CN117029904A - Intelligent cage-rearing poultry inspection system - Google Patents
Intelligent cage-rearing poultry inspection system Download PDFInfo
- Publication number
- CN117029904A CN117029904A CN202310568261.2A CN202310568261A CN117029904A CN 117029904 A CN117029904 A CN 117029904A CN 202310568261 A CN202310568261 A CN 202310568261A CN 117029904 A CN117029904 A CN 117029904A
- Authority
- CN
- China
- Prior art keywords
- poultry
- cage
- image data
- temperature
- server
- 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
- 244000144977 poultry Species 0.000 title claims abstract description 173
- 238000007689 inspection Methods 0.000 title claims abstract description 85
- 238000001514 detection method Methods 0.000 claims abstract description 56
- 230000002159 abnormal effect Effects 0.000 claims abstract description 22
- 230000007246 mechanism Effects 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 20
- 238000001931 thermography Methods 0.000 claims abstract description 16
- 235000013594 poultry meat Nutrition 0.000 claims description 162
- 230000036760 body temperature Effects 0.000 claims description 27
- 230000036541 health Effects 0.000 claims description 18
- 230000005540 biological transmission Effects 0.000 claims description 10
- 241000287828 Gallus gallus Species 0.000 claims description 9
- 235000013330 chicken meat Nutrition 0.000 claims description 9
- 230000007613 environmental effect Effects 0.000 claims description 9
- 230000003993 interaction Effects 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 7
- 230000001360 synchronised effect Effects 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000009395 breeding Methods 0.000 claims description 4
- 230000001488 breeding effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 11
- 238000013135 deep learning Methods 0.000 abstract description 7
- 208000027954 Poultry disease Diseases 0.000 abstract description 2
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000009374 poultry farming Methods 0.000 description 3
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 2
- 208000028752 abnormal posture Diseases 0.000 description 2
- 238000012364 cultivation method Methods 0.000 description 2
- 210000003746 feather Anatomy 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 208000015181 infectious disease Diseases 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 235000013601 eggs Nutrition 0.000 description 1
- 238000009313 farming Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
- B62D—MOTOR VEHICLES; TRAILERS
- B62D63/00—Motor vehicles or trailers not otherwise provided for
- B62D63/02—Motor vehicles
- B62D63/04—Component parts or accessories
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D11/00—Component parts of measuring arrangements not specially adapted for a specific variable
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J5/0022—Radiation pyrometry, e.g. infrared or optical thermometry for sensing the radiation of moving bodies
- G01J5/0025—Living bodies
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J5/48—Thermography; Techniques using wholly visual means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Remote Sensing (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Radar, Positioning & Navigation (AREA)
- General Health & Medical Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Combustion & Propulsion (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Medical Informatics (AREA)
- Automation & Control Theory (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
An intelligent cage-rearing poultry inspection system and an intelligent cage-rearing poultry inspection method relate to the technical fields of poultry disease detection technology, computer machine vision technology, deep learning technology and inspection robot, and are characterized in that: the system comprises a data acquisition system, a lifting mechanism, a server, a client and a patrol system, wherein the data acquisition system comprises an image data acquisition module and an environment data acquisition module, the image data acquisition module is used for acquiring image data of cage-reared poultry, and the environment data acquisition module is used for acquiring environment data of a poultry house and transmitting the data to the server; the image data of the poultry in the cage comprises RGB image data and infrared thermal imaging image data of the poultry in the cage, and the environment data of the poultry house comprises temperature information, humidity information and loudness information of the environment of the poultry house; the server is used for processing the image data and transmitting the data, and analyzing whether the body state of the poultry is abnormal or not and whether the temperature of the poultry is abnormal or not.
Description
Technical Field
The invention relates to the technical fields of poultry disease detection technology, computer machine vision technology, deep learning technology and inspection robots, in particular to an intelligent poultry health detection method and an inspection system.
Background
With the increasing level of living, the demand of the public for eggs and meat is increasing, and the poultry farming is rapidly developing. At present, poultry farms are developing to large scale and intensification, and the cultivation scale and density are also expanding. Therefore, most poultry farms adopt laminated cages for cultivation, and the cultivation mode can greatly improve the cultivation density and increase the economic benefit. However, the occurrence and spread of epidemic diseases are easy to occur due to the too high culture density. Therefore, the detection of the health condition of the poultry is a necessary thing in time, so that the pathological changes of the breeding poultry can be found as early as possible, the spread of diseases is reduced, and the economic loss of farmers is reduced.
