WO2020110862A1 - Système de détection de signes de l'ensemble des maladies respiratoires bovines - Google Patents

Système de détection de signes de l'ensemble des maladies respiratoires bovines Download PDF

Info

Publication number
WO2020110862A1
WO2020110862A1 PCT/JP2019/045466 JP2019045466W WO2020110862A1 WO 2020110862 A1 WO2020110862 A1 WO 2020110862A1 JP 2019045466 W JP2019045466 W JP 2019045466W WO 2020110862 A1 WO2020110862 A1 WO 2020110862A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
brdc
learning
unit
cow
Prior art date
Application number
PCT/JP2019/045466
Other languages
English (en)
Japanese (ja)
Inventor
信行 古園井
洋一 木川
一正 岡田
達也 北原
成祥 板谷
Original Assignee
日東電工株式会社
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from JP2019205357A external-priority patent/JP7534041B2/ja
Application filed by 日東電工株式会社 filed Critical 日東電工株式会社
Priority to EP19891213.1A priority Critical patent/EP3888456A4/fr
Priority to US17/296,696 priority patent/US20220053737A1/en
Priority to AU2019387813A priority patent/AU2019387813B2/en
Priority to BR112021009480-6A priority patent/BR112021009480A2/pt
Publication of WO2020110862A1 publication Critical patent/WO2020110862A1/fr

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K29/00—Other apparatus for animal husbandry
    • 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining

Definitions

  • the present invention relates to a BRDC sign detection system.
  • the process from breeding to sale of livestock cattle is roughly divided into multiple processes (eg, grass breeding process, fattening process, processing process, etc.).
  • the Feedlot fostering place
  • BRDC Bovine Respiratory Disease Complex
  • Non-Patent Document 1 proposes a system for automatically detecting the cows actually infected with BRDC among the cows infected with BRDC. According to the system, it is possible to quickly identify a cow that has developed BRDC.
  • Cargill "Cargill brings facial recognition capability to farmers through strategic equity investment in Cainthus", [online], January 31, 2018, [July 31, 2018 Search], Internet (URL: https://www.cargill .com/2018/cargill-brings-facial-recognition-capability-to-farmers)
  • the purpose is to provide a BRDC sign detection system that detects a sign of BRDC.
  • a BRDC sign detection system comprises: Data showing the state of cattle that developed BRDC within the period required for fattening during a predetermined period of time that did not develop BRDC, and data showing the state of cattle that did not develop BRDC after the period required for fattening during the predetermined period of time; An acquisition unit that acquires and A machine learning unit that machine-learns the correspondence between the acquired data indicating the state in the predetermined period and information indicating whether or not BRDC has developed; Information indicating whether or not the new cow develops BRDC by inputting data indicating a state of the new cow in the predetermined period into a learned model generated by machine learning the correspondence. And an inference unit that infers the inference result and outputs the inference result.
  • FIG. 1 is a diagram showing an example of a system configuration of a BRDC sign detection system.
  • FIG. 2 is a diagram showing a sensor mounted on a cow and a position where the sensor is mounted.
  • FIG. 3 is a diagram illustrating an arrangement example of image pickup devices in a fattening farm.
  • FIG. 4 is a diagram illustrating an example of the hardware configuration of the detection device.
  • FIG. 5 is a sequence diagram showing a processing flow of the entire BRDC sign detection system.
  • FIG. 6 is a first diagram illustrating an example of functional configurations of the learning device and the detection device.
  • FIG. 7 is a diagram showing details of the experimental data processing unit.
  • FIG. 8A is a first diagram showing an example of the learning data stored in the learning data storage unit.
  • FIG. 8A is a first diagram showing an example of the learning data stored in the learning data storage unit.
  • FIG. 8B is a second diagram illustrating an example of the learning data stored in the learning data storage unit.
  • FIG. 9 is a first diagram showing details of the learning unit.
  • FIG. 10 is a first diagram showing an example of the sign detection data stored in the sign detection data storage unit.
  • FIG. 11 is a first diagram showing details of the sign detection unit.
  • FIG. 12 is a third diagram illustrating an example of the learning data stored in the learning data storage unit.
  • FIG. 13 is a diagram showing an example of the detection result screen.
  • FIG. 14 is a second diagram illustrating an example of the functional configurations of the learning device and the detection device.
  • FIG. 15A is a fourth diagram illustrating an example of the learning data stored in the learning data storage unit.
  • FIG. 15A is a fourth diagram illustrating an example of the learning data stored in the learning data storage unit.
  • FIG. 15B is a fifth diagram illustrating an example of learning data stored in the learning data storage unit.
  • FIG. 16 is a second diagram showing details of the learning unit.
  • FIG. 17 is a second diagram illustrating an example of the sign detection data stored in the sign detection data storage unit.
  • FIG. 18 is a second diagram showing details of the sign detection unit.
  • FIG. 19 is a sixth diagram illustrating an example of the learning data stored in the learning data storage unit.
  • FIG. 20 is a third diagram showing details of the learning unit.
  • FIG. 21 is a third diagram showing details of the sign detection unit.
  • FIG. 1 is a diagram showing an example of a system configuration of a BRDC sign detection system.
  • the BRDC sign detection system 100 is a system that detects a sign of BRDC onset of each cow in the fattening farm 140 in the fattening process.
  • the BRDC sign detection system 100 includes a detection device 110, a learning device 111, an imaging device group 130, a relay device 131 (and a sensor wirelessly connected to the relay device), and a terminal device 150. , And a terminal device 160.
  • the detection device 110 is communicatively connected to the imaging device group 130, the relay 131, the terminal device 150, and the terminal device 160 via the network 120.
  • the detection device 110 receives the image data taken by the imaging device group 130 and the measurement data measured by the sensor attached to the cow via the relay 131. Further, the detection device 110 detects a sign of BRDC onset of each cow based on the received image data and measurement data (hereinafter, referred to as Feedlot data).
  • the detection device 110 determines the treatment to be performed on the cow. As a result of the determination, in the case of performing the medication treatment on the cow, the medication instruction is output to the terminal device 150. Further, as a result of the determination, when the isolation treatment is performed on the cow, the isolation instruction is issued to the terminal device 160.
  • the learning device 111 generates a learned model used when the detection device 110 detects a sign of the onset of BRDC, and provides it to the detection device 110.
  • the learning device 111 collects the image data of the experimental cows in the experimental field 112 taken between the BRDC infection and the onset of BRDC, and the measured measurement data (hereinafter, referred to as experimental data) to perform machine learning.
  • a trained model is generated by performing.
  • the learning device 111 corrects the generated learned model of the cow in the feedlot 140 based on the Feedlot data collected by the detection device 110.
  • the imaging device group 130 is arranged around the fattening place 140 and photographs the cows in the fattening place 140. Further, the imaging device group 130 transmits the captured image data to the detection device 110 via the network 120.
  • the repeater 131 receives the measurement data measured by the sensor attached to the cow in the feedlot 140 by wireless communication, and transmits the received measurement data to the detection device 110 via the network 120.
  • the terminal device 150 is a terminal carried by a fattening manager who manages each cow in the fattening farm 140, and notifies the fattening manager when the detection device 110 gives a medication instruction. Thereby, the fattening manager identifies the cow for which the medication is instructed and administers the medication.
  • the terminal device 160 is a terminal carried by another fattening manager who manages each cow in the fattening farm 140, and notifies the other fattening manager when the detection device 110 issues an isolation instruction. Thereby, the other fattening manager identifies the cow for which the isolation instruction is given and transfers it from the fattening place 140 to the medical treatment place 182 to perform the isolation treatment.
  • FIG. 1 also shows devices other than the BRDC sign detection system 100 and the transfer of cattle other than the fattening process.
  • the terminal device 181 transmits the grazing data obtained by managing the cows grazing in the pasture 180 to the detection device 110 via the network 120.
  • Each of the cattle grazing on the pasture 180 is transferred to the fattening farm 140 when the grass breeding process is completed, and the fattening process is started on the fattening farm 140.
  • Each cow is checked at the time of transfer to the feedlot 140 and returned to the pasture 180 depending on the growing condition.
  • a cow infected with BRDC or developing BRDC is transferred to the medical treatment facility 182.
  • each cow whose fattening process has been started in the fattening farm 140 is transferred to the processor 183 when the period required for the fattening process (the period required for fattening) has elapsed and the fattening process is completed.
  • the terminal device 184 manages each cow transferred to the processor 183 after the fattening process is completed. Specifically, the terminal device 184 collects the data obtained until each cattle transferred is processed by the processor 183 and shipped, and transmits it to the detection device 110 as the processed data.
  • FIG. 2 is a diagram showing a sensor mounted on a cow and a position where the sensor is mounted.
  • the cows in the feedlot 140 are equipped with sensors having different sensor IDs for each cow.
  • the sensor attached to each cow includes multiple measurement element groups, and is classified into “acceleration”, "temperature”, and “voice” according to the measurement items. Further, the sensor attached to each cow has a measurement site determined for each measurement item, and a measurement element is attached to each measurement site.
  • 2b of FIG. 2 shows an example of mounting a plurality of measurement element groups included in the sensor.
  • the mounting tool 201 is mounted on the head of a cow and various measurements are performed on the head of the cow.
  • the “acceleration sensor 1” is mounted on the wearing tool 201.
  • the mounting tool 202 is mounted on the cow's neck to perform various measurements on the cow's neck.
  • the mounting tool 202 is equipped with a “temperature sensor”.
  • the mounting tool 203 is mounted on the abdomen of the cow and performs various measurements on the abdomen of the cow.
  • the wearing tool 203 is equipped with the “acceleration sensor 2” and the “voice sensor”.
  • the measurement data measured by the sensor is transmitted to the repeater 131 in a predetermined cycle in association with the sensor ID by a transmitter (not shown).
  • the detection device 110 can receive the measurement data (acceleration data 1, acceleration data 2, temperature data, voice data) associated with the sensor ID in a predetermined cycle.
  • FIG. 3 is a diagram illustrating an arrangement example of image pickup devices in a fattening farm.
  • the example of FIG. 3 illustrates a state in which five image pickup devices 310, 320, 330, 340, and 350 are arranged around the feeding area 140 as the image pickup device group 130.
  • the imaging devices 310, 320, 330, 340, and 350 are installed on the mounts 311, 321, 331, 341, and 351, respectively, so that all areas in the feedlot 140 can be photographed.
  • An imaging range in the width direction is defined.
  • the image pickup devices 310, 320, 330, 340, and 350 have an image pickup range in the height direction so that an image can be taken from the standing state to the recumbent state of the cow.
