CN114241409A - Disease early warning system and method based on abnormal excrement and anatomical images of caged chickens - Google Patents
Disease early warning system and method based on abnormal excrement and anatomical images of caged chickens Download PDFInfo
- Publication number
- CN114241409A CN114241409A CN202111521393.7A CN202111521393A CN114241409A CN 114241409 A CN114241409 A CN 114241409A CN 202111521393 A CN202111521393 A CN 202111521393A CN 114241409 A CN114241409 A CN 114241409A
- Authority
- CN
- China
- Prior art keywords
- excrement
- abnormal
- image
- image information
- early warning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
- G06K17/0029—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device the arrangement being specially adapted for wireless interrogation of grouped or bundled articles tagged with wireless record carriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/182—Level alarms, e.g. alarms responsive to variables exceeding a threshold
Abstract
The invention discloses a disease early warning system based on abnormal excrement of caged chicken and anatomical images, which comprises a client for acquiring the abnormal excrement of caged chicken and analyzing image information, a server for processing and analyzing the abnormal excrement and analyzing the image information, a management system and a human-computer interface used in a matched manner; the client comprises: an abnormal excrement monitoring unit, an abnormal coop positioning unit and an anatomical image analysis unit; the server side comprises: the system comprises a deep learning algorithm model, a sample database and a Web service module. The invention also provides a use method of the early disease early warning system. The early warning of the diseases of the cage-raised chickens is achieved by identifying abnormal excrement and anatomical images.
Description
Technical Field
The invention relates to the technical field of breeding monitoring of cage-raised chickens, in particular to a disease early warning system based on abnormal excrement and anatomical images of cage-raised chickens.
Background
With the increase of the world population and the improvement of the quality of life, the demand of human beings on protein products such as meat, eggs and the like is steadily increasing. The related data shows that over 680 million chickens raised globally in 2018 account for one third of the global meat production, and in addition, 1.38 trillion eggs for human consumption, and the chicken breeding industry provides a large amount of protein products for human beings. With the continuous development and progress of scale and mechanization of chicken breeding industry, growth promoters, antibiotics and the like are gradually forbidden, and the risk of disease transmission and outbreak in the breeding process is greatly increased. The digestive system is a main channel for the chicken to communicate with the external environment, and has high pathogen contact probability, high morbidity and high propagation speed. Therefore, if the health status of the chickens can be judged in the early stage of the infectious diseases of the chickens, the further spread of the diseases can be prevented, the breeding welfare can be improved, and the breeding loss can be greatly reduced.
Patent document CN105574899B discloses a method and system for detecting feces of caged chickens, which comprises placing a coop above a conveyor belt, conveying the feces of chickens in the coop on the conveyor belt along the conveyor belt, sequentially acquiring images of the feces of chickens on the conveyor belt by a CCD camera, obtaining color and thickness degree of the feces on the conveyor belt through image processing, obtaining monitoring data, comparing the color and thickness degree of the feces with a known feces-disease relationship to obtain the type of the feces, and monitoring the health of chickens in each coop. The invention can accurately judge the health status of the poultry, but the invention can not lock the area where the abnormal poultry is located and can not implement targeted solution.
Patent document CN113223035A discloses a cage chicken intelligence system of patrolling and examining, including marginal data acquisition device, server, visual analysis platform, specifically do: the edge data acquisition device is used for acquiring image data of the caged chickens and environment information in the cage and transmitting the image data and the environment information to the server; the image data of the caged chicken comprises an RGB image and a thermal infrared image, and the environment information in the cage comprises temperature and humidity information and crowding degree information; the server analyzes the image data to obtain abnormal chicken head scores and the temperatures of individual caged chickens, obtains inversion temperatures of the caged chickens based on the temperatures of the individual caged chickens and the environment information in the cage, obtains comprehensive health evaluation of the caged chickens based on the abnormal chicken head scores and the inversion temperatures of the caged chickens and issues the comprehensive health evaluation to a visual analysis platform; and the visual analysis platform displays information. The RGB image and the thermal infrared image of the invention are easily influenced by external environment, so that the reliability of the final evaluation result is not high.