In order to solve the problems, the efficiency and the quality of the poultry farming industry are improved, the health condition of the poultry needs to be monitored and evaluated timely, accurately and comprehensively, in the traditional farming mode, the traditional poultry health detection method mainly depends on manual observation, sampling detection or later diagnosis, the methods have the defects of time consumption, labor consumption, inefficiency, inaccuracy and the like, the living state, pathological conditions and the like of the poultry completely need to be judged by a farmer by experience, symptoms of certain diseases are not obvious at the early stage of infection, the diseases are difficult to observe through human eyes, and the requirement of a large number of inspection is difficult to meet by adopting traditional manual inspection along with the expansion of the poultry farming scale. Moreover, manual inspection has high labor cost and low efficiency, and is easy to cause cross contact infection of poultry. Therefore, the automatic inspection equipment is used for replacing manual work to monitor the health condition of the poultry, and the automatic inspection equipment becomes a necessary development trend.
There are several studies and inventions on intelligent poultry health detection systems and methods. For example, CN201910303625.0 discloses an intelligent health monitoring system for laying hens based on deep learning and a method thereof, the system comprises an image acquisition module, an image processing module, a data analysis module and a data display module, the abnormal posture of the laying hens can be detected by an image recognition technology, and the detection result is displayed on a terminal device. The system can automatically identify the abnormal posture of the laying hen, but can not detect the abnormal body temperature of the laying hen and can not detect other types of poultry.
Therefore, the invention aims to provide an intelligent cage poultry inspection system and method, the system can collect and process image data of cage poultry and environment data of a poultry house, detect the posture and the body temperature of the poultry through an improved YOLOv8 algorithm, judge whether the poultry is abnormal in health, and display the detection result and the environment information to a user so as to realize remote control; the system can be suitable for cage-raised poultry of different types and varieties, and has the characteristics of high efficiency, accuracy and intelligence.
Disclosure of Invention
Aiming at the problems in the background technology, the invention provides an intelligent cage-rearing poultry inspection system, which aims at taking cage-rearing poultry as a research object, non-contact, accurate and automatic cage-rearing poultry health detection as a target, takes a computer vision technology, a deep learning technology and an inspection robot technology as main technical means, develops the cage-rearing poultry health detection system and the automatic inspection system, realizes real-time automatic inspection of the health condition of the cage-rearing poultry, can feed back the detected result in real time, and provides an effective method for ensuring the health of the cage-rearing poultry and intelligent poultry cultivation.
The invention provides an intelligent cage-rearing poultry inspection system, which is characterized in that: the system comprises a data acquisition system, a lifting mechanism, a server, a client and a patrol system, wherein the data acquisition system comprises an image data acquisition module and an environment data acquisition module, the image data acquisition module is used for acquiring image data of cage-reared poultry, and the environment data acquisition module is used for acquiring environment data of a poultry house and transmitting the data to the server;
the image data of the poultry in the cage comprises RGB image data and infrared thermal imaging image data of the poultry in the cage, and the environment data of the poultry house comprises temperature information, humidity information and loudness information of the environment of the poultry house;
the server is used for processing the image data and transmitting the data, analyzing whether the body state of the poultry is abnormal or not and whether the temperature of the poultry is abnormal or not, sending the analysis result to the client, and transmitting a signal to the inspection system according to the analysis result;
the client is used for displaying data processing results and environmental information, and can realize remote control;
the inspection system is used for carrying the data acquisition system and the server, and can automatically inspect in the poultry house.
As a further limitation of the technical scheme, the data acquisition system comprises a high-definition camera, an infrared camera, a temperature sensor, a humidity sensor and a decibel sensor, and is used for acquiring image data of poultry and environmental data of a poultry house;
the high-definition camera, the infrared camera, the temperature sensor, the humidity sensor and the decibel sensor are connected with the server in a wired connection mode;
the high-definition camera and the infrared camera are carried on the synchronous belt guide rail which is vertically arranged and can move in the vertical direction, so that shooting angles are ensured to be right against a poultry breeding cage;
the synchronous belt guide rail, the temperature sensor, the humidity sensor and the decibel sensor are mounted on a self-propelled inspection chassis in the inspection system.