  • the imaging devices 310, 320, 330, 340, 350 may all be visible light cameras that detect visible light, or some of them may be infrared cameras that detect infrared light.
  • the mounts 311, 321, 331, 341, and 351 may be fixed mounts or movable mounts (mounts that move in the pan, tilt, and roll directions).
  • FIG. 4 is a diagram showing an example of the hardware configuration of the detection device.
  • the detection device 110 includes a CPU (Central Processing Unit) 401, a ROM (Read Only Memory) 402, and a RAM (Random Access Memory) 403.
  • the CPU 401, ROM 402, and RAM 403 form a so-called computer.
  • the detection device 110 also includes an auxiliary storage device 404, an operation device 405, a display device 406, a communication device 407, and a drive device 408.
  • the hardware of the detection device 110 is connected to each other via the bus 409.
  • the CPU 401 executes various programs installed in the auxiliary storage device 404 (for example, a sign detection program described later).
  • the ROM 402 is a non-volatile memory and functions as a main storage device.
  • the ROM 402 stores various programs and data necessary for the CPU 401 to execute various programs installed in the auxiliary storage device 404.
  • the ROM 402 stores a boot program such as BIOS (Basic Input/Output System) and EFI (Extensible Firmware Interface).
  • the RAM 403 is a volatile memory such as DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory), and functions as a main storage device.
  • the RAM 403 provides a work area in which various programs installed in the auxiliary storage device 404 are expanded when being executed by the CPU 401.
  • the auxiliary storage device 404 stores various programs, data collected by the various programs executed by the CPU 401, data generated by processing the collected data, and the like.
  • the sign detection data storage unit described later is realized in the auxiliary storage device 404.
  • the operation device 405 is an input device used by the administrator of the detection device 110 when inputting various instructions to the detection device 110.
  • the display device 406 is a display device that displays various information to the administrator of the detection device 110.
  • the communication device 407 is a communication device that is connected to the network 120 and communicates with each device (imaging device group 130, repeater 131, terminal device 150, terminal device 160, terminal device 181, terminal device 184, etc.).
  • the drive device 408 is a device for setting the recording medium 410.
  • the recording medium 410 referred to here includes a medium such as a CD-ROM, a flexible disk, a magneto-optical disk, etc. that records information optically, electrically or magnetically.
  • the recording medium 410 may include a semiconductor memory such as a ROM or a flash memory that electrically records information.
  • the various programs installed in the auxiliary storage device 404 are installed, for example, by setting the distributed recording medium 410 in the drive device 408 and reading the various programs recorded in the recording medium 410 by the drive device 408. To be done.
  • the various programs installed in the auxiliary storage device 404 may be installed by being downloaded via the network 120.
  • FIG. 5 is a sequence diagram showing a processing flow of the entire BRDC sign detection system.
  • the processing of the BRDC sign detection system 100 can be divided into a learning phase, a sign detection and learning phase, and a sign detection phase.
  • the learning phase is a phase in which the detection device 110 generates a learned model used when detecting a sign of the onset of BRDC.
  • the sign detection and learning phase is a phase in which the detection device 110 detects the sign of the onset of BRDC and corrects the learned model using the learned model. Further, the sign detection phase is a phase in which the detection device 110 detects the sign of the onset of BRDC using the modified learned model.
  • step S501 the learning device 111 starts monitoring cows in the experimental field 112 and collects experimental data.
  • the experimental system 112 also reproduces the same system as the fattening site 140. That is, the learning device 111 can collect the image data of the experimental cow and the measurement data measured by the sensor attached to the experimental cow as the experimental data.
  • step S502 the experimenter administers a virus or bacterium that causes BRDC to each experimental cow in the experimental site 112 to infect each experimental cow with BRDC.
  • step S503 the experimenter determines, from among the experimental cows infected with BRDC, the experimental cows that have developed BRDC and the experimental cows that have had severe BRDCs. Whether or not each experimental cow has developed BRDC and whether or not it has become severe is determined by, for example, whether or not the experimenter judges the daily clinical score and whether the clinical score exceeds a predetermined threshold value. It shall be based on.
  • cows that develop BRDC have symptoms such as decreased appetite, cough, runny nose, and high fever. Therefore, it is possible to determine that they have developed BRDC by combining these judgments.
  • the symptoms become more severe than when BRDC occurs, and symptoms such as weakness, cough, dyspnea, and extreme suppression of weight gain appear.
  • Determine that BRDC has become severe The determination result of whether or not each experimental cow has developed BRDC and the determination result of whether or not it has become severe are input to the learning device 111 in association with the experimental data.
  • step S504 the learning device 111 generates learning data (details will be described later using FIGS. 8A and 8B) using the experimental data and the determination result, and analyzes the learning data.
  • the learning device 111 analyzes the learning data by, for example, machine learning the learning model using the learning data.
  • step S505 the learning device 111 generates a learned model by analyzing the learning data.
  • step S506 the detection device 110 installs the learned model generated by the learning device 111.
  • the BRDC sign detection system 100 shifts to the sign detection and learning phase.
  • step S511 each cow is transferred from the grazing farm 180 to the fattening farm 140.
  • step S512 the fattening manager of the fattening farm 140 checks the condition of each cow transferred from the grazing farm 180.
  • step S513 the detection device 110 starts monitoring Feedlot data by starting monitoring each cow in the feedlot 140.
  • step S514 the fattening manager of the fattening farm 140 starts the fattening process for each cow in good condition.
  • step S515 the detection device 110 detects the sign of the onset of BRDC using the current Feedlot data for each cattle that has started the fattening process in the fattening farm 140.
  • step S5166 the fattening manager of the fattening plant 140 performs isolation treatment and medication treatment on the cow in which the sign of the onset of BRDC is detected.
  • step S517 the fattening process is terminated in the fattening farm 140 because the period required for the fattening process has elapsed.
  • step S5128 the fattening manager of the fattening farm 140 determines the state of each cow.
  • the detection device 110 uses the determination result determined by the fattening manager of the feedlot 140 and the sign detection data used for sign detection (details will be described later with reference to FIG. 10) for learning data ( Details will be described later using FIG. 12) and transmitted to the learning device 111.
  • step S519 the learning device 111 uses the learning data transmitted from the detection device 110 to machine-learn the learning model again. Thereby, the learning device 111 corrects the learned model.
  • step S520 the detection device 110 installs the learned model modified by the learning device 111.
  • the BRDC sign detection system 100 shifts to the sign detection phase.
  • step S532 the fattening manager of the fattening farm 140 checks the condition of each of the following cattle transferred from the grazing farm 180.
  • step S533 the detection device 110 starts monitoring Feedlot data by starting monitoring each cow in the feedlot 140.
  • the fattening manager of the fattening plant 140 starts the fattening process for each cow in good condition as a result of the check.
  • step S535 the detection device 110 detects the sign of the onset of BRDC for each cow whose fattening process has started in the feedlot 140, using the sign detection data generated based on the current Feedlot data.
  • step S536 the fattening manager of the fattening farm 140 performs isolation treatment and medication treatment on the cow in which the sign of the onset of BRDC is detected.
  • step S537 the fattening process is completed in the fattening farm 140 because the period required for the fattening process has elapsed.
  • FIG. 6 is a first diagram illustrating an example of functional configurations of the learning device and the detection device.
  • a learning program is installed in the learning device 111, and by executing the program, the learning device 111 includes the experiment data collecting unit 610, the experiment data processing unit 611, the learning data collecting unit 612, and the learning unit. Function as 613.
  • the experimental data collection unit 610 collects the image data and measurement data transmitted from the experimental site 112 as experimental data. In addition, the experimental data collection unit 610 associates the determination result of whether the experimental cow has developed BRDC and the determination result of whether the experimental cow has become severe (information indicating whether BRDC has been developed) with the experimental data. Attach and get.
  • the experimental data processing unit 611 is an example of a processing unit, and extracts the characteristic amount from the collected experimental data and analyzes the extracted characteristic amount to generate learning data for each individual.
  • the learning data collection unit 612 is an example of an acquisition unit. In the learning phase, the learning data collection unit 612 stores the learning data (including the corresponding determination result) generated by the experimental data processing unit 611 in the learning data storage unit 614.
  • the learning data collection unit 612 stores the learning data (including the corresponding determination result) generated by the detection device 110 in the learning data storage unit 614 in the sign detection and learning phases.
  • the learning unit 613 uses the learning data stored in the learning data storage unit 614 to perform machine learning on the learning model and generate a learned model.
  • the generated learned model is provided to the detection device 110 and installed in the sign detection unit 624.
  • the learning unit 613 uses the learning data newly stored in the learning data storage unit 614 to perform machine learning again on the learning model to obtain the learned model. Fix it.
  • the corrected learned model is provided again to the detection device 110 and installed in the sign detection unit 624.
