A rapid detection system for porcine enteric coronavirus

Abstract

The present invention relates to the technical field of pig detection, and specifically discloses a rapid detection system for porcine enteric coronavirus, which includes a pig data extraction module, a pig data analysis module, a pig data judgment module, a pen data collection module, and a pen data analysis module; by setting a monitoring period, based on the monitoring instruments arranged in the target pen, the basic parameters of each pig in the target pen on each monitoring day during the corresponding monitoring period are extracted; the state abnormality index of each pig in the target pen on each monitoring day is comprehensively evaluated; and the pen environment data of each pig in the target pen on each monitoring day during the corresponding monitoring period is collected by an automatic sampling device; the pen environment data of each pig in the target pen on each monitoring day during the corresponding monitoring period is comprehensively analyzed to analyze the infection probability of each pig in the target pen on each monitoring day, and then the diseased pigs in the target pen are screened out, so as to timely grasp the health status of each pig and timely discover the abnormal state of the pigs.

Description

Technical Field

[0001] The present invention belongs to the technical field of pig detection, and relates to a rapid detection system for porcine enteric coronavirus. Background Art

[0002] Porcine Enteric Coronavirus (PEC) is of great significance in the pig industry. This virus can cause severe intestinal diseases, affecting the growth and production performance of pigs. At the same time, porcine enteric coronavirus mainly includes Porcine Epidemic Diarrhea Virus (PEDV), Transmissible Gastroenteritis Virus of Swine (TGEV), and Porcine Delta Coronavirus (PDCoV). These viruses can cause diarrhea, vomiting, dehydration, and high mortality in pigs, especially for piglets and sows.

[0003] Current technologies for the detection of porcine enteric coronavirus mainly rely on chemical assay methods, such as Reverse Transcription Polymerase Chain Reaction (RT-PCR) and Enzyme-Linked Immunosorbent Assay (ELISA). Chemical assay methods usually require complex experimental equipment and a professional laboratory environment. This means that in areas with limited resources or small farms, it may be practically difficult to conduct these detections.

[0004] Secondly, chemical assay methods often take a long time to obtain results. For RT-PCR, steps such as sample processing, RNA extraction, reverse transcription, and amplification all require time, which may lead to delays in test results. Such delays may affect the timeliness and effectiveness of controlling the epidemic during a virus outbreak. Summary of the Invention

[0005] In view of the above problems existing in the prior art, the present invention provides a rapid detection system for porcine enteric coronavirus to solve the above technical problems.

[0006] To achieve the above and other purposes, the technical solution adopted by the present invention is as follows:

[0007] The present invention provides a rapid detection system for porcine enteric coronavirus, including a pig data extraction module, a pig data analysis module, a pig data judgment module, a pen data collection module, and a pen data analysis module. The above-mentioned modules are connected by wired and / or wireless connection methods to achieve data transmission between modules;

[0008] Pig data extraction module: used to set the monitoring period, and extract the basic parameters of each pig in the target pen during each monitoring day within the monitoring period according to the monitoring instruments arranged in the target pen;

[0009] Live pig data analysis module: Analyze the basic parameters of each live pig in the target pen for each monitoring day, and comprehensively evaluate to obtain the status anomaly index of each live pig in the target pen for each monitoring day;

[0010] Live pig data judgment module: Analyze and judge based on the status anomaly index of each live pig in the target pen for each monitoring day;

[0011] Pen data collection module: Collect the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period according to the automatic sampling equipment;

[0012] Pen data analysis module: Comprehensively analyze the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period, analyze the infection probability of each live pig in the target pen for each monitoring day, and then screen out the diseased live pigs in the target pen.

[0013] Exemplarily, the basic parameters of each live pig in the target pen for each monitoring day during the monitoring period include the total diarrhea amount, the average individual temperature, and the total food intake.

[0014] Exemplarily, the process of extracting the total diarrhea amount of each live pig in the target pen for each monitoring day during the monitoring period is as follows:

[0015] According to the monitoring instruments arranged in the target pen, obtain the monitoring videos of the target pen for each monitoring day during the monitoring period; preprocess the monitoring videos, and simultaneously use the background modeling method to separate the background in the target pen and each live pig in the target pen;

[0016] Use the object detection algorithm to detect each live pig in the monitoring video, and use the object tracking algorithm to track each live pig in the target pen to obtain the positions of each live pig in the target pen at each monitoring time point for each monitoring day;