As a further limitation of the technical scheme, the server comprises a data processing module and a data transmission module, wherein the data processing module comprises an industrial personal computer, and the data transmission module comprises a communication module;
the industrial personal computer is placed in the self-propelled inspection chassis and used for processing and analyzing poultry image data, and the communication module is used for data transmission among the server, the client and the inspection system;
as a further limitation of the technical scheme, the client comprises a visual interaction interface, namely a display, of the computer end and the mobile phone end, wherein the visual interaction interface of the computer end and the mobile phone end is used for receiving the image data of the poultry raising in cages, the environment data of the poultry houses and the information transmitted by the server, which are acquired by the display data acquisition module, and the user can control the inspection system through the visual interaction interface of the computer end and the mobile phone end.
As a further limitation of the technical scheme, the inspection system comprises the self-propelled inspection chassis, the self-propelled inspection chassis is connected with the support of the lifting mechanism through bolts, a control module, the server, a power module and a walking module are arranged in the self-propelled inspection chassis, the environment data acquisition module is arranged on the self-propelled inspection chassis, and the image data acquisition module 2 is connected with the L-shaped platform of the lifting mechanism through bolts;
the control module is used for controlling the whole work of the chassis;
the power module is used for supplying power to the data acquisition module, the server and the self-propelled inspection chassis;
the walking module comprises a direct current motor and a walking wheel.
The invention also provides a novel method: the method is characterized in that: the method comprises the following steps:
s1, a data acquisition system acquires RGB image data and infrared thermal imaging image data of poultry and transmits the RGB image data and the infrared thermal imaging image data to a server;
s2, processing the RGB image data through a deep neural network in a server to position poultry;
and S3, disposing a poultry gesture recognition detection model and a poultry body temperature recognition detection model in the server, extracting characteristics of the region of interest, judging whether the gesture of the poultry is abnormal or not, and judging whether the body temperature of the poultry exceeds a normal range or not.
As a further limitation of the technical scheme, the deep neural network adopts an improved YOLOv8 algorithm, named as a C-YOLOv8 algorithm, and is used as an identification algorithm for detecting the posture, the body temperature and the camera positioning of the poultry, the algorithm is improved according to the detected part of the poultry, the detection precision and the detection speed are improved, and the network structure of the C-YOLOv8 algorithm is divided into: the C-YOLOv8 algorithm is mainly modified in the main network, and after three C2f modules in the main network of the YOLOv8 algorithm, an attention mechanism module CBAM is added, wherein the CBAM is used for connecting a channel attribute and a spatial attribute in series, so that the number of layers of the network is slightly increased, and the detection precision is improved.
By further limiting the technical scheme, the poultry gesture recognition detection model carries out chicken head anomaly scoring based on anomalies of all local parts, and combines the inversion temperature and crowding degree of the cage-raised chickens to form a comprehensive score, so that comprehensive evaluation of health of the cage-raised chickens is realized.
As a further limitation of the technical scheme, the camera positioning recognition model recognizes the central position of the cultivation cage according to the rectangular frame of the cultivation cage, and when the central position of the cultivation cage is at the central position of the image, the server sends a signal to the controller through the serial port to control the lifting mechanism to stop lifting.
As a further limitation of the technical scheme, the infrared image temperature extraction method comprises the following steps: acquiring a thermal imaging image, extracting temperature information of pixel points in the image, identifying an interested region according to an infrared image poultry identification algorithm model, outputting coordinate information of the interested region, extracting the temperature of the interested region through coordinate conversion, filtering out the ambient temperature or abnormal low temperature of the interested region, extracting the highest temperature and the average temperature of the region in the interested region after filtering the ambient temperature, comparing with the set abnormal temperature, and judging whether the poultry body temperature is abnormal or not;
the calculation formula of the average body temperature of the region of interest is as follows:in the above, P W For average body temperature, T W In order to remove the total temperature of the pixels after the ambient temperature, N is the number of pixels left after the ambient temperature is removed.
The invention has the following beneficial effects:
(1) An intelligent cage-rearing poultry inspection system. The recognition algorithm in the server adopts an improved YOLOv8 algorithm to recognize the poultry head, so that the detection accuracy is improved; the lifting guide rail is arranged on the self-propelled inspection chassis, the high-definition camera and the infrared camera are carried on the lifting guide rail, so that the device can adapt to a laminated cage cultivation method adopted by a modern poultry farm, the lifting guide rail can be lifted to the top layer from the first layer of the laminated cage, the image data acquisition work of poultry can be completed only by using one group of cameras, the multi-layer camera structure is not carried, and the whole structure is simplified.
