of Internet of Things devices.

[0119] S3: Long-term maintenance and upgrade. Sensor maintenance: Regularly maintain and calibrate spectral, fluorescence, and electrochemical sensors to ensure detection accuracy. Replace sensor probes regularly as needed. Software upgrade: The cloud platform regularly pushes software updates to improve system performance and add new functions, such as introducing new machine learning models or improving alarm strategies.

[0120] As Figure 3 shown, the embodiment of the present invention provides an animal skin harmful factor detection method for animal disease source prevention and control based on the animal skin harmful factor detection system for animal disease source prevention and control. The method specifically includes:

[0121] S21: Use the multi-modal sensor detection module, which integrates various detection means such as spectral analysis, fluorescence sensing, and micro-electrochemical sensing, and can respectively detect microorganisms, chemical residues, and other harmful factors on animal skins.

[0122] S22: Use the intelligent analysis and machine learning module. The system fuses the multi-modal data collected by the sensors and classifies, identifies, and analyzes various data through machine learning algorithms such as random forest and support vector machine (SVM), so as to achieve rapid detection and classification of different types of harmful factors in the skin. Continuously optimize the detection model through machine learning to gradually improve the recognition accuracy of pathogens and harmful factors.

[0123] S23: Internet of Things real-time monitoring and automatic alarm module. Through Internet of Things technology, the system uploads the detection data of each sensor to the cloud platform in real time for managers and technicians to monitor the status of hides at any time. An automatic alarm function is set. When detecting excessive harmful factors, such as the number of microorganisms or the concentration of chemical substances, the system will send alarm notifications in the form of text messages, emails, etc., to remind relevant personnel to take prevention and control measures in a timely manner. Through the cloud management platform, users can remotely access the detection data, generate reports, and adjust the detection strategy as needed.

[0124] An embodiment of the present invention provides a computer device, which includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor is caused to execute the steps of the method for detecting harmful factors of animal hides for animal disease source prevention and control.

[0125] An embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor is caused to execute the steps of the method for detecting harmful factors of animal hides for animal disease source prevention and control.

[0126] An embodiment of the present invention provides an information data processing terminal for implementing the system for detecting harmful factors of animal hides for animal disease source prevention and control.

[0127] Signal and data processing process of the system for detecting harmful factors of animal hides:

[0128] The multimodal sensor detection module is responsible for collecting information on harmful factors on the surface of animal hides. The spectral sensor extracts chemical residue information on the hide by scanning with near-infrared or mid-infrared light; the fluorescence sensor uses specific fluorescent labels to bind to pathogens to detect microorganisms on the hide, such as bacteria and viruses; the electrochemical sensor obtains relevant information on the concentration of heavy metals or chemical agents by measuring changes in potential, current or resistance. Each sensor transmits the real-time collected data to the data processing unit through a dedicated interface.

[0129] The signal data collected by the sensors is diverse and complex. The system first extracts features from different types of sensor signals. For example, the absorption peak position information is extracted from the spectral data, the intensity and wavelength features are extracted from the fluorescence signal, and the potential change and the corresponding concentration value are extracted from the electrochemical signal. These features are converted into feature vectors for subsequent data fusion and analysis.

[0130] The system uses a weighted fusion algorithm to fuse multi-modal data from spectral, fluorescence, and electrochemical sensors. By assigning reasonable weights to different sensors, it ensures that the data fusion result can reflect the comprehensive state of various harmful factors on the hide. The weighted fusion algorithm dynamically adjusts the weights according to indicators such as the detection accuracy and sensitivity of the sensors, making the fused feature data more accurate.

[0131] The intelligent analysis module classifies and identifies the feature data through machine learning methods. The system uses a known dataset of harmful factors on animal hides to train random forest (RF) and support vector machine (SVM) models, establishing a benchmark model for classification. After training, the models are deployed into the detection system to quickly classify the real-time collected data, determine whether there are harmful factors on the hide, and classify their types.

[0132] The Internet of Things real-time monitoring module transmits the real-time data collected by each sensor to the cloud platform through wireless communication technologies (such as Wi-Fi, 4G / 5G, LoRa). The data transmission uses the MQTT or CoAP protocol to ensure low-latency and high-reliability transmission performance. The upload of real-time data facilitates the management personnel to monitor the state of the hide at any time.