Disclosure of Invention
In order to solve the problems, the invention provides a disease early warning system based on abnormal excrement of cage-raised chickens and anatomical images, the system is simple to operate, and early warning of diseases of the cage-raised chickens is achieved by identifying the abnormal excrement.
A disease early warning system based on abnormal excrement of cage-raised chickens and anatomical images comprises a client for acquiring the abnormal excrement of cage-raised chickens and analyzing image information, a server for processing and analyzing the abnormal excrement and analyzing the image information, a management system and a human-computer interface used in a matched mode;
the client comprises:
the abnormal excrement monitoring unit is used for acquiring excrement image information on the excrement cleaning belt;
the abnormal coop positioning unit is used for recording the distribution position of the coop where abnormal excrement is located, wherein the abnormal excrement is located;
the anatomical image analysis unit is used for acquiring the anatomical image information of the digestive system of the chicken in the coop where the abnormal excrement is located;
the server side comprises:
the deep learning algorithm model is used for analyzing and identifying the excrement image information and the digestive system anatomical image information sent by the client and outputting a result;
the sample database is used for providing sample image information for the analysis and identification of the deep learning algorithm model;
and the Web service module is used for identifying and analyzing the excrement image information and the digestive system anatomical image information in a remote manual mode and outputting an analysis report.
Preferably, the abnormal excrement monitoring unit is arranged at the upper end of the excrement outlet of the excrement cleaning belt and comprises a bracket, a high-resolution industrial camera and an illumination device, wherein the bracket spans the excrement cleaning belt, the high-resolution industrial camera and the illumination device are arranged on the bracket, and the imaging range of the high-resolution industrial camera covers the whole excrement outlet area.
Preferably, the abnormal coop locating unit includes an RFID tag and an RFID reader for receiving tag information:
the RFID tag is arranged on the back of the excrement cleaning belt and used for marking the position of the group of excrement;
and the RFID reader and the abnormal excrement monitoring unit are arranged at the upper end of the excrement outlet of the excrement cleaning belt together, and are used for recording the time difference T when each RFID tag passes through and sending the time difference T to the management system.
Preferably, the sample image information includes: abnormal stool image information and digestive system anatomical map information corresponding to various infectious diseases, and normal stool image information and digestive system anatomical map information.
The invention also provides a use method of the early warning system based on the diseases, and the invention identifies and judges the types of the diseases by combining the abnormal excrement image and the digestive system diagram of the sick chicken, thereby facilitating the selection of subsequent diagnosis and treatment methods.
A use method of a disease early warning system based on abnormal excrement and anatomical images of cage-raised chickens comprises the following steps:
s1, setting an early warning threshold value for the management system through a human-computer interface;
s2, when the feces clearing belt runs, video recording is carried out through the abnormal feces monitoring unit, frame extraction is carried out on the video, and feces images to be identified are extracted;
s3, the deep learning algorithm module of the server receives the excrement image, analyzes and identifies the excrement image, and outputs an analysis result;
s4, selecting sample image information with the maximum probability value in the analysis result of S3 as a result and feeding the result back to the client;
the abnormal coop positioning unit of the S5 client analyzes and obtains the distribution condition of the coops corresponding to the excrement image information based on the result in the S4;
s6, when the distribution area of the coop corresponding to the abnormal excrement image is larger than the early warning threshold value, an alarm is sent out to prompt a worker to check;
s7, dissecting the abnormal excrement image corresponding to the sick chicken in the coop, and extracting a digestive system dissection image to be identified through a dissection image analysis unit;
s8, the deep learning algorithm module of the server receives the digestive system anatomical map, analyzes and identifies the digestive system anatomical map, and outputs an analysis result;
s9 calculates a comprehensive assessment value for determining the disease category based on the analysis result of S3 and the analysis result of S8, and feeds back the final result to the client.
Preferably, the distribution of the coops corresponding to the feces image information obtained by analyzing in S5 is based on the fact that the time t for the feces of each coop to be delivered to the feces clearing and bringing-out opening is stored in the management systemnAnd comparing the distribution position of the coop corresponding to the excrement image information represented by the RFID tag with the time difference T recorded by the RFID reader to judge: when | T-TnWhen | ≦ Δ T, the fecal image corresponds to the position of the coop with number n, and Δ T is a time threshold preset in the management system; the excrement cleaning belt rotates one wheel every 12 hours, and excrement image information is updated every 12 hours.