(2) And (5) establishing a mixed detection model of the poultry posture and the poultry body temperature. The method comprises the steps of simultaneously shooting by using a high-definition camera and an infrared camera, simultaneously collecting RGB images and infrared images of poultry, and simultaneously identifying and detecting the posture of the poultry and the body temperature of the poultry by using a deep convolutional neural network detection model. And transmitting the detection result to the client through the data transmission system.
(3) And (5) collecting poultry house environment multi-metadata. And when the poultry image data are detected, environment information of the poultry house including temperature information, humidity information and loudness information is acquired, and the quality of the poultry cultivation environment is evaluated in a multi-element manner, so that the welfare cultivation of the poultry is promoted.
(4) The self-propelled inspection chassis in the inspection system can bear equipment to automatically inspect in the poultry house. The camera carried on the front end of the self-propelled inspection chassis can identify the road in the poultry house, carry out path planning, automatically avoid obstacles, automatically return to the journey when the electric quantity of the power supply is insufficient, return to a charging point for charging, and the inspection system can realize whole-course unmanned detection.
Drawings
FIG. 1 is a schematic diagram of a patrol equipment according to an embodiment of the present invention;
FIG. 2 is a flow chart of the identification model establishment of the poultry health detection method according to the embodiment of the invention;
FIG. 3 is a flow chart of a camera positioning method according to an example of the present invention;
FIG. 4 is a flow chart of infrared image temperature extraction according to an example of the present invention;
FIG. 5 is a flow chart of an inspection system according to an example of the present invention;
FIG. 6 is a diagram of a C-YOLOv8 network architecture shown in an example of the invention;
FIG. 7 is a diagram of RGB image recognition results according to an example of the present invention;
FIG. 8 is a graph of the infrared image recognition result shown in an example of the present invention;
in the figure: 1. self-propelled inspection chassis; 2. an image acquisition module; 3. an environmental data acquisition module; 4. a lifting mechanism; 5. a display;
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the scope of the present invention is not limited to the embodiments.
As shown in fig. 1, the system comprises a data acquisition system, a lifting mechanism 4, a server, a client and a patrol system, wherein the data acquisition system comprises an image data acquisition module 2 and an environment data acquisition module 3, the image data acquisition module 2 is used for acquiring image data of cage-reared poultry, the environment data acquisition module 3 is used for acquiring environment data of a poultry house and transmitting the data to the server, and the client comprises the display 5;
the image data of the poultry in the cage comprises RGB image data and infrared thermal imaging image data of the poultry in the cage, and the environment data of the poultry house comprises temperature information, humidity information and loudness information of the environment of the poultry house;
the server is used for processing the image data and transmitting the data, analyzing whether the body state of the poultry is abnormal or not and whether the temperature of the poultry is abnormal or not, sending the analysis result to the client, and transmitting a signal to the inspection system according to the analysis result;
the client is used for displaying data processing results and environmental information, and can realize remote control;
the inspection system is used for carrying the data acquisition system and the server, and can automatically inspect in the poultry house.
The data acquisition system comprises a high-definition camera, an infrared camera, a temperature sensor, a humidity sensor and a decibel sensor and is used for acquiring image data of poultry and environmental data of a poultry house;
the high-definition camera, the infrared camera, the temperature sensor, the humidity sensor and the decibel sensor are connected with the server in a wired connection mode;
the high-definition camera and the infrared camera are carried on the synchronous belt guide rail which is vertically arranged and can move in the vertical direction, so that shooting angles are ensured to be right against a poultry breeding cage;
the synchronous belt guide rail, the temperature sensor, the humidity sensor and the decibel sensor are mounted on a self-propelled inspection chassis 1 in the inspection system.
The image data acquisition module 2 comprises an RGB video camera and an infrared video camera which are connected to the synchronous belt guide rail of the lifting mechanism 4 through screws;
the environment data acquisition module 3 and the display 5 are arranged on the self-propelled inspection chassis 1 and connected with the server.
As a further limitation of the technical scheme, the server comprises a data processing module and a data transmission module, wherein the data processing module comprises an industrial personal computer, and the data transmission module comprises a communication module;
the industrial personal computer is placed in the self-propelled inspection chassis 1 and used for processing and analyzing poultry image data, and the communication module is used for data transmission among the server, the client and the inspection system;
the client comprises a visual interaction interface of a computer end and a mobile phone end, namely a portable display, wherein the visual interaction interface of the computer end and the mobile phone end is used for receiving image data of cage-reared poultry, environment data of a poultry house and information transmitted by a server, which are acquired by a display data acquisition module, by a user, and the user can control a patrol system through the visual interaction interface of the computer end and the mobile phone end.