[0017] Use image processing and computer vision techniques to identify the monitoring time points of each diarrhea event of each live pig in the target pen for each monitoring day , ZL is the frame rate of the monitoring video of the target pen for each monitoring day during the monitoring period, is the number of frames of the i-th diarrhea event of the c-th live pig in the target pen on the j-th monitoring day in the corresponding monitoring video, c is the number of each live pig in the target pen, c = 1, 2,... C; j is the number of each monitoring day, j = 1, 2,... J, i is the number of each diarrhea event, i = 1, 2,... ;

[0018] Mark the monitoring time points of each diarrhea event corresponding to each live pig in the target pen on each monitoring day in the monitoring video, and overlap them with the positions of each live pig in the target pen at each monitoring time point within each monitoring day, so as to obtain the diarrhea areas of each diarrhea event corresponding to each live pig in the target pen on each monitoring day;

[0019] Using image segmentation technology, obtain the total number of pixels in the diarrhea areas corresponding to each diarrhea event of each live pig in the target pen on each monitoring day ;

[0020] Collect the image pixel width of the monitoring screen in the monitoring video and the image pixel height ; Thus, calculate the predicted diarrhea area of the diarrhea area corresponding to each diarrhea event of each live pig in the target pen on each monitoring day , and respectively represent the actual width and actual height corresponding to the camera view;

[0021] Furthermore, through the formula , calculate the predicted diarrhea liquid thickness of the diarrhea area corresponding to the i-th diarrhea event of the c-th live pig in the target pen on the j-th monitoring day , is the set liquid characteristic constant;

[0022] Finally, obtain the total diarrhea volume of each live pig in the target pen on each monitoring day during the monitoring period .

[0023] Exemplarily, extract the individual temperature means of each live pig in the target pen on each monitoring day during the monitoring period, and the specific process is as follows:

[0024] According to the monitoring instruments arranged in the target pen, collect the infrared temperatures of each live pig in the target pen at each monitoring time point within each monitoring day; synchronously collect the outdoor temperature, outdoor humidity, and the indoor temperature and indoor humidity of the pen at each monitoring time point corresponding to each monitoring day;

[0025] Based on the positions of each live pig in the target pen at each monitoring time point within each monitoring day and the positions of the infrared devices arranged in the target pen, obtain the position angle coefficients and position angle functions of the positions of each live pig in the target pen at each monitoring time point corresponding to the positions of the infrared devices;

[0026] Perform temperature calibration operations on the infrared temperatures of each live pig in the target pen at each monitoring time point within each monitoring day to obtain the calibrated infrared temperatures of each live pig in the target pen at each monitoring time point within each monitoring day , respectively represent predefined calibration coefficients, d is the number of each monitoring time point, d = 1, 2,... D;

[0027] Calculate the individual temperature mean of each live pig in the target pen for each monitoring day during the corresponding monitoring period , are respectively the indoor temperature, outdoor temperature, indoor humidity and outdoor humidity of the pen corresponding to the d-th monitoring time point on the j-th monitoring day; are respectively the position angle coefficient and position angle function corresponding to the position of the infrared device at the position of the c-th live pig in the target pen for the d-th monitoring time point within the j-th monitoring day, are respectively the indoor-outdoor temperature difference influence coefficient and humidity difference influence coefficient corresponding to the d-th monitoring time point on the j-th monitoring day, and D is the total number of monitoring time points.

[0028] Exemplarily, obtaining the position angle coefficient and position angle function corresponding to the position of the infrared device at the position of each live pig in the target pen for each monitoring time point within each monitoring day includes:

[0029] Obtain the position coordinates of the infrared devices deployed in the target pen and the position coordinates of each live pig for each monitoring time point within each monitoring day ;

[0030] Calculate the distance between the position of each live pig in the target pen and the position of the infrared device for each monitoring time point within each monitoring day ;

[0031] Calculate the relative pitch angle and relative yaw angle ;

[0032] Furthermore, the position angle coefficient corresponding to the position of the infrared device at the position of each live pig in the target pen for each monitoring time point within each monitoring day is expressed as , where k4, k5 and k6 are all predefined calculation experience coefficients;

[0033] Finally, obtain the position angle function , corresponding to the position of the infrared device at the position of each live pig in the target pen for each monitoring time point within each monitoring day, where

[0034] Exemplarily, the total food intake of each live pig in the target pen for each monitoring day during the corresponding monitoring period is extracted, and the specific process is as follows:

[0035] Identify the feeding time points and the end feeding time points of each pig's each feeding during each monitoring day through behavior recognition algorithms and target tracking algorithms;

[0036] And based on image recognition technology, obtain the estimated total feed amount in the feed trough at the feeding time point of each pig's each feeding during each monitoring day and the estimated remaining total feed amount in the feed trough at the corresponding end feeding time point, perform a subtraction calculation on them to obtain the feeding amount of each pig's each feeding during each monitoring day;

[0037] Perform a summation operation on the feeding amounts of each pig's each feeding during each monitoring day to obtain the total feeding amount of each pig in the target pen during each monitoring day of the monitoring period.