The management system also stores the number of chickens in each chicken coop, the age of the chickens in day, an illumination program, a feeding program, the illness history and the vaccination condition.
When a worker checks a coop where abnormal excrement is located, calling information data of all chickens in the coop according to the corresponding coop position, and preliminarily judging the source of the problem; furthermore, the remote expert doctor can also call the information through the web module to provide a theoretical basis for disease diagnosis and treatment scheme design.
Preferably, the early warning threshold is set based on the number of coops in the chicken house: when the number of the coops is more, the early warning threshold value is larger, otherwise, the early warning threshold value is smaller.
Preferably, the analysis result in S3 includes probability values P of the stool images of the respective categories; the analysis result in S8 includes probability values Q of the digestive system images of each category.
Preferably, in S9, the weighted summation is performed based on the probability value P of each type of stool image and the probability value Q of each type of digestive system image, the weight of the probability value P of each type of stool image is set to 0.4, the weight of the probability value Q of each type of digestive system image is set to 0.6, and the comprehensive evaluation value of each type of stool image and the corresponding digestive system image is obtained.
Compared with the prior art, the invention has the beneficial effects that:
(1) compared with the traditional early warning system, the method adopts the excrement image and the digestive system image to identify and judge the category of the disease;
(2) furthermore, the proposal of the specialist doctor can be obtained on line through the web module, the time for waiting the specialist doctor to go home or go out is reduced, and the state of illness in the chicken coop can be controlled in time.
Drawings
FIG. 1 is a system structure diagram of an early disease warning system based on abnormal feces and anatomical images of a caged chicken according to the present invention;
FIG. 2 is a flow chart of the establishment of a stool image detection model in an embodiment of the present invention;
fig. 3 is a flowchart illustrating the process of constructing the classification model of the anatomical image according to the embodiment of the present invention.
Detailed Description
As shown in fig. 1, the early disease warning system based on abnormal excrement of caged chicken and anatomical images comprises a client for acquiring the abnormal excrement of caged chicken and analyzing image information, a server for processing and analyzing the abnormal excrement and analyzing the image information, a management system and a human-computer interface used in cooperation.
The client comprises:
the abnormal excrement monitoring unit is used for acquiring excrement image information on the excrement cleaning belt;
the abnormal coop positioning unit is used for recording the distribution position of the coop where abnormal excrement is located, and comprises an RFID tag and an RFID reader used for receiving tag information:
the RFID tag is arranged on the back of the excrement cleaning belt and used for marking the position of the group of excrement;
the RFID reader and the abnormal excrement monitoring unit are arranged at the upper end of the excrement outlet of the excrement cleaning belt together, and are used for recording the time difference T when each RFID tag passes through and sending the time difference T to the management system;
and the anatomical image analysis unit is used for acquiring the anatomical image information of the digestive system of the chicken in the coop where the abnormal excrement is located.
The server side comprises:
the deep learning algorithm model is used for analyzing and identifying the excrement image information and the digestive system anatomical image information sent by the client and outputting a result;
the sample database is used for providing sample image information for the analysis and identification of the deep learning algorithm model;
and the Web service module is used for identifying and analyzing the excrement image information and the digestive system anatomical image information in a remote manual mode and outputting an analysis report.
The abnormal excrement monitoring unit is arranged at the upper end of an excrement outlet of the excrement cleaning belt and comprises a support, a high-resolution industrial camera and an illumination device, wherein the support stretches across the excrement cleaning belt, the high-resolution industrial camera and the illumination device are arranged on the support, and the imaging range of the high-resolution industrial camera covers the whole excrement outlet area.
Establishing a sample database:
s1, breeding the chickens in cages in laboratories or under other conditions with controllable environmental parameters, dividing the chickens into a healthy group and a diseased group after the chickens are familiar with the environment, wherein the groups are not interfered with each other, and the diseased group enables the chickens to suffer from various infectious diseases in an artificial active infection mode, wherein the infectious diseases comprise avian influenza, Newcastle disease, fowl typhoid, chicken colibacillosis, fowl cholera, histomoniasis, pullorum disease, infectious bursal disease, small intestinal coccidiosis, cecal coccidiosis, necrotic enteritis, fish meal putrescence poisoning, mycotoxin poisoning and other infectious diseases;
s2, collecting fresh excrement images of a healthy group and a diseased group under different light environment conditions every 12 hours after infection, and collecting digestive system anatomical images of a corresponding group every 24 hours, wherein the digestive system anatomical images comprise images of 6 parts of glandular stomach, muscular stomach, liver, small intestine, cecum and rectum, and natural light, LED lamps and fluorescent lamps are adopted to simulate different light environment conditions;
s3, dividing the stool image and the corresponding digestive system anatomical image into a training set, a verification set and a test set according to the ratio of 6:2:2, wherein 1 stool image is a sample, and 6 corresponding digestive system anatomical images are samples;
s4, because a plurality of fecal masses exist in one fecal image, the fecal image analysis is set as a target detection task, Labelimg is adopted to label each fecal mass in the image as healthy or infected with a certain disease plus the number of sick days, such as avian influenza-0.5, or Newcastle disease-1, etc., namely, N x N +1 types, and a corresponding xml file is generated after each image is labeled; marking the anatomical image of the digestive system in the same way to generate a corresponding xml file, wherein N is the name of a disease category, and N is the number of days of illness;
s5 saves the stool image and the digestive system anatomical image in S4, and the corresponding xml file in the sample data.
As shown in fig. 2 and 3, the deep learning algorithm model is established, which includes a stool image detection model and an anatomical image classification model:
1. establishing a fecal image detection model:
s1 adopts yolov5 network structure;
s2 model environment configuration;
s3, the target category parameter in the network structure corresponds to the category of the sample database;
s4, the original network parameters adopt parameters obtained by training on the public data sets such as ImageNet and the like, and the newly added parameters are initialized by Kaiming;
s5, loading a training set to perform iterative training, and storing a model under each iteration;
s6, loading a verification set to perform iterative verification on the model under each iteration, stopping when the model converges or reaches the maximum iteration times, storing the model before verification effect 10 for testing, wherein the model effect is evaluated by adopting an mAP index, the mAP is the mean value of the area enclosed by the Precion-Recall curves of various categories and coordinate axes, and the larger the value is, the better the model effect is represented:
precision: precision ratio, which is the correct ratio of all targets predicted by the model;
recall: recall ratio, predicting the correct proportion for the model in all real targets;
TP: true Positive, which is determined to be a Positive sample, and in fact is also a Positive sample;
TN: true Negative, determined as a Negative sample, in fact also a Negative sample;
FP: false Positive, judged as Positive, but in fact negative;
FN: false Negative, judged as Negative, but in fact positive;
s7, loading a test set to carry out iterative test on the 10 models, stopping when the models are converged or reach the maximum iterative times, and taking the model with the optimal test effect, namely the maximum mAP value, as a final fecal image detection model.
2. Establishing an anatomical image classification model:
s1 adopts a CoAtNet-7 network structure;
s2 model environment configuration;
s3 associates the target category parameter in the network structure with the category of the sample database: building a network structure formed by connecting 6 CoAtNet-7 networks in parallel for the anatomical image classification model based on 6 anatomical images as a sample;
s4, the original network parameters adopt parameters obtained by training on the public data sets such as ImageNet and the like, and the newly added parameters are initialized by Kaiming;
s5, loading a training set to perform iterative training, and storing a model under each iteration;
s6, loading a verification set to perform iterative verification on the model under each iteration, stopping when the model converges or reaches the maximum iteration times, storing the model 10 before the verification effect for testing, and evaluating by adopting Macro-F1 score:
Macro-F1 score is the mean of the various classes F1 score;
f1 score, which is the harmonic mean of Precision and Recall.