The inspection system comprises a self-propelled inspection chassis 1, wherein the self-propelled inspection chassis 1 is connected with a bracket of a lifting mechanism 4 through a bolt, a control module, a server, a power module and a traveling module are arranged in the self-propelled inspection chassis 1, an environment data acquisition module 3 is arranged on the self-propelled inspection chassis 1, and an image data acquisition module 2 is connected with an L-shaped platform of the lifting mechanism 4 through a bolt;
the control module is used for controlling the whole work of the chassis;
the power module is used for supplying power to the data acquisition module, the server and the self-propelled inspection chassis;
the walking module comprises a direct current motor and a walking wheel.
The self-propelled inspection chassis 1 can automatically inspect in a poultry house without manual remote control, an RGB camera is arranged at the front end of the self-propelled inspection chassis 1, the camera shoots the advancing route of the chassis and transmits the advancing route to a server, the server performs path planning through an algorithm and then sends an instruction to a controller of the self-propelled inspection chassis 1 to control the advancing direction and the advancing route of the self-propelled inspection chassis 1.
The invention also provides a novel method: the method is characterized in that: the method comprises the following steps:
s1, a data acquisition system acquires RGB image data and infrared thermal imaging image data of poultry and transmits the RGB image data and the infrared thermal imaging image data to a server;
s2, processing the RGB image data through a deep neural network in a server to position poultry;
and S3, disposing a poultry gesture recognition detection model and a poultry body temperature recognition detection model in the server, extracting characteristics of the region of interest, judging whether the gesture of the poultry is abnormal or not, and judging whether the body temperature of the poultry exceeds a normal range or not.
The deep neural network adopts an improved YOLOv8 algorithm, which is named as a C-YOLOv8 algorithm, and is used as an identification algorithm for detecting the posture, detecting the body temperature and positioning a camera of poultry, the algorithm is improved according to the detected part of the poultry, the detection precision and the detection speed are improved, and the network structure of the C-YOLOv8 algorithm is divided into: the C-YOLOv8 algorithm is mainly modified in the main network, and after three C2f modules in the main network of the YOLOv8 algorithm, an attention mechanism module CBAM is added, wherein the CBAM is used for connecting a channel attribute and a spatial attribute in series, so that the number of layers of the network is slightly increased, and the detection precision is improved.
The poultry gesture recognition detection model is used for carrying out chicken head anomaly scoring based on anomalies of all local parts, and the comprehensive scoring is formed by combining the inversion temperature and crowding degree of the cage-raised chickens, so that the comprehensive evaluation of the health of the cage-raised chickens is realized.
The camera positioning recognition model recognizes the central position of the cultivation cage according to the rectangular frame of the cultivation cage, and when the central position of the cultivation cage is at the central position of the image, the server sends a signal to the controller through the serial port to control the lifting mechanism to stop lifting.
The infrared image temperature extraction method comprises the following steps: acquiring a thermal imaging image, extracting temperature information of pixel points in the image, identifying an interested region according to an infrared image poultry identification algorithm model, outputting coordinate information of the interested region, extracting the temperature of the interested region through coordinate conversion, filtering out the ambient temperature or abnormal low temperature of the interested region, extracting the highest temperature and the average temperature of the region in the interested region after filtering the ambient temperature, comparing with the set abnormal temperature, and judging whether the poultry body temperature is abnormal or not;
the calculation formula of the average body temperature of the region of interest is as follows:in the above, P W For average body temperature, T W In order to remove the total temperature of the pixel points after the ambient temperature, N is the number of the pixel points left after the ambient temperature is removed;
specifically, in step S1, the camera shoots the cage-raised poultry at different positions according to a preset time interval or according to a control instruction of a user, and sends the RGB image data and the infrared thermal imaging image data obtained by shooting to the server. The sensor monitors and records environmental parameters such as temperature, humidity, ammonia concentration and the like in the poultry house in real time, and sends the parameters to the server as environmental data.