  • the detection device 110 has a sign detection program installed.
  • the detection device 110 functions as the feed lot data collection unit 621, the feed lot data processing unit 622, the sign detection unit 624, and the first output unit 625 when the program is executed.
  • the feedlot data collection unit 621 collects the image data and measurement data transmitted from the feedlot 140 as Feedlot data.
  • the feed lot data processing unit 622 is an example of a processing unit.
  • the feature amount is extracted from the collected Feedlot data, and the extracted feature amount is analyzed to generate the sign detection data for each individual.
  • the data is stored in the storage unit 623.
  • the sign detection unit 624 has, for example, a learned model provided by the learning unit 613.
  • the sign detection unit 624 uses the sign detection data read from the sign detection data storage unit 623 to execute the learned model and detects the sign of the onset of BRDC.
  • the sign detection unit 624 notifies the first output unit 625 when the sign of the onset of BRDC is detected.
  • the first output unit 625 When the first detection unit 625 is notified by the sign detection unit 624 that the sign of the onset of BRDC has been detected, the first output unit 625 outputs information for identifying the cow and notifies the administrator of the detection device 110 of the onset of BRDC. The cow whose warning sign was detected is notified.
  • the detection device 110 ⁇ Sign detection data stored in the sign detection data storage unit 623; -The judgment result judged by the fattening manager of the fattening farm 140, Is used to generate learning data (details will be described later with reference to FIG. 12).
  • FIG. 7 is a diagram showing details of the experimental data processing unit.
  • the experimental data processing unit 611 includes a feature amount extraction unit 710, a behavior data analysis unit 720, a non-action data analysis unit 730, and a vital data analysis unit 740.
  • the feature amount extraction unit 710 extracts the feature amount from the image data (image data 1 to image data n) included in the experimental data. Further, the feature quantity extraction unit 710 extracts the feature quantity from the measurement data (acceleration data 1, acceleration data 2, temperature data, voice data) included in the experimental data.
  • the feature amount extracted by the feature amount extraction unit 710 includes various feature amounts.
  • position coordinates indicating the position of the feet of each cow can be cited.
  • walking or running
  • walking distance or the number of times of walking
  • mileage or the number of times of running
  • the behavior data analysis unit 720 generates behavior data 750 by analyzing the feature amount extracted by the feature amount extraction unit 710. As shown in FIG. 7, the action data 750 generated by the action data analysis unit 720 includes, for example, the number of feedings per day, the walking distance per day (or the number of walks), the traveling distance per day ( Or the number of times of travel) is included.
  • the behavior data 750 generated by the behavior data analysis unit 720 includes, for example, the number of standing rests (or time) per day, the number of standing ruminations per day, the number of recumbent rests per day ( Or hours) includes the number of recumbent ruminations per day.
  • the action data 750 generated by the action data analysis unit 720 includes, for example, the amount (or number of times) of drinking water per day, the amount (or number of times) of salt licking per day, the presence or absence of estrus, etc. Be done.
  • the non-action data analysis unit 730 generates the non-action data 760 by analyzing the feature amount extracted by the feature amount extraction unit 710. As shown in FIG. 7, the non-action data 760 generated by the non-action data analysis unit 730 includes, for example, the degree of inactivation.
  • the degree of inactivation is an index showing the degree to which the movement of cattle has decreased from the original movement.
  • the vital data analysis unit 740 generates vital data 770 by analyzing the feature amount extracted by the feature amount extraction unit 710. As shown in FIG. 7, the vital data 770 generated by the vital data analysis unit 740 includes, for example, the loudness of lung sound for each unit time, the body temperature for each unit time, and the respiratory rate for each unit time.
  • the vital data 770 generated by the vital data analysis unit 740 includes, for example, the presence or absence of nasal discharge and weight.
  • the experimental data processing unit 611 of the learning device 111 generates learning data using the behavior data 750, the non-behavior data 760, and the vital data 770 (data indicating the state of the cow), and stores the learning data in the learning data storage unit 614.
  • the learning data 800 includes “identification data” and “result data” as items of header information.
  • the example of FIG. 8A shows that “judgment result: BRDC onset (onset time: XX day), aggravation (severity time: YY day)” is recorded as result data.
  • the example of FIG. 8B shows that “judgment result: no BRDC onset” was recorded as result data.
  • the learning data storage unit 614 stores both the learning data 800 for cows that have developed BRDC and the learning data 800′ for cows that have not developed BRDC.
  • the learning data 800 includes “data item” and “time-series data” as main body information items.
  • Data item stores each data item of action data 750, non-action data 760, and vital data 770 generated by the experimental data processing unit 611.
  • Time-series data stores time-series data of each data item included in the action data 750, the non-action data 760, and the vital data 770 generated by the experimental data processing unit 611.
  • a broken line 850 shown in FIG. 8A overlaid on the “time series data” indicates the timing after a predetermined period (see arrow 840) has elapsed from the timing of infection with BRDC.
  • the learning unit 613 uses the time-series data within the predetermined period indicated by the arrow 840 among the time-series data when generating the learned model.
  • the time-series data before the onset of BRDC is used. If the time-series data after the onset of BRDC is used, the accuracy of predicting the onset of BRDC decreases before the onset of BRDC. This is because there is a concern that it will happen. In other words, by using the time series data before the onset of BRDC, the accuracy of predicting the onset of BRDC can be improved.
  • a broken line 850 superimposed on the “time series data” shows the timing after a predetermined period (see arrow 840) has elapsed from the timing of infection with BRDC.
  • the learning unit 613 which will be described later, generates time-series data within a predetermined period indicated by an arrow 840 among time-series data when generating a learned model. To use.
  • FIG. 8A and FIG. 8B show the learning data stored in the learning data storage unit 614 in the learning phase
  • the learning data stored in the learning data storage unit 614 in the sign detection and learning phase is also included. It is assumed to have the same configuration.
  • FIG. 9 is a first diagram showing details of the learning unit.
  • the learning unit 613 generates a learned model installed in the sign detection unit 624.
  • the learning unit 613 has a learning model 901 and a comparison changing unit 902.
  • the learning model 901 is an example of a machine learning unit, and is, for example, a convolutional neural network (CNN: Convolutional Neural Network)-based learning model.
  • CNN Convolutional Neural Network
  • the learning model 901 executes the process when the time-series data (predetermined period) of the learning data (for example, the learning data 800, 800′) read from the learning data storage unit 614 is input. As a result, the learning model 901 outputs the output result (BRDC onset, onset time, seriousness, seriousness time, or no BRDC onset) to the comparison change unit 902.
  • CNN Convolutional Neural Network
  • the comparison changing unit 902 calculates an error for the learning model 901 to perform machine learning. Specifically, the comparison changing unit 902 Output results output from the learning model 901 (BRDC onset, onset time, severity, severity time, or BRDC onset), A determination result (BRDC onset, onset time, severity, severity time, or BRDC onset) included in the item “result data” of the learning data read from the learning data storage unit 614; The error is calculated by comparing Further, the comparison changing unit 902 changes the model parameter in the learning model 901 based on the calculated error to perform machine learning on the learning model 901.
  • the learning unit 613 shown in FIG. 9 generates a learned model by performing machine learning on the learning model 901 using the learning data for all the experimental cows stored in the learning data storage unit 614. ..
  • the determination result is described to include the occurrence of BRDC, the onset time, the severity of the disease, the time of seriousness, or the occurrence of BRDC.
  • the determination result may include only one or more of them.
  • FIG. 10 is a first diagram showing an example of the sign detection data stored in the sign detection data storage unit. As shown in FIG. 10, the sign detection data 1001 includes “identification data” as an item of header information.
  • the sign detection data 1001 includes “data item” and “time-series data” as main body information items.
  • each data item of behavior data, non-action data, and vital data output from the feed lot data processing unit 622 is stored.
  • Time-series data stores time-series data of each data item included in action data, non-action data, and vital data output from the feed lot data processing unit 622.
  • a broken line 1050 shown in FIG. 10 overlaid on “time-series data” indicates the timing after a predetermined period (arrow 1040) has elapsed since the start of the fattening process at the fattening farm 140.
  • the predetermined period shown by the arrow 1040 in FIG. 10 is approximately the same length as the predetermined period shown by the arrow 840 in FIG. 8A.
  • the predetermined period (arrow 1040) is referred to as a sign monitoring period, and the time series data within the sign monitoring period is called sign monitoring data.
  • FIG. 11 is a first diagram showing details of the sign detection unit. As shown in FIG. 11, the sign detection unit 624 has a learned model 1101.
  • the learned model 1101 is an example of an inference unit, and is generated by the learning unit 613. Whether the learned model 1101 develops BRDC when the time series data (symptom monitoring period) of the sign detection data (for example, the sign detection data 1001) read from the sign detection data storage unit 623 is input. Information indicating whether or not it is inferred and the inference result is output.
  • the information indicating whether or not to develop BRDC includes information indicating that BRDC is developed and information indicating that BRDC is not developed, and one of them is output.