A use method of a disease early warning system based on abnormal excrement and anatomical images of cage-raised chickens comprises the following steps:
s1, setting an early warning threshold value for the management system through a human-computer interface;
s2, when the feces clearing belt runs, video recording is carried out through the abnormal feces monitoring unit, frame extraction is carried out on the video, and feces images to be identified are extracted;
the deep learning algorithm module of the S3 server receives the excrement image, analyzes and identifies the excrement image through an excrement image detection model, and outputs the probability value P of each type of excrement image;
s4, selecting sample image information with the maximum probability value in the analysis result of S3 as a result and feeding the result back to the client;
the abnormal coop positioning unit of the S5 client side is based on the result in the S4 and the time t for the excrement of each coop to be conveyed to the excrement outlet of the excrement cleaning belt stored in the management systemnAnd comparing the distribution position of the coop corresponding to the excrement image information represented by the RFID tag with the time difference T recorded by the RFID reader to judge: when | T-TnWhen | ≦ Δ T, the fecal image corresponds to the position of the coop with number n, and Δ T is a time threshold preset in the management system; wherein the feces cleaning belt rotates one wheel every 12 hours, and simultaneously, the feces image information is updated once every 12 hours;
s6, when the distribution area of the coops corresponding to the abnormal excrement images is larger than an early warning threshold value, an alarm is sent to prompt a worker to check, and when the worker checks the coops where the abnormal excrement is located, the worker preliminarily judges the source of the problem by calling information data of all chickens in the corresponding coops stored in a management system, wherein the information data include the number of chickens in each coop, the day age of each chicken, an illumination program, a feeding program, illness history and vaccination conditions;
s7, dissecting the abnormal excrement image corresponding to the sick chicken in the coop, and extracting a digestive system dissection image to be identified through a dissection image analysis unit, wherein the digestive system dissection image comprises images of 6 parts of glandular stomach, muscular stomach, liver, small intestine, cecum and rectum;
s8, a deep learning algorithm module of the server receives the digestive system anatomical map, analyzes and identifies the digestive system anatomical map through an anatomical image classification model, and outputs a probability value Q of each type of digestive system image;
s9 carries out weighted summation based on the probability value P of each type of excrement image and the probability value Q of each type of digestive system image, the weight of the probability value P of each type of excrement image is set to be 0.4, the weight of the probability value Q of each type of digestive system image is set to be 0.6, the comprehensive evaluation value of each type of excrement image and the corresponding digestive system image is obtained, and the final result is fed back to the client.
The final result comprises that the system identifies the judged disease category through the comprehensive evaluation value, and remotely obtains the treatment suggestion provided by the expert doctor through the web module based on the identified disease category.
Meanwhile, an expert doctor accesses a management system of the client through the web module, and provides a more accurate diagnosis result and a treatment scheme with higher adaptability for a user according to the information data of the corresponding coop.
Claims (9)
1. A disease early warning system based on abnormal excrement of cage-raised chickens and anatomical images is characterized by comprising a client for acquiring the abnormal excrement of the cage-raised chickens and analyzing image information, a server for processing and analyzing the abnormal excrement and analyzing the image information, a management system and a human-computer interface used in a matched mode;
the client comprises:
the abnormal excrement monitoring unit is used for acquiring excrement image information on the excrement cleaning belt;
the abnormal coop positioning unit is used for recording the distribution position of the coop where abnormal excrement is located;
the anatomical image analysis unit is used for acquiring the anatomical image information of the digestive system of the chicken in the coop where the abnormal excrement is located;
the server side comprises:
the deep learning algorithm model is used for analyzing and identifying the excrement image information and the digestive system anatomical image information sent by the client and outputting a result;
the sample database is used for providing sample image information for the analysis and identification of the deep learning algorithm model;
and the Web service module is used for identifying and analyzing the excrement image information and the digestive system anatomical image information in a remote manual mode and outputting an analysis report.
2. The early warning system for diseases according to claim 1, wherein the abnormal stool monitoring unit is installed at the upper end of the place where the feces outlet of the feces cleaning belt is located.
3. The early warning system of disease as claimed in claim 1, wherein the abnormal coop locating unit comprises an RFID tag and an RFID reader for receiving tag information:
the RFID tag is arranged on the back of the excrement cleaning belt and used for marking the position of the group of excrement;
and the RFID reader and the abnormal excrement monitoring unit are arranged at the upper end of the excrement outlet of the excrement cleaning belt together, and are used for recording the time difference T when each RFID tag passes through and sending the time difference T to the management system.