[0133] The system sets a detection threshold. When the sensor detects excessive harmful factors (such as microbial concentration or chemical pollutant concentration), it automatically triggers an alarm. The alarm information is pushed to the management personnel through text messages, emails, or mobile applications, and the content includes detailed information such as the detection location, the types and concentrations of the excessive factors, helping the relevant personnel to take prevention and control measures in time.

[0134] All detection data is uploaded to the cloud platform in real-time for long-term storage and management. The cloud platform supports historical backtracking and statistical analysis of the data, facilitating users to retrieve past detection data for trend analysis or comparative research. The cloud storage uses an encryption method to ensure the security and privacy of the data.

[0135] Through the cloud management platform, users can remotely access real-time data and historical detection data, generate detection reports, and dynamically adjust the detection strategy according to the detection results. For example, users can remotely change parameters such as the detection frequency and alarm threshold of the sensors to adapt to different prevention and control requirements. The dynamic adjustment function greatly improves the flexibility and prevention and control efficiency of the system.

[0136] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated designed hardware. Those of ordinary skill in the art can understand that the above devices and methods can be implemented using computer-executable instructions and / or included in processor control code, such as provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits of programmable hardware devices such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above hardware circuits and software such as firmware.

[0137] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be covered by the protection scope of the present invention.

Claims

1. An animal skin harmful factor detection system for animal disease source prevention and control, characterized in that, The system includes: A multimodal sensor detection module, which integrates three detection methods: spectral sensing, fluorescence sensing, and electrochemical sensors; An intelligent analysis and machine learning module. The system fuses the multimodal data collected by the sensors and classifies, identifies, and analyzes various data through machine learning algorithms. The machine learning algorithms include Random Forest (RF) and Support Vector Machine (SVM); An Internet of Things (IoT) real-time monitoring and automatic alarm module. The system uploads the detection data of each sensor to the cloud platform in real time through IoT technology; an automatic alarm function is set; through the cloud management platform, users can remotely access the detection data, generate reports, and adjust the detection strategy as needed; The specific solution of the multimodal sensor detection module includes: Spectral sensing: Scanning the surface of the hide using near-infrared or mid-infrared spectroscopy to detect whether there are harmful chemical residues or pollutants on the hide; Fluorescence sensing: Detecting the fluorescence signal by binding a specific fluorescent labeling substance to harmful microorganisms, where the harmful microorganisms include bacteria and viruses; Electrochemical sensors: Detecting specific harmful factors on the surface of the hide through electrochemical analysis technology, where the specific harmful factors include heavy metals and chemical agent concentrations.

2. The animal skin harmful factor detection system for preventing and controlling animal diseases as described in claim 1, characterized in that, The intelligent analysis and machine learning module specifically includes: Training data set: The system uses a known data set of harmful factors of animal hides for model training to enable it to identify and classify different types of harmful factors; Adaptive learning: As data accumulates during actual use, the system automatically optimizes the algorithm.

3. The animal skin harmful factor detection system for preventing and controlling animal diseases as described in claim 1, characterized in that, The implementation method of the intelligent analysis and machine learning module specifically includes: (1) Multi-source data fusion Data acquisition: Collect and uniformly transmit the multimodal data generated by spectral, fluorescence, and electrochemical sensors to the data processing unit; Feature extraction: Extract features from the output data of each sensor, such as absorption peaks in spectral data, fluorescence signal intensity, and potential changes in electrochemical signals, to form feature vectors; Data fusion algorithm: Use a weighted fusion algorithm to weightedly fuse data from different sensors; (2) Machine learning implementation Model training: Use a pre-prepared database of harmful factors of animal hides, which contains the spectral, fluorescence, and electrochemical characteristics of various known harmful factors; Use this data for training and build a classification model using the Random Forest (RF) and Support Vector Machine (SVM) algorithms; Model deployment: The trained model is deployed into the system. By processing the data collected by the sensors in real time, it quickly determines whether there are harmful factors on the hide and classifies them; Adaptive learning: The system continuously accumulates new detection data and uses an online learning algorithm to update the model in real time. The online learning algorithm includes incremental SVM.