[0038] Exemplarily, comprehensively evaluate to obtain the status anomaly index of each pig in the target pen corresponding to each monitoring day. The specific evaluation process is as follows:

[0039] Obtain the total diarrhea amount of each pig in the target pen during each monitoring day of the monitoring period and the individual temperature average value as well as the total feeding amount ;

[0040] Calculate the possible infection index of each pig corresponding to each monitoring day , where are the total diarrhea amount, the individual temperature average value, and the total feeding amount of the c-th pig corresponding to the (j - 1)-th monitoring day respectively, where are the total diarrhea amount, the individual temperature average value, and the total feeding amount of the c-th pig corresponding to the (j + 1)-th monitoring day respectively, where are the set weight coefficients respectively, and C is the total number of monitoring days;

[0041] Comprehensively evaluate to obtain the status anomaly index of each pig in the target pen corresponding to each monitoring day , where is the set status anomaly permission threshold.

[0042] Exemplarily, the pen environment data of each pig in the target pen during each monitoring day of the monitoring period includes ventilation density, feed quality, and pig density.

[0043] Exemplarily, analyze the infection probability of each pig in the target pen corresponding to each monitoring day. The specific analysis logic is as follows:

[0044] Arbitrarily select a monitoring day from the pen environment data of each pig in the target pen during each monitoring day of the monitoring period as the target analysis monitoring day, and obtain the ventilation density V, feed quality F, and pig density U of the target analysis monitoring day;

[0045] Let G be the state value of a live pig infected with the coronavirus, where G takes the value of 1 or 0, 1 indicates infection, and 0 indicates non-infection. Thus, a Bayesian network corresponding to each live pig on the target analysis and monitoring day is constructed. ;

[0046] The prior probability of a live pig being infected with the coronavirus on the target analysis and monitoring day is , then the prior probability of a live pig not being infected with the coronavirus is ;

[0047] Thus, the joint probability of a live pig being infected with the coronavirus on the target analysis and monitoring day is calculated ; Calculate the joint probability that a live pig is not infected with the coronavirus on the target analysis and monitoring day ;

[0048] Calculate the probability that ventilation conditions, feed quality, and pig density occur simultaneously under all possible infection states for each live pig on the target analysis and monitoring day = × + × ;

[0049] Comprehensively calculate the infection probability of each live pig corresponding to ventilation density, feed quality, and pig density on the target analysis and monitoring day ;

[0050] In this analysis method, the infection probability of each live pig in the target pen corresponding to each monitoring day is further obtained.

[0051] Exemplarily, the screening logic for screening out diseased live pigs in the target pen is as follows:

[0052] Weights Q1 and Q2 are respectively assigned to the state anomaly index and the infection probability;

[0053] Thus, the comprehensive index value of each live pig in the target pen is calculated , is the infection probability of the c-th live pig in the target pen corresponding to the j-th monitoring day;

[0054] The comprehensive index value of each live pig in the target pen is compared with the threshold . If , it is determined that the c-th live pig has the coronavirus.

[0055] As described above, a rapid detection system for porcine enteric coronavirus provided by the present invention has at least the following beneficial effects:

[0056] (1)The rapid detection system for porcine enteric coronavirus provided by the present invention can grasp the health status of each live pig in real time by setting a monitoring period and using monitoring instruments to obtain the basic parameters of live pigs. By comprehensively evaluating these data, the abnormal status of live pigs can be detected in a timely manner. Compared with traditional manual inspections, it is more timely and accurate, and can detect potential problems at the early stage of the disease, so as to take effective measures for intervention and avoid the spread of the disease.

[0057] (2)The present invention helps to accurately screen out diseased live pigs and reduce misjudgment and missed judgment. Traditional disease screening methods often rely on experience and intuitive judgment, and are prone to misjudgment and missed judgment. By means of data-driven, scientific analysis can be carried out based on objective data to screen out truly diseased live pigs. This can not only improve the accuracy of disease prevention and control, but also avoid unnecessary isolation and treatment, saving resources and costs.

[0058] (3)The present invention sets a monitoring period, uses monitoring instruments to obtain the basic parameters of live pigs, comprehensively evaluates the status anomaly index, and combines the pen environment data for analysis to screen out diseased live pigs. This method not only improves the efficiency and accuracy of disease monitoring and prevention and control, but also has important practical significance and popularization value. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0060] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the claims of the present invention, they should all belong to the protection scope of the present invention.