In step S2, a deep neural network model is operated in the server, the model consists of a plurality of convolution layers, a pooling layer, a full connection layer and an activation function, the input RGB image data can be subjected to feature extraction and classification, and the position coordinates and the category labels of the poultry are output. The model adopts an improved YOLOv8 algorithm, and by adding the attention mechanism module CBAM behind the C2f module of the backbone network, the detection precision of poultry can be improved, the detection speed and accuracy are higher, different illumination conditions and poultry posture changes can be adapted, the robustness and reliability of the system are improved, and the modified network structure is shown in figure 6.
In step S3, two detection models, namely a poultry gesture recognition detection model and a poultry body temperature recognition detection model, are deployed in the server, and the poultry gesture recognition detection model is a classifier based on deep learning, and can perform feature extraction and classification on the region of interest in the input RGB image data, and output the gesture label of the poultry, such as normal, inverted, rolling and the like. The poultry body temperature identification detection model is a regression device based on deep learning, can perform feature extraction and regression on an interested region in input infrared thermal imaging image data, and outputs a poultry body temperature value. Judging whether the poultry has health abnormality according to the posture label and the body temperature value of the poultry, and sending the judging result to the client and the inspection system.
The poultry gesture recognition and detection model is established in the following manner:
the invention mainly takes the head of the poultry as a detection target, the temperature of the feathers of the poultry is lower than the temperature of the poultry, the real temperature cannot be detected through an infrared camera, the head and the legs of the poultry are not covered by the feathers, and the relatively real temperature can be detected through the infrared camera, but because the modern cultivation of the poultry adopts a cultivation method of a laminated cage, more than 6 poultry are usually arranged in one cultivation cage, the poultry is seriously overlapped, the legs of the poultry cannot be shot by the camera, the poultry head is selected as the detection target in the poultry temperature detection, and the poultry detection model establishment flow based on the C-YOLOv8 algorithm is shown in figure 2.
Taking poultry body temperature detection as an example, acquiring 1000 poultry original images in advance, marking an interested area in the original images by software, expanding the marked images by data through a data enhancement program, wherein the expansion is 20 times, the total number of the images is 20000, the images are used as image data sets for model training, the data sets are divided, and the dividing ratio of a training set, a verification set and a test set is 7:2:1, setting training parameters according to hardware configuration of a deep learning environment after setting up an operation environment of an algorithm, setting an iteration period to 150 times, and setting a learning rate to 0.001.
The average detection precision of the RGB image obtained after training is above 0.989, the detection result is shown in figure 6, the average detection precision of the thermal imaging image is above 0.980, and the detection result is shown in figure 7.
And (3) adopting a detection speed, an accuracy rate P, a recall rate R, F1 _cut and an average accuracy mean value mAP index to judge the performance of an algorithm on the test set.
Accuracy (Precision), i.e., the proportion of the total predicted positive class correctly predicted positive class, the accuracy P formula is:
the Recall (Recall), i.e., the proportional Recall R of all the actual positive classes correctly predicted to be positive, is given by:
in the above formula, TP represents the number of correct prediction on positive samples, FP represents the number of incorrect prediction on positive samples, and FN represents the number of incorrect prediction on negative samples;
f1_cut is a measure of classification, is a harmonic mean function of accuracy and recall, and is between 0,1, the larger the better the f1_cut formula is:
the average accuracy mean mAP index represents calculating the AP for all pictures in each class at different IoU thresholds and then averaging all classes.
The intelligent poultry health detection method and the inspection system work flow are as follows:
the self-propelled inspection chassis 1 is opened from the charging station, the first pavement is reached, the pavement is straight, the platform of the lifting mechanism 4 is at the bottommost part, when the cultivation cage is detected, the server sends out an instruction, the self-propelled inspection chassis 1 is controlled to stop moving, poultry starts to be detected, the server controls the lifting mechanism 4 to ascend, after the cultivation cage is detected, the lifting guide rail keeps high, the self-propelled inspection chassis 1 moves forward, when the cultivation cage is detected again, the self-propelled inspection chassis 1 stops moving again, the server controls the camera to perform detection from top to bottom, and the self-propelled inspection chassis automatically returns to the home for charging when the shortage of battery power is detected, and the detection is performed again after the charging is finished.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions may be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. An intelligent cage-rearing poultry inspection system which is characterized in that: the system comprises a data acquisition system, a lifting mechanism (4), a server, a client and a patrol system, wherein the data acquisition system comprises an image data acquisition module 2 and an environment data acquisition module (3), the image data acquisition module (2) is used for acquiring image data of cage-reared poultry, and the environment data acquisition module (3) is used for acquiring environment data of a poultry house and transmitting the data to the server;
the image data of the poultry in the cage comprises RGB image data and infrared thermal imaging image data of the poultry in the cage, and the environment data of the poultry house comprises temperature information, humidity information and loudness information of the environment of the poultry house;
the server is used for processing the image data and transmitting the data, analyzing whether the body state of the poultry is abnormal or not and whether the temperature of the poultry is abnormal or not, sending the analysis result to the client, and transmitting a signal to the inspection system according to the analysis result;
the client is used for displaying data processing results and environmental information, and can realize remote control;
the inspection system is used for carrying the data acquisition system and the server, and can automatically inspect in the poultry house.