4. The animal skin harmful factor detection system for preventing and controlling animal diseases as described in claim 1, wherein The implementation method of the IoT real-time monitoring and automatic alarm module specifically includes: (1) Real-time data collection and transmission Sensor data collection: The system uploads the real-time data detected by various sensors to the cloud platform for unified processing through the integrated IoT sensing module via wireless communication. The wireless communication includes Wi-Fi, 4G / 5G, and LoRa; Data transmission protocol: Adopt standard Internet of Things data transmission protocols, including MQTT or CoAP; (2) Implementation of automatic alarm function Threshold setting: The system triggers an alarm automatically according to pre-set thresholds, which include concentrations of bacteria, viruses, heavy metals, and chemical agents when they exceed the standard; Alarm notification: When harmful factors exceeding the standard are detected, the system automatically generates alarm information and notifies the administrator in the form of SMS, email, or APP push; The alarm information includes the detection location, the type of exceeded standard, and detailed concentration data; (3) Cloud platform and remote management Cloud data storage: Through Internet of Things technology, all detection data is uploaded to the cloud platform in real time for long-term storage and management; Remote access: Users can remotely access cloud data through a Web interface or a mobile application, monitor the status of animal skins, and generate detection reports; Dynamic policy adjustment: According to real-time data and detection results, users can remotely adjust detection policies through the cloud platform, including changing the detection frequency of sensors and threshold setting.

5. The animal skin harmful factor detection system for preventing and controlling animal diseases as described in claim 1, characterized in that, The integration and implementation method of the harmful factor detection system for animal skins used in animal disease source prevention and control is as follows: S1: System integration, integrating multi-modal sensors, a data processing unit, a machine learning module, and an Internet of Things monitoring module into an embedded system; The hardware part adopts a modular design for easy expansion and maintenance; The software part adopts a layered architecture, and each module of data acquisition, analysis, storage, and monitoring is relatively independent; S2: On-site deployment and debugging, On-site deployment: The system can be portably installed at detection points or on factory production lines for skin detection through mobile devices or fixed detection stations; System debugging: Before the actual application of the system, the system needs to be debugged according to specific scenarios, including sensor calibration, model parameter adjustment, and network connection of Internet of Things devices; S3: Long-term maintenance and upgrade, Sensor maintenance: Regularly maintain and calibrate spectral, fluorescence, and electrochemical sensors to ensure detection accuracy; Replace sensor probes regularly as needed; Software upgrade: The cloud platform regularly pushes software updates, including introducing new machine learning models or improving alarm strategies.

6. A method for detecting harmful factors of animal skins used in animal disease source prevention and control as described in claims 1-5 A method for detecting harmful factors of animal hides for animal disease source prevention and control of a system, characterized in that, The method specifically includes: S21: Using a multi-modal sensor detection module, which integrates three detection methods: spectral sensing, fluorescence sensing, and electrochemical sensors; S22: Using an intelligent analysis and machine learning module, the system fuses multi-modal data collected by sensors and classifies, identifies, and analyzes various data through machine learning algorithms, which include random forest RF and support vector machine SVM; S23: Internet of Things real-time monitoring and automatic alarm module. The system uploads the detection data of each sensor to the cloud platform in real time through Internet of Things technology. An automatic alarm function is set. When harmful factors exceeding the standard are detected, the harmful factors include bacteria, viruses, heavy metals, and chemical agent concentrations. The system will send alarm notifications in the form of text messages and emails to remind relevant personnel to take prevention and control measures in a timely manner. Through the cloud management platform, users can remotely access the detection data, generate reports, and adjust the detection strategy as needed.

7. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor executes the steps of the method for detecting harmful factors on animal skins for preventing and controlling animal diseases as claimed in claim 6.

8. A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the method for detecting harmful factors on animal skins for preventing and controlling animal diseases as claimed in claim 6.

9. An information data processing terminal, characterized in that, The information data processing terminal is used to process the method for detecting harmful factors on animal skins for preventing and controlling animal diseases as claimed in claim 6.

Images