  • outputting information indicating that BRDC develops means that, for example, a node including "BRDC onset” or "BRDC onset, severe” in the output layer of the CNN configuring the learning model 901 in FIG. Is output as.
  • outputting information indicating that BRDC does not occur means that, for example, in the CNN output layer in FIG. 9, a node including “BRDC does not occur” is output as an inference result.
  • time series data of the sign detection data is input every predetermined period (for example, every 3 hours).
  • the learned model 1101 infers information indicating whether or not to develop BRDC every predetermined period, and outputs the inference result. That is, in the first embodiment, ⁇ "Detecting signs of BRDC onset", or ⁇ "Notify that a sign of the onset of BRDC has been detected", That is, the learned model 1101 outputs information indicating that BRDC develops as an inference result.
  • the detection device 110 In the sign detection and learning phase, the detection device 110 generates learning data using the sign detection data used for sign detection and the judgment result judged by the fattening manager after the fattening process. Then, the data is transmitted to the learning data collection unit 612. As a result, the learning data generated in the detection device 110 is stored in the learning data storage unit 614.
  • FIG. 13 is a diagram showing an example of the detection result screen.
  • the first output unit 625 generates the detection result screen 1300 and displays it on the display device 406 when the sign detection unit 624 notifies that the sign of BRDC onset has been detected.
  • the detection result screen 1300 includes a message indicating that the sign of the onset of BRDC has been detected.
  • the detection result screen 1300 includes a cow ID for specifying a cow in which a sign of the onset of BRDC is detected.
  • the detection result screen 1300 includes the date and time when the sign of the onset of BRDC was detected, the body temperature of the cow at the time of detection, the weight, and the like.
  • the detection result screen 1300 may include the current position of the cow in which the sign of the onset of BRDC is detected.
  • BRDC sign detection system 100 it is possible to detect the sign before the onset of BRDC. As a result, it is possible to reduce the loss associated with the disposal of cattle. This is because the absolute amount of cattle that become severe can be reduced by detecting and treating the disease at an early stage before the onset or seriousness.
  • the BRDC sign detection system 100 is Obtain data (time-series data of each data item) indicating the state of a cow that has developed BRDC within the period required for the fattening process during a predetermined period when BRDC has not developed.
  • data time series data of each data item
  • indicating the state of a cow that has not developed BRDC after a period required for the fattening process has passed during a predetermined period is acquired.
  • Machine learning is performed on the correspondence relationship between the acquired data indicating the state in the predetermined period and the information indicating whether or not BRDC has developed.
  • information indicating whether the new cow develops BRDC is displayed. It infers and outputs the inference result.
  • the sign of the onset of BRDC can be detected at an early stage when no BRDC has developed.
  • the learning model 901 is described as inputting time-series data.
  • the data input to the learning model 901 is not limited to time series data.
  • the processed data obtained by processing the time series data may be input.
  • the amount of change of the time series data from the reference data is calculated as the processed data, and the amount of change is input to the learning model 901.
  • time series data you may enter data other than time series data.
  • advance data data relating to breeding before the experimental cow is brought into the experimental field 112
  • the preliminary data is input to the learning model 901.
  • FIG. 14 is a second diagram illustrating an example of the functional configurations of the learning device and the detection device.
  • the learning device 111 of FIG. 14 also functions as the grazing data collection unit 1410. Further, in the case of the learning device 111 in FIG. 14, the functions of the learning data collection unit 1401 and the learning unit 1403 are different from the functions of the learning data collection unit 612 and the learning unit 613 shown in FIG.
  • the difference from FIG. 6 is that the detection device 110 of FIG. 14 also functions as a grazing data collection unit 1421, a fattening data generation unit 1425, and a second output unit 1426.
  • the function of the sign detection unit 1424 is different from the function of the sign detection unit 624 shown in FIG. 6.
  • the grazing data collection unit 1410 collects advance data (data relating to breeding before the experimental cows are brought into the experimental field 112) and stores the advance data in the learning data collection unit 1401.
  • the prior data means grazing data (genetic information, past medical history, date of birth of cattle, place of origin, etc.) obtained by managing cattle grazing in the pasture 180. Shall be pointed out.
  • the grazing data collection unit 1410 collects grazing data from the terminal device 181 via the network 120.
  • the learning data collection unit 1401 is an example of an acquisition unit.
  • the learning data collection unit 1401 stores the learning data (including the corresponding determination result) generated by the experimental data processing unit 611 in the learning data storage unit 1402.
  • the learning data collection unit 1401 reads the corresponding grazing data from the grazing data storage unit 1411 and includes the learning data in the learning data storage unit 1402.
  • the learning data collection unit 1401 uses the learning data (including the corresponding determination result and the corresponding grazing data) generated by the detection device 110 as the learning data storage unit. This is stored in 1402.
  • the learning unit 1403 performs machine learning on the learning model using the learning data stored in the learning data storage unit 1402, and generates a learned model.
  • the learning unit 1403 instead of the time series data, the amount of change in the time series data from the reference data is input to the learning model 901.
  • the grazing data is input to the learning model 901 in addition to the time series data.
  • the grazing data collection unit 1421 collects grazing data and stores it in the grazing data storage unit 1422, like the grazing data collection unit 1410 of the learning device 111. Note that, of the grazing data stored in the grazing data storage unit 1422, the grazing data of cattle whose fattening process has started at the feedlot 140 is included in the corresponding symptom detection data and the symptom detection data storage unit is included. 1423.
  • the sign detection unit 1424 has, for example, a learned model provided by the learning unit 1403.
  • the sign detection unit 1424 uses the sign detection data read from the sign detection data storage unit 1423 to execute the learned model and detects the sign of the onset of BRDC.
  • the sign detection unit 1424 instead of the time series data, the amount of change in the time series data from the reference data is input to the learned model.
  • grazing data is input to the learned model in addition to the time series data.
  • the fattening data generation unit 1425 reads, from the sign detection data storage unit 1423, sign detection data (sign detection data including grazing data) for cattle transferred to the processor 183 after the fattening process is completed. Generate fattening data.
  • the second output unit 1426 transmits the generated fattening data to the terminal device 184 of the processor 183.
  • the processor 183 can obtain fattening data of the cattle transferred from the fattening farm 140 and having completed the fattening process.
  • 15A and 15B are fourth and fifth diagrams showing an example of the learning data stored in the learning data storage unit.
  • the difference from the learning data 800 and 800′ shown in FIGS. 8A and 8B is that the learning data 1500 and 1500′ include “preliminary data” as an item of header information.
  • the pasture data is stored in the “preliminary data”.
  • FIG. 16 is a second diagram showing details of the learning unit.
  • the learning unit 1403 includes a preprocessing unit 1601, a learning model 1602, and a comparison changing unit 1603.
  • the preprocessing unit 1601 generates data to be input to the learning model 1602 based on the learning data.
  • the data input to the learning model 1602 is ⁇ Grazing data, ⁇ Amount of change in time series data for each data item (predetermined period), (See 1611).
  • the grazing data is extracted from the “preliminary data” item of the learning data (for example, the learning data 1500, 1500′) read from the learning data storage unit 1402.
  • the amount of change in the time-series data of each data item is data indicating the state of the cow, and is the learning data read from the learning data storage unit 1402 (for example, learning data 1500, 1500′). It is calculated based on the time series data of. Specifically, for each data item (“foraging”, “walking”,%) -Time series data within a predetermined period indicated by an arrow 840, -Reference data (for example, a representative value (for example, an average value) of time-series data within a predetermined period indicated by an arrow 820), The amount of change in the time-series data is calculated by calculating the difference.
  • the learning model 1602 is an example of a machine learning unit, and is, for example, a convolutional neural network (CNN: Convolutional Neural Network)-based learning model.
  • CNN Convolutional Neural Network
  • the learning model 1602 executes the processing by inputting the data (grazing data, the amount of change in the time series data (predetermined period)) generated by the preprocessing unit 1601. As a result, the learning model 1602 outputs the output result (BRDC onset, onset time, severity, severity, or no BRDC onset) to the comparison change unit 1603.
  • CNN Convolutional Neural Network
  • the comparison change unit 1603 calculates an error for the learning model 1602 to perform machine learning. Specifically, in the comparison change unit 1603, Output results output from the learning model 1602 (BRDC onset time, onset time, severity, severity time, or BRDC onset), A determination result (BRDC onset, onset time, severity, severity time, or BRDC onset) included in the item “result data” of the learning data read from the learning data storage unit 1402; The error is calculated by comparing Further, the comparison changing unit 1603 changes the model parameter in the learning model 1602 based on the calculated error to perform machine learning on the learning model 1602.
  • the learning unit 1403 illustrated in FIG. 16 generates a learned model by performing machine learning on the learning model 1602 using the learning data stored in the learning data storage unit 1402 for all experimental cows. ..