2. An intelligent home-based poultry inspection system according to claim 1, wherein: the data acquisition system comprises a high-definition camera, an infrared camera, a temperature sensor, a humidity sensor and a decibel sensor and is used for acquiring image data of poultry and environmental data of a poultry house;
the high-definition camera, the infrared camera, the temperature sensor, the humidity sensor and the decibel sensor are connected with the server in a wired connection mode;
the high-definition camera and the infrared camera are carried on the synchronous belt guide rail which is vertically arranged and can move in the vertical direction, so that shooting angles are ensured to be right against a poultry breeding cage;
the synchronous belt guide rail, the temperature sensor, the humidity sensor and the decibel sensor are mounted on a self-propelled inspection chassis (1) in the inspection system.
3. An intelligent home-based poultry inspection system according to claim 2, wherein: the server comprises a data processing module and a data transmission module, wherein the data processing module comprises an industrial personal computer, and the data transmission module comprises a communication module;
the industrial personal computer is placed in the self-propelled inspection chassis (1) and used for processing and analyzing poultry image data, and the communication module is used for data transmission among the server, the client and the inspection system.
4. An intelligent home-based poultry inspection system according to claim 3, wherein: the client comprises a visual interaction interface of a computer end and a mobile phone end, namely a display (5), wherein the visual interaction interface of the computer end and the mobile phone end is used for receiving image data of cage-reared poultry, environment data of a poultry house and information transmitted by a server, which are acquired by a display data acquisition module, by a user, and the user can control a patrol system through the visual interaction interface of the computer end and the mobile phone end.
5. An intelligent home-based poultry inspection system according to claim 4, wherein: the inspection system comprises the self-propelled inspection chassis (1), wherein the self-propelled inspection chassis (1) is connected with a bracket of the lifting mechanism (4) through a bolt, a control module, a server, a power module and a walking module are arranged in the self-propelled inspection chassis (1), the environment data acquisition module (3) is arranged on the self-propelled inspection chassis (1), and the image data acquisition module (2) is connected with an L-shaped platform of the lifting mechanism (4) through a bolt;
the control module is used for controlling the whole work of the chassis;
the power module is used for supplying power to the data acquisition module, the server and the self-propelled inspection chassis;
the walking module comprises a direct current motor and a walking wheel.
6. The method for detecting an intelligent cage-rearing poultry inspection system according to claim 1, wherein: the method comprises the following steps:
s1, a data acquisition system acquires RGB image data and infrared thermal imaging image data of poultry and transmits the RGB image data and the infrared thermal imaging image data to a server;
s2, processing the RGB image data through a deep neural network in a server to position poultry;
and S3, disposing a poultry gesture recognition detection model and a poultry body temperature recognition detection model in the server, extracting characteristics of the region of interest, judging whether the gesture of the poultry is abnormal or not, and judging whether the body temperature of the poultry exceeds a normal range or not.
7. The method for detecting the intelligent cage-rearing poultry inspection system according to claim 6, wherein: the deep neural network adopts an improved YOLOv8 algorithm, which is named as a C-YOLOv8 algorithm, and is used as an identification algorithm for detecting the posture, detecting the body temperature and positioning a camera of poultry, the algorithm is improved according to the detected part of the poultry, the detection precision and the detection speed are improved, and the network structure of the C-YOLOv8 algorithm is divided into: the C-YOLOv8 algorithm is mainly modified in the main network, and after three C2f modules in the main network of the YOLOv8 algorithm, an attention mechanism module CBAM is added, wherein the CBAM is formed by connecting Channel Attention and Spatial Attention in series, the number of layers of the network is slightly increased, and the detection precision is improved.