4. The early warning system of disease as claimed in claim 1, wherein the sample image information includes: abnormal stool image information and digestive system anatomical map information corresponding to various infectious diseases, and normal stool image information and digestive system anatomical map information.
5. A use method of the cage chicken abnormal excrement and anatomical image-based disease early warning system according to any one of claims 1-4, characterized by comprising the following steps:
s1, setting an early warning threshold value for the management system through a human-computer interface;
s2, when the feces clearing belt runs, video recording is carried out through the abnormal feces monitoring unit, frame extraction is carried out on the video, and feces images to be identified are extracted;
s3, a deep learning algorithm module of the server receives the excrement image, analyzes and identifies the excrement image, and outputs an analysis result;
s4, selecting sample image information with the maximum probability value in the analysis result of S3 as a result and feeding the result back to the client;
the abnormal coop positioning unit of the S5 client analyzes and obtains the distribution condition of the coops corresponding to the excrement image information based on the result in the S4;
s6, when the distribution area of the coop corresponding to the abnormal excrement image is larger than the early warning threshold value, an alarm is sent out to prompt a worker to check;
s7 dissects the abnormal excrement image corresponding to the sick chicken or the dead chicken in the coop, and extracts a digestive system dissection image to be identified through a dissection image analysis unit;
s8, a deep learning algorithm module of the server receives the digestive system anatomical map, analyzes and identifies the digestive system anatomical map, and outputs an analysis result;
s9 calculates a comprehensive assessment value for determining the disease category based on the analysis result of S3 and the analysis result of S8, and feeds back the final result to the client.
6. The method of claim 5The application method of the disease early warning system based on abnormal excrement and anatomical images of the caged chickens is characterized in that the distribution condition of the coops corresponding to the excrement image information is obtained by analyzing in the S5, and the distribution condition is based on the fact that the time t when the excrement of each coop is transmitted to the excrement removal belt excrement outlet is stored in the management systemnAnd comparing the distribution position of the coop corresponding to the excrement image information represented by the RFID tag with the time difference T recorded by the RFID reader to judge the distribution position of the coop corresponding to the excrement image information represented by the RFID tag, wherein n is the coop number.
7. The use method of the early warning system for diseases based on abnormal feces and anatomical images of caged chicken as claimed in claim 5, wherein the warning threshold is set based on the number of cages in the henhouse: when the number of the coops is more, the early warning threshold value is larger, otherwise, the early warning threshold value is smaller.
8. The use method of the early disease warning system based on abnormal feces and anatomical images of caged chicken according to claim 5, wherein the analysis result in S3 comprises probability values P of feces images of each category; the analysis result in S8 includes probability values Q of the digestive system images of each category.
9. The use method of the early warning system for diseases based on abnormal feces of caged chicken and anatomical images as claimed in claim 5, wherein in the step S9, based on the probability value P of each type of feces image and the probability value Q of each type of digestive system image, weighted summation is performed, the weight of the probability value P of each type of feces image is set to 0.4, the weight of the probability value Q of each type of digestive system image is set to 0.6, and the comprehensive evaluation value of each type of feces image and the corresponding digestive system image is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111521393.7A CN114241409A (en) | 2021-12-13 | 2021-12-13 | Disease early warning system and method based on abnormal excrement and anatomical images of caged chickens |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111521393.7A CN114241409A (en) | 2021-12-13 | 2021-12-13 | Disease early warning system and method based on abnormal excrement and anatomical images of caged chickens |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114241409A true CN114241409A (en) | 2022-03-25 |
Family
ID=80755375