[0062] Embodiment 1

[0063] Please refer to Figure 1 As shown, a rapid detection system for porcine enteric coronavirus includes a live pig data extraction module, a live pig data analysis module, a live pig data judgment module, a pen data collection module, and a pen data analysis module. The above-mentioned modules are connected by wired and / or wireless connection methods to realize data transmission between each module;

[0064] Live pig data extraction module: used to set the monitoring period, and based on the monitoring instruments arranged in the target pen, extract the basic parameters of each live pig in the target pen for each monitoring day during the monitoring period;

[0065] In pig farming, there is a kind of breeding facility called a small pen, which is specially used for small-scale penning. This kind of small pen usually has an area of about 10 square meters, is equipped with a feeding trough with a length of 1.5 - 2.0 meters, and has a sunny exercise field. Each pen can raise 3 - 5 weaned sows. This design aims to provide a more refined breeding environment to meet the management needs of specific pig groups, such as sows with estrus return and dystocia, which need to be raised and managed in small pens.

[0066] In a possible design, the basic parameters of each live pig in the target pen for each monitoring day during the monitoring period include the total diarrhea amount, the average individual temperature, and the total feeding amount.

[0067] In a possible design, the process of extracting the total diarrhea amount of each live pig in the target pen for each monitoring day during the monitoring period is as follows:

[0068] Based on the monitoring instruments arranged in the target pen, obtain the monitoring videos of the target pen for each monitoring day during the monitoring period; preprocess the monitoring videos, and simultaneously use the background modeling method to separate the background in the target pen and each live pig in the target pen;

[0069] Use the object detection algorithm to detect each live pig in the monitoring video, and use the object tracking algorithm to track each live pig in the target pen to obtain the positions of each live pig in the target pen at each monitoring time point within each monitoring day;

[0070] Use image processing and computer vision techniques to identify the monitoring time points of each diarrhea event of each live pig in the target pen for each monitoring day , ZL is the frame rate of the monitoring video of the target pen for each monitoring day during the monitoring period, is the number of frames of the i-th diarrhea event of the c-th live pig in the target pen on the j-th monitoring day in the corresponding monitoring video. c is the number of each live pig in the target pen, c = 1, 2,... C; j is the number of each monitoring day, j = 1, 2,... J, and i is the number of each diarrhea event, i = 1, 2,... ;

[0071] Mark the monitoring time points of each diarrhea event of each live pig in the target pen in the monitoring video, and overlap them with the positions of each live pig in the target pen at each monitoring time point within each monitoring day, so as to obtain the diarrhea area of each diarrhea event of each live pig in the target pen for each monitoring day;

[0072] Using image segmentation technology, obtain the total number of pixels in the diarrhea areas corresponding to each diarrhea event for each live pig in the target pen on each monitoring day. ;

[0073] Collect the image pixel width of the monitoring screen in the monitoring video. and the image pixel height. ; From this, calculate the predicted diarrhea area of the diarrhea areas corresponding to each diarrhea event for each live pig in the target pen on each monitoring day. , and respectively represent the actual width and actual height corresponding to the camera's viewing angle;

[0074] Furthermore, through the formula , calculate the predicted diarrhea liquid thickness of the diarrhea areas corresponding to the i-th diarrhea event for the c-th live pig in the target pen on the j-th monitoring day. , is a set liquid characteristic constant;

[0075] Finally, obtain the total diarrhea amount of each live pig in the target pen on each monitoring day during the monitoring period. .

[0076] In a possible design, extract the average individual temperature of each live pig in the target pen on each monitoring day during the monitoring period. The specific process is as follows:

[0077] Based on the monitoring instruments arranged in the target pen, collect the infrared temperatures of each live pig in the target pen at each monitoring time point on each monitoring day; simultaneously collect the outdoor temperature, outdoor humidity, and the indoor temperature and indoor humidity of the pen at each monitoring time point corresponding to each monitoring day;

[0078] Based on the positions of each live pig in the target pen at each monitoring time point on each monitoring day and the positions of the infrared devices arranged in the target pen, obtain the position angle coefficient and position angle function of the positions of each live pig in the target pen at each monitoring time point on each monitoring day corresponding to the positions of the infrared devices;

[0079] The position angle coefficient is used to adjust the amplitude of the position angle function so that the correction value is closer to the actual measurement error.

[0080] The position angle function describes the measurement error of the infrared device at different positions and angles.

[0081] The position angle coefficient is a specific value used to quantify the influence of the position and angle of the infrared device on the temperature measurement result. It is usually a coefficient calculated through experimental data or empirical formulas and is used to correct the measurement result. It specifically reflects the influence of the distance and angle between the infrared device and the live pig on the measured temperature.

[0082] The position - angle function is a mathematical expression formula used to describe the influence relationship of the position and angle of an infrared device on the temperature measurement result.

[0083] The position - angle coefficient is a specific output value of the position - angle function. The position - angle function describes the influence of position and angle on the measurement result through a mathematical expression, and the position - angle coefficient is the calculation result of this function under specific position and angle conditions.

[0084] For the infrared temperatures of each live pig in the target pigsty corresponding to each monitoring time point within each monitoring day perform temperature calibration operations to obtain the calibrated infrared temperatures of each live pig in the target pigsty corresponding to each monitoring time point within each monitoring day , respectively represent predefined calibration coefficients, d is the number of each monitoring time point, d = 1, 2,... D;

[0085] Calculate the individual temperature means of each live pig in the target pigsty corresponding to each monitoring day within the monitoring period , are respectively the indoor temperature, outdoor temperature, indoor humidity, and outdoor humidity of the pigsty corresponding to the d - th monitoring time point on the j - th monitoring day; are respectively the position - angle coefficient and the position - angle function corresponding to the position of the infrared device where the position of the c - th live pig in the target pigsty is located at the d - th monitoring time point within the j - th monitoring day, are respectively the indoor - outdoor temperature difference influence coefficient and the humidity difference influence coefficient corresponding to the d - th monitoring time point on the j - th monitoring day, and D is the total number of monitoring time points.

[0086] As homeothermic animals, the body temperature regulation mechanism of live pigs is significantly affected by the surrounding environment. When the environmental temperature is high, live pigs regulate their body temperature by increasing skin blood flow and evaporative heat dissipation, which may cause their surface temperature to rise; conversely, when the environmental temperature is low, live pigs reduce skin blood flow and increase metabolic heat production to maintain body temperature, which may lead to a decrease in their surface temperature. In addition, air humidity also affects the heat dissipation efficiency of live pigs. In a high - humidity environment, the evaporative heat dissipation efficiency decreases, and the body temperature regulation ability of live pigs is limited, which may cause the body temperature to rise. Therefore, considering the influence of environmental temperature and humidity in the calculation formula is a practical consideration based on the physiology and thermodynamics principles of live pigs. By using the calculation formula to reflect the influence of these external factors on the body temperature of live pigs, the actual body temperature can be estimated more accurately.

[0087] In a possible design, obtaining the position - angle coefficient and the position - angle function corresponding to the position of the infrared device where the position of each live pig in the target pigsty is located at each monitoring time point within each monitoring day includes:

[0088] Obtain the position coordinates of the infrared devices deployed in the target pen and the position coordinates of each live pig corresponding to each monitoring time point within each monitoring day ;

[0089] Calculate the distances between the positions of each live pig corresponding to each monitoring time point within each monitoring day in the target pen and the positions of the infrared devices ;

[0090] Calculate the relative pitch angles between the positions of each live pig corresponding to each monitoring time point within each monitoring day in the target pen and the positions of the infrared devices and the relative yaw angles ;

[0091] Furthermore, the position angle coefficients of the positions of each live pig corresponding to each monitoring time point within each monitoring day in the target pen with respect to the positions of the infrared devices are expressed as , where k4, k5, and k6 are all predefined calculation experience coefficients;

[0092] Finally, obtain the position angle functions of the positions of each live pig corresponding to each monitoring time point within each monitoring day in the target pen with respect to the positions of the infrared devices , which are respectively the fitting experience coefficients

[0093] The basic constant term represents the basic correction value, the distance term reflects the linear influence of the distance on the measurement result, and the angle term describes the influence of the pitch angle and the yaw angle

[0094] In a possible design, the total intake of each live pig corresponding to each monitoring day during the monitoring period in the target pen is extracted, and the specific process is as follows:

[0095] Identify the feeding start time points and feeding end time points of each feeding of each live pig corresponding to each monitoring day through the behavior recognition algorithm and the target tracking algorithm;

[0096] And based on the image recognition technology, obtain the estimated total feed amount in the feed trough at the feeding start time point of each feeding of each live pig corresponding to each monitoring day and the estimated remaining feed amount in the feed trough at the corresponding feeding end time point, and perform a subtraction calculation on them to obtain the intake amount of each feeding of each live pig corresponding to each monitoring day;

[0097] Sum up the intake amounts of each feeding of each live pig corresponding to each monitoring day to obtain the total intake of each live pig corresponding to each monitoring day during the monitoring period in the target pen

[0098] Live pig data analysis module: Analyze the basic parameters of each live pig in the target pen for each monitoring day, and comprehensively evaluate to obtain the status anomaly index of each live pig in the target pen for each monitoring day;

[0099] In a possible design, to comprehensively evaluate and obtain the status anomaly index of each live pig in the target pen for each monitoring day, the specific evaluation process is as follows:

[0100] Obtain the total diarrhea volume of each live pig in the target pen for each monitoring day during the monitoring period , the average individual temperature and the total food intake ;

[0101] Calculate the possible infection index of each live pig for each monitoring day , are respectively the total diarrhea volume, the average individual temperature and the total food intake of the c-th live pig corresponding to the (j - 1)-th monitoring day, are respectively the total diarrhea volume, the average individual temperature and the total food intake of the c-th live pig corresponding to the (j + 1)-th monitoring day, are respectively the set weight coefficients, and C is the total number of monitoring days;

[0102] Comprehensively evaluate to obtain the status anomaly index of each live pig in the target pen for each monitoring day , is the set status anomaly permission threshold.

[0103] Live pig data judgment module: Analyze and judge based on the status anomaly index of each live pig in the target pen for each monitoring day;

[0104] If the status anomaly index of a certain live pig is 1, continue to execute the pen data collection module. If the status anomaly index of a certain live pig is 0, do not execute the pen data collection module and jump out of the system module.

[0105] Pen data collection module: Collect the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period;

[0106] In a possible design, the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period includes ventilation density, feed quality and live pig density.

[0107] Pen data analysis module: Conduct a comprehensive analysis of the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period, analyze the infection probability of each live pig in the target pen for each monitoring day, and then screen out the diseased live pigs in the target pen.

[0108] In a possible design, to analyze the infection probability of each live pig in the target pen for each monitoring day, the specific analysis logic is as follows:

[0109] For each live pig in the target pen, randomly select a monitoring day as the target analysis monitoring day from the pen environment data of each monitoring day during the corresponding monitoring period, and obtain the ventilation density V, feed quality F, and live pig density U on the target analysis monitoring day;

[0110] Let G be the state value of a live pig infected with the coronavirus, where G takes the value of 1 or 0, 1 indicates infection, and 0 indicates non-infection. Thus, a Bayesian network for each live pig corresponding to the target analysis monitoring day is constructed ;

[0111] The prior probability of each live pig corresponding to the target analysis monitoring day being infected with the coronavirus is , and the prior probability of the live pig not being infected with the coronavirus is ;

[0112] Thus, calculate the joint probability of each live pig corresponding to the target analysis monitoring day being infected with the coronavirus ; Calculate the joint probability of each live pig corresponding to the target analysis monitoring day not being infected with the coronavirus ;

[0113] Calculate the probability that ventilation conditions, feed quality, and live pig density appear simultaneously under all possible infection states for each live pig corresponding to the target analysis monitoring day = × + × ;

[0114] Comprehensively calculate the infection probability of each live pig corresponding to the ventilation density, feed quality, and live pig density on the target analysis monitoring day ;

[0115] Using this analysis method, further obtain the infection probability of each live pig in the target pen corresponding to each monitoring day.

[0116] In a possible design, the screening logic for screening out diseased live pigs in the target pen is:

[0117] Assign weights Q1 and Q2 to the state anomaly index and infection probability respectively;

[0118] Thus, calculate the comprehensive index value of each live pig in the target pen , is the infection probability of the c-th live pig in the target pen corresponding to the j-th monitoring day;

[0119] Compare the comprehensive index value of each live pig in the target pen with the threshold . If , then it is determined that the c-th live pig has coronavirus.

[0120] It should be understood that in various embodiments of the present application, the magnitude of the serial numbers of the above processes does not indicate the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0121] It should be understood that determining B based on A does not mean determining B solely based on A, and B can also be determined based on A and / or other information.

[0122] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claimed rights.

[0123] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A rapid detection system for porcine enteric coronavirus, characterized in that, Including: Live pig data extraction module: used to set the monitoring period, and extract the basic parameters of each live pig in the target pen for each monitoring day during the monitoring period according to the monitoring instruments arranged in the target pen; Live pig data analysis module: analyze the basic parameters of each live pig in the target pen for each monitoring day, and comprehensively evaluate to obtain the status anomaly index of each live pig in the target pen for each monitoring day; Live pig data judgment module: analyze and judge based on the status anomaly index of each live pig in the target pen for each monitoring day; Pen data collection module: collect the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period according to the automatic sampling equipment; Pen data analysis module: comprehensively analyze the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period, analyze the infection probability of each live pig in the target pen for each monitoring day, and then screen out the diseased live pigs in the target pen; The pen environment data of each live pig in the target pen for each monitoring day during the monitoring period includes ventilation density, feed quality, and live pig density; Analyze the infection probability of each live pig in the target pen for each monitoring day. The specific analysis logic is as follows: Arbitrarily select a monitoring day from the pen environment data of each live pig in the target pen for each monitoring day during the monitoring period as the target analysis monitoring day, and obtain the ventilation density V, feed quality F, and live pig density U of the target analysis monitoring day; Let \(G\) be the state value of a live pig infected with the coronavirus, where \(G\) takes the value of 1 or 0, 1 indicates infection, and 0 indicates no infection. Thus, a Bayesian network corresponding to each live pig on the target analysis and monitoring date is constructed. ; The prior probability that each live pig is infected with the coronavirus on the target analysis and monitoring day is , then the prior probability that the live pig is not infected with the coronavirus is ; Calculate the joint probability of each live pig being infected with the coronavirus on the target analysis and monitoring date accordingly. ; Calculate the joint probability that the pigs are not infected with the coronavirus for each pig on the target analysis and monitoring date ; Calculate the probability that ventilation conditions, feed quality, and pig density occur simultaneously under all possible infection states for each live pig corresponding to the target analysis and monitoring date ; Comprehensively calculate the infection probability of each live pig corresponding to the ventilation density, feed quality, and live pig density on the corresponding target analysis and monitoring date ; Obtain the infection probability of each live pig in the target pen for each monitoring day through this analysis method.

2. The rapid detection system for porcine enteric coronavirus according to claim 1, characterized in that, The basic parameters of each live pig in the target pen for each monitoring day during the monitoring period include the total diarrhea amount, average individual temperature, and total food intake.

3. A rapid detection system for porcine enteric coronavirus according to claim 2, characterized in that, The process of extracting the total diarrhea amount of each live pig in the target pen for each monitoring day during the monitoring period is as follows: According to the monitoring instruments arranged in the target pen, obtain the monitoring videos of the target pen for each monitoring day during the monitoring period; preprocess the monitoring videos, and simultaneously use the background modeling method to separate the background in the target pen and each live pig in the target pen; Use the target detection algorithm to detect each live pig in the monitoring video, and use the target tracking algorithm to track each live pig in the target pen to obtain the positions of each live pig in the target pen at each monitoring time point for each monitoring day; Use image processing and computer vision techniques to identify the monitoring time points of each diarrhea event corresponding to each live pig in the target pen on each monitoring day , ZL is the frame rate of the monitoring videos on each monitoring day during the monitoring period corresponding to the target pen, is the number of frames in the corresponding monitoring video of the i-th diarrhea event that occurred to the c-th live pig in the target pen on the j-th monitoring day. c is the number of each live pig in the target pen, c = 1, 2,... C; j is the number of each monitoring day, j = 1, 2,... J, and i is the number of each diarrhea event, i = 1, 2,... ; Mark the monitoring time points of each diarrhea event of each live pig in the target pen for each monitoring day in the monitoring video, and overlap it with the positions of each live pig in the target pen at each monitoring time point for each monitoring day, so as to obtain the diarrhea area of each diarrhea event of each live pig in the target pen for each monitoring day; Using image segmentation technology, obtain the total number of pixels in the diarrhea area corresponding to each diarrhea event for each live pig in the target pen on each monitoring day ; The image pixel width of the monitored screen in the collected monitoring video and the image pixel height ; Calculate the expected diarrhea area corresponding to each diarrhea event in each diarrhea area for each live pig in the marked pen on each monitoring day accordingly. , and represent the actual width and actual height corresponding to the camera view, respectively. Furthermore, through the formula , calculate the predicted diarrhea liquid thickness of the diarrhea area corresponding to the i-th diarrhea event of the c-th live pig in the target pen on the j-th monitoring day , is the set liquid characteristic constant; Finally, the total diarrhea volume of each live pig in the target pen during each monitoring day within the corresponding monitoring period is obtained. .

4. The rapid detection system for porcine enteric coronavirus according to claim 3, characterized in that, The process of extracting the average individual temperature of each live pig in the target pen for each monitoring day during the monitoring period is as follows: According to the monitoring instruments arranged in the target pen, collect the infrared temperatures of each live pig in the target pen at each monitoring time point for each monitoring day; simultaneously collect the outdoor temperature, outdoor humidity, indoor temperature of the pen, and indoor humidity of the pen at each monitoring time point corresponding to each monitoring day; Based on the positions of each live pig in the target pen at each monitoring time point within each monitoring day and the positions where the infrared devices are arranged in the target pen, the position angle coefficients and position angle functions corresponding to the positions of the infrared devices for each live pig in the target pen at each monitoring time point within each monitoring day are obtained; The infrared temperature of each live pig in the target pigsty at each monitoring time point within each monitoring day Perform a temperature calibration operation to obtain the calibrated infrared temperature of each live pig in the target pigsty at each monitoring time point within each monitoring day , respectively represent predefined calibration coefficients, d is the number of each monitoring time point, d = 1, 2,... D; Calculate the individual temperature mean of each live pig in the target pen during each monitoring day within the corresponding monitoring period , are the indoor temperature, outdoor temperature, indoor humidity, and outdoor humidity of the pen corresponding to the d-th monitoring time point on the j-th monitoring day, respectively; are the position angle coefficient and position angle function corresponding to the position of the infrared device at the d-th monitoring time point within the j-th monitoring day for the c-th live pig in the target pen, respectively, are the indoor-outdoor temperature difference influence coefficient and humidity difference influence coefficient corresponding to the d-th monitoring time point on the j-th monitoring day, respectively, and D is the total number of monitoring time points.

5. The rapid detection system for porcine enteric coronavirus according to claim 4, wherein Obtaining the position angle coefficients and position angle functions corresponding to the positions of the infrared devices for each live pig in the target pen at each monitoring time point within each monitoring day includes: Obtain the position coordinates of the infrared devices deployed in the target pen and the position coordinates of each live pig at each monitoring time point within each monitoring day ; Calculate the distances between the positions of each live pig in the target pen corresponding to each monitoring time point within each monitoring day and the positions of the infrared devices ; Calculate the relative pitch angles of the positions of each live pig in the target pen corresponding to each monitoring time point within each monitoring day with respect to the positions of the infrared devices and the relative yaw angles ; Furthermore, the position angle coefficient of the position of each live pig in the target pen corresponding to each monitoring time point within each monitoring day with respect to the position of the infrared device is expressed as , where k4, k5, and k6 are all predefined calculation experience coefficients; Finally, the position angle function of the positions of each live pig in the target pen corresponding to the positions of the infrared devices at each monitoring time point within each monitoring day is obtained. , They are respectively the fitted empirical coefficients.

6. The rapid detection system for porcine enteric coronavirus according to claim 2, wherein Extract the total food intake of each live pig in the target pen for each monitoring day during the monitoring period. The specific process is as follows: Identify the feeding time points and the end-feeding time points of each feeding of each live pig within each monitoring day through a behavior recognition algorithm and a target tracking algorithm; And based on image recognition technology, obtain the estimated total feed amount in the feed trough at the feeding time point of each feeding of each live pig within each monitoring day and the estimated remaining total feed amount in the feed trough at the corresponding end-feeding time point, and perform a subtraction calculation on them to obtain the food intake of each feeding of each live pig within each monitoring day; Perform a summation operation on the food intakes of each feeding of each live pig within each monitoring day to obtain the total food intake of each live pig in the target pen for each monitoring day during the monitoring period.

7. A rapid detection system for porcine enteric coronavirus according to claim 1, characterized in that Comprehensively evaluate to obtain the status anomaly index of each live pig in the target pen for each monitoring day. The specific evaluation process is: Obtain the total diarrhea volume, the average individual temperature, and the total food intake of each live pig in the target pen during each monitoring day within the corresponding monitoring period , the average individual temperature , and the total food intake ; Calculate the possible infection index of each live pig corresponding to each monitoring day , which are the total diarrhea volume, the average individual temperature, and the total food intake of the c-th live pig corresponding to the (j - 1)-th monitoring day, respectively, which are the total diarrhea volume, the average individual temperature, and the total food intake of the c-th live pig corresponding to the (j + 1)-th monitoring day, respectively, which are the set weight coefficients, and C is the total number of monitoring days; The comprehensive evaluation obtains the status anomaly index of each live pig in the target pen corresponding to each monitoring day , is the set status anomaly permission threshold value.

8. The rapid detection system for porcine enteric coronavirus according to claim 1, characterized in that, The screening logic for screening out the diseased live pigs in the target pen is: Assign weights Q1 and Q2 to the status anomaly index and the infection probability respectively; Calculate the comprehensive index values of each live pig in the target pen accordingly , is the infection probability of the c-th live pig in the target pen corresponding to the j-th monitoring day; The comprehensive index values of each live pig in the target pen are compared with the threshold value . If , it is determined that the c-th live pig has coronavirus.

Images