  • FIG. 17 is a second diagram illustrating an example of the sign detection data stored in the sign detection data storage unit.
  • the difference from the sign detection data 1001 shown in FIG. 10 is that “advance data” is included as an item of header information.
  • grazing data is stored in the “preliminary data”.
  • FIG. 18 is a second diagram showing details of the sign detection unit.
  • the sign detection unit 1424 includes a preprocessing unit 1801 and a learned model 1802.
  • the preprocessing unit 1801 generates data to be input to the learned model 1802 based on the sign detection data.
  • the data input to the trained model 1802 is ⁇ Grazing data, ⁇ Amount of change in time series data for each data item (prediction monitoring period), (See 1811).
  • the grazing data is extracted from the "advance data" item of the sign detection data (for example, sign detection data 1701) read from the sign detection data storage unit 1423.
  • the amount of change in the time-series data of each data item is calculated based on the time-series data of the sign detection data (for example, the sign detection data 1701) read from the sign detection data storage unit 1423. To do. Specifically, for each data item (“foraging”, “walking”,...) -Time series data within a predetermined period indicated by arrow 1040, Reference data (for example, a representative value (for example, an average value) of time-series data within a predetermined period indicated by an arrow 1020), The amount of change in the time-series data is calculated by calculating the difference.
  • the learned model 1802 is an example of an inference unit, and is generated by the learning unit 1403.
  • the learned model 1802 executes the processing by inputting the data (grazing data, change amount of time series data (prediction monitoring period)) generated by the preprocessing unit 1801. As a result, the learned model 1802 infers information indicating whether to develop BRDC and outputs the inference result.
  • the detection device 110 In the sign detection and learning phase, the detection device 110 generates learning data using the sign detection data used for sign detection and the judgment result judged by the fattening manager after the fattening process. Then, the data is transmitted to the learning data collection unit 1401. As a result, the learning data generated in the detection device 110 is stored in the learning data storage unit 1402.
  • the BRDC sign detection system 100 Obtain data (amount of change in time-series data of each data item) showing the state of a cow that has developed BRDC within the period required for the fattening process during a predetermined period in which BRDC has not developed.
  • data indicating the state of a cow that has not developed BRDC after a period required for the fattening process has elapsed during a predetermined period is acquired.
  • the sign of the onset of BRDC can be detected at an early stage when BRDC has not yet occurred.
  • the learning units 613 and 1403 have learning models 901 and 1602, and the learning model is used to perform machine learning to generate a learned model. .. Further, in the first and second embodiments, the sign detection units 624 and 1424 are described as executing the learned model using the sign detection data and outputting the inference result.
  • the individual data storage units are arranged in the learning units 613 and 1403, and the prior data and the time series data of each data item (or the change amount of the time series data of each data item) are stored. , Is accumulated for each individual in association with the determination result.
  • the sign detection units 624 and 1424 are ⁇ Newly acquired prior data and time series data of each data item (or change amount of time series data of each data item), ⁇ Preliminary data accumulated in the past and time series data of each data item (or change amount of time series data of each data item), And search for similar ones. Then, in the third embodiment, the sign detection units 624 and 1424 output the determination result associated with the search result as the inference result.
  • the sign detection units 624 and 1424 output the determination result associated with the search result as the inference result.
  • FIG. 20 is a third diagram showing details of the learning unit.
  • the learning unit 613 has an individual data storage processing unit 2001.
  • the individual data storage processing unit 2001 is an example of a storage unit, and stores the time series data of each data item read from the learning data storage unit 614 and the determination result in the individual data storage unit 2002 in association with each other. ..
  • FIG. 21 is a third diagram showing details of the sign detection unit. As shown in FIG. 21, the sign detection unit 624 includes a similarity determination unit 2101.
  • the similarity determination unit 2101 is an example of a search unit.
  • the similarity determination unit 2101 reads the sign detection data (for example, the sign detection data 1001) read from the sign detection data storage unit 623 from the time-series data of each data item stored in the individual data storage unit 2002.
  • the time series data of each data item similar to is searched.
  • searching time-series data of similar data items is equivalent to inferring information indicating whether or not BRDC develops.
  • the similarity determination unit 2101 outputs, as an inference result, a determination result (information indicating whether to develop BRDC) associated with the time-series data of each searched data item.
  • the BRDC sign detection system 100 is ⁇ Data (time series data of each data item or amount of change of time series data of each data item) showing the state of a cow that has developed BRDC within the period required for the fattening process in a predetermined period when BRDC has not been developed.
  • BRDC are stored in association with information indicating that BRDC has developed.
  • ⁇ Onset of BRDC data indicating the state of cows that have not developed BRDC after the period required for the fattening process has elapsed during a predetermined period (time series data of each data item or change amount of time series data of each data item). It is stored in association with the information indicating that it is not done.
  • learning data is generated as learning data for both cows that develop BRDC (or experimental cows) and cows that do not develop BRDC (or experimental cows). Described as a thing.
  • the target for generating the learning data is not limited to this, and for example, the learning data may be generated only for the cow (or the experimental cow) that did not develop BRDC.
  • the sign detection unit 624 in the first and second embodiments, when the information indicating that BRDC is not developed is not output from the learned model 1101, it indicates that BRDC is developed. Information is output as an inference result.
  • the individual data storage unit 2002 is searched, and when the data is not similar to any data, information indicating that BRDC is developed is output as an inference result.
  • the learning data may be divided into a plurality of groups similar to each other, and machine learning may be performed for each group to generate a learned model.
  • the learning data of the group used when generating each learned model is associated with the learned model, and the sign detection units 624 and 1424 are associated with each other. To store.
  • the sign detection units 624 and 1424 search for learning data similar to the sign detection data read from the sign detection data storage units 623 and 1423, and execute the learned model associated with the similar learning data. By doing so, the sign of the onset of BRDC is detected. Note that the sign detection units 624 and 1424 compare time-series data of a predetermined period when searching for learning data similar to the sign detection data.
  • the grazing data (genetic information, past medical history, date of birth of cattle, place of origin, etc.) is recorded by including "advance data" in the item of header information of learning data.
  • data other than grazing data may be recorded in the "preliminary data" as the preliminary data.
  • the machine learning when machine learning is performed on the learning model 901, the machine learning is performed including the onset time and the aggravation time, but the onset time and the aggravation time are excluded.
  • Machine learning may be performed.
  • the machine learning may be performed after replacing the onset time and the aggravation time with a time range having a predetermined length.
  • isolation treatment or medication is performed. It has been described as a treatment.
  • the information indicating the onset of BRDC is output as the inference result, neither the medication treatment nor the isolation treatment may be performed if the seriousness does not occur. This makes it possible to administer only to severely ill cows that have been administered so far and not to those that do not require medication. As a result, the effect that the medication cost can be reduced can be obtained.
  • the learning device and the detection device are configured as separate bodies, but the learning device and the detection device may be configured as one body.
  • the head, neck, and abdomen are cited as the measurement sites of the sensor, and the sensor is attached to the measurement site, but the sensors are attached to other measurement sites. You may. Further, the method of mounting on each measurement site is also arbitrary.
  • the experimental data collecting unit 610 or the feedlot data collecting unit 621 collects image data, acceleration data, temperature data, and audio data as experimental data or Feedlot data. explained.
  • the data collected by the experimental data collection unit 610 or the feedlot data collection unit 621 is not limited to image data, acceleration data, temperature data, and audio data, and may be other data.
  • the experimental data processing unit 611 or the feedlot data processing unit 622 has been described as generating behavior data, non-behavior data, and vital data.
  • the data generated by the experimental data processing unit 611 or the feedlot data processing unit 622 is not limited to the action data, the non-action data, and the vital data, and may be other data.
  • BRDC sign detection system 110 Detection device 111: Learning device 130: Imaging device group 131: Repeater 140: Feedlot 150, 160: Terminal device 610: Experimental data collection unit 611: Experimental data processing unit 612: Learning data Collection unit 613: Learning unit 621: Feedlot data collection unit 622: Feedlot data processing unit 624: Sign detection unit 625: First output unit 710: Feature amount extraction unit 720: Behavior data analysis unit 730: Non-action data analysis unit 740: Vital data analysis unit 800, 800': Learning data 901: Learning model 902: Comparison change unit 1001: Prediction detection data 1101: Learned model 1300: Detection result screen 1403: Learning unit 1424: Prediction detection unit 1601: Pre-processing unit 1801: Pre-processing unit 1802: Learned model

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Business, Economics & Management (AREA)
  • Animal Husbandry (AREA)
  • Environmental Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Primary Health Care (AREA)
  • Mining & Mineral Resources (AREA)
  • Agronomy & Crop Science (AREA)
  • Health & Medical Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

L'invention fournit un système de détection de signes de l'ensemble des maladies respiratoires bovines qui détecte les signes de l'ensemble des maladies respiratoires bovines. Ce système de détection de signes de l'ensemble des maladies respiratoires bovines est caractéristique en ce qu'il présente : une unité acquisition qui acquiert d'une part des données indiquant, pendant une période prédéfinie au cours de laquelle l'ensemble des maladies respiratoires bovines ne s'est pas développé, l'état d'un bovin ayant développé l'ensemble des maladies respiratoires bovines à l'intérieur de la période nécessaire à l'engraissement, et d'autre part des données indiquant, pendant une période prédéfinie, l'état d'un bovin n'ayant pas développé l'ensemble des maladies respiratoires bovines après écoulement de la période nécessaire à l'engraissement ; une unité apprentissage automatique qui assure un apprentissage automatique des correspondances entre lesdites données indiquant l'état d'un bovin pendant la période prédéfinie ainsi acquises, et des informations indiquant le développement ou non de l'ensemble des maladies respiratoires bovines ; et une unité inférence qui entre des données indiquant l'état d'un nouveau bovin pendant ladite période prédéfinie dans un modèle appris généré par l'apprentissage automatique de ladite correspondance, qui infère des informations indiquant si le nouveau bovin développe ou non l'ensemble des maladies respiratoires bovines, et qui émet en sortie des résultats d'inférence.
PCT/JP2019/045466 2018-11-29 2019-11-20 Système de détection de signes de l'ensemble des maladies respiratoires bovines WO2020110862A1 (fr)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP19891213.1A EP3888456A4 (fr) 2018-11-29 2019-11-20 Système de détection de signes de l'ensemble des maladies respiratoires bovines
US17/296,696 US20220053737A1 (en) 2018-11-29 2019-11-20 Brdc sign detecting system
AU2019387813A AU2019387813B2 (en) 2018-11-29 2019-11-20 BRDC sign detection system
BR112021009480-6A BR112021009480A2 (pt) 2018-11-29 2019-11-20 sistema de detecção de sinal de brdc

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2018223742 2018-11-29
JP2018-223742 2018-11-29
JP2019205357A JP7534041B2 (ja) 2018-11-29 2019-11-13 Ｂｒｄｃ予兆検知システム
JP2019-205357 2019-11-13

Publications (1)

Publication Number Publication Date
WO2020110862A1 true WO2020110862A1 (fr) 2020-06-04

Family

ID=70853210

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2019/045466 WO2020110862A1 (fr) 2018-11-29 2019-11-20 Système de détection de signes de l'ensemble des maladies respiratoires bovines

Country Status (1)

Country Link
WO (1) WO2020110862A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016171077A1 (fr) * 2015-04-24 2016-10-27 Ｊｃアライアンス株式会社 Système de traitement d'informations
JP2018023377A (ja) * 2016-08-08 2018-02-15 板垣 和子 人工哺乳牛用乳首、および人工哺乳牛用嚥下補助具
WO2018105222A1 (fr) * 2016-12-08 2018-06-14 ソニー株式会社 Dispositif de commande d'affichage, procédé de commande d'affichage et programme
JP2018170969A (ja) * 2017-03-31 2018-11-08 Ｎｔｔテクノクロス株式会社 行動特定装置、行動特定方法、及びプログラム
JP2019205357A (ja) 2018-05-28 2019-12-05 日清フーズ株式会社 ゲル状ソース

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016171077A1 (fr) * 2015-04-24 2016-10-27 Ｊｃアライアンス株式会社 Système de traitement d'informations
JP2018023377A (ja) * 2016-08-08 2018-02-15 板垣 和子 人工哺乳牛用乳首、および人工哺乳牛用嚥下補助具
WO2018105222A1 (fr) * 2016-12-08 2018-06-14 ソニー株式会社 Dispositif de commande d'affichage, procédé de commande d'affichage et programme
JP2018170969A (ja) * 2017-03-31 2018-11-08 Ｎｔｔテクノクロス株式会社 行動特定装置、行動特定方法、及びプログラム
JP2019205357A (ja) 2018-05-28 2019-12-05 日清フーズ株式会社 ゲル状ソース

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CARGILL, CARGILL BRINGS FACIAL RECOGNITION CAPABILITY TO FARMERS THROUGH STRATEGIC EQUITY INVESTMENT IN CAINTHUS, 31 January 2018 (2018-01-31), Retrieved from the Internet <URL:https://www.cargill.com/2018/cargill-brings-facial-recognition-capability-to-farmers>

Similar Documents

Publication Publication Date Title
Van Hertem et al. Lameness detection in dairy cattle: Single predictor v. multivariate analysis of image-based posture processing and behaviour and performance sensing
Peng et al. Dam behavior patterns in Japanese black beef cattle prior to calving: Automated detection using LSTM-RNN
WO2016171077A1 (fr) Système de traitement d&#39;informations
WO2011120529A1 (fr) Modèle destiné à classer l&#39;activité d&#39;un animal
Bowen et al. Early prediction of respiratory disease in preweaning dairy calves using feeding and activity behaviors
Terrell et al. Incidence of lameness and association of cause and severity of lameness on the outcome for cattle on six commercial beef feedlots
US20220248641A1 (en) Information provision device, information provision method, and storage medium
Ayadi et al. Dairy cow rumination detection: A deep learning approach
Smith et al. Development of a model to simulate infection dynamics of Mycobacterium bovis in cattle herds in the United States
US20210400920A1 (en) Livestock stillbirthing alerting system
Yaseer et al. A review of sensors and Machine Learning in animal farming
Dittrich et al. Variable selection for monitoring sickness behavior in lactating dairy cattle with the application of control charts
Mei et al. Identification of aflatoxin-poisoned broilers based on accelerometer and machine learning
Liakos et al. Machine Learning Based Computational Analysis Method for Cattle Lameness Prediction.
JP7092624B2 (ja) 行動特定装置、行動特定方法及びプログラム
WO2020110862A1 (fr) Système de détection de signes de l&#39;ensemble des maladies respiratoires bovines
AU2019387813B2 (en) BRDC sign detection system
Dalton et al. Validation of HOBO Pendant® data loggers for automated step detection in two age classes of male turkeys: growers and finishers
JP7410200B2 (ja) 特定装置、特定方法、及びプログラム
WO2021255898A1 (fr) Dispositif de traitement d&#39;informations, procédé de traitement d&#39;informations, système de traitement d&#39;informations et support de stockage lisible par ordinateur
Barwick et al. On-animal motion sensing using accelerometers as a tool for monitoring sheep behaviour and health status
Thakur et al. Digitalization of livestock farms through blockchain, big data, artificial intelligence, and Internet of Things
Pirinen et al. Validation of a tail-mounted triaxial accelerometer for measuring foals' lying and motor behavior
Panda et al. Development of a Decision Support System for Animal Health Management Using Geo-Information Technology: A Novel Approach to Precision Livestock Management
Dissanayakea et al. Evaluating ROCKET and Catch22 features for calf behaviour classification from accelerometer data using Machine Learning models

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19891213

Country of ref document: EP

Kind code of ref document: A1

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112021009480

Country of ref document: BR

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019387813

Country of ref document: AU

Date of ref document: 20191120

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2019891213

Country of ref document: EP

Effective date: 20210629

ENP Entry into the national phase

Ref document number: 112021009480

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20210514