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111521393.7A Pending CN114241409A (en) | 2021-12-13 | 2021-12-13 | Disease early warning system and method based on abnormal excrement and anatomical images of caged chickens |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114241409A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114898405A (en) * | 2022-05-27 | 2022-08-12 | 南京农业大学 | Portable broiler chicken abnormity monitoring system based on edge calculation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110888108A (en) * | 2019-11-14 | 2020-03-17 | 天津大学 | Positioning method based on RFID and phase calibration |
| CN111709333A (en) * | 2020-06-04 | 2020-09-25 | 南京农业大学 | Traceability early warning system and health monitoring method based on abnormal excrement of cage-raised chickens |
| CN111867465A (en) * | 2018-03-12 | 2020-10-30 | 勒维斯公司 | System and method for generating thin image slices from thick image slices |
| CN113329622A (en) * | 2018-11-30 | 2021-08-31 | 惠托斯饲料添加剂股份有限公司 | System for analyzing animal waste images |
-
2021
- 2021-12-13 CN CN202111521393.7A patent/CN114241409A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111867465A (en) * | 2018-03-12 | 2020-10-30 | 勒维斯公司 | System and method for generating thin image slices from thick image slices |
| CN113329622A (en) * | 2018-11-30 | 2021-08-31 | 惠托斯饲料添加剂股份有限公司 | System for analyzing animal waste images |
| CN110888108A (en) * | 2019-11-14 | 2020-03-17 | 天津大学 | Positioning method based on RFID and phase calibration |
| CN111709333A (en) * | 2020-06-04 | 2020-09-25 | 南京农业大学 | Traceability early warning system and health monitoring method based on abnormal excrement of cage-raised chickens |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114898405A (en) * | 2022-05-27 | 2022-08-12 | 南京农业大学 | Portable broiler chicken abnormity monitoring system based on edge calculation |
| CN114898405B (en) * | 2022-05-27 | 2023-08-25 | 南京农业大学 | Portable broiler chicken anomaly monitoring system based on edge calculation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Automated techniques for monitoring the behaviour and welfare of broilers and laying hens: towards the goal of precision livestock farming | |
| Wurtz et al. | Recording behaviour of indoor-housed farm animals automatically using machine vision technology: A systematic review | |
| Barreto et al. | Emerging indicators of fish welfare in aquaculture | |
| EFSA Panel on Animal Health and Welfare (AHAW) | Statement on the use of animal‐based measures to assess the welfare of animals | |
| US20140155756A1 (en) | System and methods for health monitoring of anonymous animals in livestock groups | |
| US20220039357A1 (en) | System For Analyzing Animal Secretion Images | |
| KR102165891B1 (en) | Livestock data analysis-based farm health state diagnosis system | |
| WO2021172677A1 (en) | Image artificial intelligence-based method for diagnosing prognosis of companion animal disease | |
| Brünger et al. | ‘Tailception’: using neural networks for assessing tail lesions on pictures of pig carcasses | |
| CN114898405B (en) | Portable broiler chicken anomaly monitoring system based on edge calculation | |
| CN114241409A (en) | Disease early warning system and method based on abnormal excrement and anatomical images of caged chickens | |
| Li et al. | Image processing for analyzing broiler feeding and drinking behaviors | |
| CN109416351B (en) | System and method for detecting intestinal diseases, in particular animal intestinal diseases, based on odor emission | |
| Mei et al. | Identification of aflatoxin-poisoned broilers based on accelerometer and machine learning | |
| Exadaktylos et al. | Chapter Automatic Identification and Interpretation of Animal Sounds, Application to Livestock Production Optimisation | |
| Abbas et al. | Prospects and challenges of adopting and implementing smart technologies in poultry production. | |
| Zhou et al. | Detection of abnormal chicken droppings based on improved Faster R-CNN | |
| Lupo et al. | Feasibility of screening broiler chicken flocks for risk markers as an aid for meat inspection | |
| CN115359418A (en) | Livestock delivery monitoring and early warning system and method based on CLIP model | |
| CN111652084B (en) | Abnormal layer identification method and device | |
| JP2022164589A (en) | Determination system | |
| WO2020116775A1 (en) | Farm productivity and disease analysis device using big data, and productivity and disease analysis method using same | |
| Dawkins et al. | Poultry welfare monitoring: group-level technologies | |
| WO2023229327A1 (en) | Image-based apparatus and method for analyzing herding behavior pattern of herding individuals | |
| Siegford et al. | Practical considerations for the use of precision livestock farming to improve animal welfare. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220325 |
|
| RJ01 | Rejection of invention patent application after publication |