8. The method for detecting an intelligent cage-rearing poultry inspection system according to claim 7, wherein: the poultry gesture recognition detection model is used for carrying out chicken head anomaly scoring based on anomalies of all local parts, and the comprehensive scoring is formed by combining the inversion temperature and crowding degree of the cage-raised chickens, so that the comprehensive evaluation of the health of the cage-raised chickens is realized.
9. The method for detecting an intelligent cage-rearing poultry inspection system according to claim 8, wherein: the camera positioning recognition model recognizes the central position of the cultivation cage according to the rectangular frame of the cultivation cage, and when the central position of the cultivation cage is at the central position of the image, the server sends a signal to the controller through the serial port to control the lifting mechanism to stop lifting.
10. The method for detecting an intelligent cage-rearing poultry inspection system according to claim 9, wherein: the infrared image temperature extraction method comprises the following steps: acquiring a thermal imaging image, extracting temperature information of pixel points in the image, identifying an interested region according to an infrared image poultry identification algorithm model, outputting coordinate information of the interested region, extracting the temperature of the interested region through coordinate conversion, filtering out the ambient temperature or abnormal low temperature of the interested region, extracting the highest temperature and the average temperature of the region in the interested region after filtering the ambient temperature, comparing with the set abnormal temperature, and judging whether the poultry body temperature is abnormal or not;
the calculation formula of the average body temperature of the region of interest is as follows:in the above, P W For average body temperature, T W In order to remove the total temperature of the pixels after the ambient temperature, N is the number of pixels left after the ambient temperature is removed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310568261.2A CN117029904A (en) | 2023-05-19 | 2023-05-19 | Intelligent cage-rearing poultry inspection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310568261.2A CN117029904A (en) | 2023-05-19 | 2023-05-19 | Intelligent cage-rearing poultry inspection system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117029904A true CN117029904A (en) | 2023-11-10 |
Family
ID=88625051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310568261.2A Pending CN117029904A (en) | 2023-05-19 | 2023-05-19 | Intelligent cage-rearing poultry inspection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117029904A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117958800A (en) * | 2024-04-01 | 2024-05-03 | 四川省畜牧科学研究院 | Weighing type pig only height measuring device |
-
2023
- 2023-05-19 CN CN202310568261.2A patent/CN117029904A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117958800A (en) * | 2024-04-01 | 2024-05-03 | 四川省畜牧科学研究院 | Weighing type pig only height measuring device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113223035B (en) | Intelligent inspection system for cage chickens | |
| JP7382899B2 (en) | Dead chicken detection system | |
| CN111709333B (en) | Tracing early warning system based on abnormal excrement of cage-raised chickens and health monitoring method | |
| CN114037552B (en) | Method and system for polling physiological growth information of meat ducks | |
| CN110612921A (en) | Monitoring system and method for positioning gilts | |
| CN109784200B (en) | Binocular vision-based cow behavior image acquisition and body condition intelligent monitoring system | |
| CN110728259A (en) | Chicken group weight monitoring system based on depth image | |
| CN114407051A (en) | Livestock and poultry farm inspection method and livestock and poultry farm robot | |
| CN117029904A (en) | Intelligent cage-rearing poultry inspection system | |
| CN111696139A (en) | System and method for estimating group weight of white feather breeding hens based on RGB image | |
| CN111767794A (en) | Cage-rearing poultry abnormal behavior detection method and detection system based on machine vision | |
| CN113762113A (en) | Livestock parturition behavior monitoring method and device | |
| CN114898405B (en) | Portable broiler chicken anomaly monitoring system based on edge calculation | |
| CN116824626A (en) | Artificial intelligent identification method for abnormal state of animal | |
| CN116019023A (en) | Electronic information management system and management method | |
| CN111833331A (en) | Livestock body temperature image processing method based on artificial intelligence and active gun-ball linkage | |
| CN111160271A (en) | Indoor chicken farm sick chicken detecting system based on deep learning | |
| CN114550334A (en) | Bridge robot inspection teaching training system, method and storage medium | |
| CN113724250A (en) | Animal target counting method based on double-optical camera | |
| CN114931112B (en) | Sow body ruler detection system based on intelligent inspection robot | |
| CN210247950U (en) | Intelligent detection device and intelligent detection system | |
| CN116277073A (en) | Chicken breeding inspection robot equipment, control system and method | |
| CN115226650B (en) | Sow oestrus state automatic detection system based on interaction characteristics | |
| CN211532340U (en) | Monitoring system for positioning fence sows | |
| CN111652084B (en) | Abnormal layer identification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |