CN115250950B - Method and system for inspecting livestock and poultry pig farm based on artificial intelligence - Google Patents
Method and system for inspecting livestock and poultry pig farm based on artificial intelligence Download PDFInfo
- Publication number
- CN115250950B CN115250950B CN202210923788.8A CN202210923788A CN115250950B CN 115250950 B CN115250950 B CN 115250950B CN 202210923788 A CN202210923788 A CN 202210923788A CN 115250950 B CN115250950 B CN 115250950B
- Authority
- CN
- China
- Prior art keywords
- animal
- determining
- audio
- unit
- culture
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 27
- 244000144972 livestock Species 0.000 title claims abstract description 25
- 244000144977 poultry Species 0.000 title claims abstract description 25
- 241001465754 Metazoa Species 0.000 claims abstract description 135
- 238000012544 monitoring process Methods 0.000 claims abstract description 102
- 238000007689 inspection Methods 0.000 claims abstract description 71
- 238000005070 sampling Methods 0.000 claims abstract description 54
- 238000009434 installation Methods 0.000 claims abstract description 34
- 238000009395 breeding Methods 0.000 claims abstract description 16
- 230000001488 breeding effect Effects 0.000 claims abstract description 16
- 230000002159 abnormal effect Effects 0.000 claims description 25
- 238000012216 screening Methods 0.000 claims description 15
- 230000007613 environmental effect Effects 0.000 claims description 11
- 238000006073 displacement reaction Methods 0.000 claims description 10
- 230000001133 acceleration Effects 0.000 claims description 7
- 238000002255 vaccination Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000009360 aquaculture Methods 0.000 abstract 1
- 244000144974 aquaculture Species 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 10
- 235000015277 pork Nutrition 0.000 description 7
- 241000282887 Suidae Species 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009313 farming Methods 0.000 description 2
- 210000000577 adipose tissue Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003863 physical function Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000007958 sleep Effects 0.000 description 1
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
- A01K29/005—Monitoring or measuring activity, e.g. detecting heat or mating
-
- 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
- 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
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/26—Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Environmental Sciences (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Biodiversity & Conservation Biology (AREA)
- General Physics & Mathematics (AREA)
- Animal Husbandry (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Image Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of intelligent breeding inspection, and particularly discloses an artificial intelligence-based livestock and poultry pig farm inspection method and system, wherein the method comprises the steps of receiving image information acquired by a monitoring end and marking an aquaculture unit according to the image information and the installation position of the monitoring end; a statistically marked breeding unit for determining sampling points; when the motion end moves to the sampling point, audio information is queried and played in a preset audio library, feedback audio of the animal is obtained, and whether a close-range image is acquired or not is judged based on the feedback audio. According to the invention, the animal image is acquired through the traditional monitoring camera, the animal image is intelligently identified, so that the inspection instruction of the moving end is determined, the moving end interacts with the animal in the inspection process through audio information, whether the close-range image is acquired or not is determined according to feedback audio of the animal, the intelligent identification process is additionally arranged on the original monitoring architecture, and the working pressure of staff is greatly reduced.
Description
Technical Field
The invention relates to the technical field of intelligent breeding inspection, in particular to an artificial intelligence-based inspection method and an artificial intelligence-based inspection system for livestock and poultry pig farms.
Background
The pig industry is one of important industries indispensable in modern agriculture in China, plays a very important role in guaranteeing the safe supply of meat foods, and is being changed from the traditional pig industry to the modern pig industry.
The most obvious characteristic of the modern pig industry is unmanned management, and a large number of pork pigs can be cultivated remotely by only a plurality of workers and a plurality of intelligent devices; in the process, automatic feeding machines, automatic cleaning machines and the like are involved, and the intelligent machines can greatly reduce the workload of staff; however, when the number of pork pigs is large, the staff still needs to check the status of the pork pigs at regular and high frequency; if a remote monitoring method can be provided, the workload of workers can be further reduced.
The existing remote monitoring method is mainly based on a monitoring system framework of a camera, and the mode is that special staff monitors in real time, because the abnormal reaction of the pork pig is sometimes quite insignificant, for example, the pork pig is prone to be motionless in a certain place for a long time, the staff cannot quickly judge whether the pork pig is abnormal or not, and when the number of videos is large, the staff can hardly find the abnormal state of the pork pig in the first time; therefore, how to reduce the monitoring pressure of the staff and improve the supervision efficiency of the staff is a technical problem to be solved by the technical scheme of the invention.
Disclosure of Invention
The invention aims to provide an artificial intelligence-based inspection method and system for livestock and poultry pig farms, which are used for solving the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions:
an artificial intelligence-based inspection method for livestock and poultry pig farms, comprising the following steps:
acquiring a culture model containing regional labels, and determining a patrol path of a moving end and the installation position of a monitoring end based on the culture model; the regional tag comprises a passing region and a breeding unit;
receiving image information acquired by a monitoring end, and marking a culture unit according to the image information and the installation position of the monitoring end;
the statistical marked cultivation units determine sampling points based on the position relation between the cultivation units and the inspection path, and the inspection path containing the sampling points is sent to the moving end;
when the motion end moves to a sampling point, inquiring and playing audio information in a preset audio library, acquiring feedback audio of animals, and acquiring a close-range image according to the feedback audio;
the audio library comprises a tag item and an audio item, wherein the tag item is used for representing the position information of the culture unit; the audio item is the sound of the animal in the pre-collected breeding unit.
As a further scheme of the invention: the step of obtaining the culture model containing the regional tag and determining the patrol path of the moving end and the installation position of the monitoring end based on the culture model comprises the following steps:
establishing a connection channel with a drawing database, reading a cultivation model containing cultivation units, and determining a passing area according to the cultivation units;
establishing a connection channel with a culture record database, and inquiring the culture state of each culture unit; the culture state is used for representing the physical sign parameters of the cultured animals, and the physical sign parameters at least comprise age and vaccination information;
determining the stability level of each animal based on the physical sign parameters, and determining the installation position of the monitoring end according to the stability level;
and determining the routing inspection path of the moving end according to the passing area.
As a further scheme of the invention: the step of determining the stability level of each animal based on the physical sign parameters and determining the installation position of the monitoring end according to the stability level comprises the following steps:
inquiring the stability level of each animal in a preset level table based on the sign parameters;
determining monitoring definition according to the stability level, and determining the height of a monitoring end according to the monitoring definition;
determining a monitoring range of the monitoring end according to the height of the monitoring end and the wide angle of the monitoring end, and determining the installation position of the monitoring end according to the monitoring range;
wherein the union of the monitoring ranges is greater than the union of the culture units.
As a further scheme of the invention: the step of receiving the image information acquired by the monitoring end and marking the culture unit according to the image information and the installation position of the monitoring end comprises the following steps:
receiving image information containing time information obtained in real time by a monitoring terminal, and sequencing the image information according to the time information to obtain an image group;
sequentially comparing adjacent images in the image group, calculating the coincidence ratio, and screening the image group according to the coincidence ratio;
performing contour recognition on the screened image group to obtain animal contours and obtain center points of the animal contours;
determining the motion trail of each animal according to the center point, and marking the corresponding culture unit according to the motion trail;
the rule for screening the image group according to the contact ratio is that when the contact ratio is the same, deleting the later image; when the overlap ratio is different, both images remain.
As a further scheme of the invention: the step of determining the motion trail of each animal according to the center point and marking the corresponding breeding unit according to the motion trail comprises the following steps:
sequentially reading the center points of animal outlines of adjacent images in the screened image group, and determining animal displacement according to a preset scale;
calculating animal speed and acceleration according to the animal displacement and the time information of the adjacent images, and generating a motion parameter table in a preset time period; the motion parameter table comprises a time item;
inputting the motion parameter table into a trained comparison model containing a historical motion parameter table, and calculating an abnormal value;
when the abnormal value reaches a preset abnormal threshold value, marking a culture unit corresponding to the animal outline; and when the abnormal value is smaller than a preset abnormal threshold value, updating the historical motion parameter table according to the motion parameter table.
As a further scheme of the invention: the step of determining sampling points by the culture units of the statistical marks based on the position relation between each culture unit and the inspection path and sending the inspection path containing the sampling points to the moving end comprises the following steps:
the statistical marked cultivation unit inquires a corresponding passage section of the cultivation unit in the passage area;
determining image acquisition points in the traffic segment according to a preset positioning rule;
marking sampling points corresponding to the image acquisition points in the inspection path, and sending the inspection path containing the sampling points to a moving end.
As a further scheme of the invention: when the motion end moves to a sampling point, inquiring and playing audio information in a preset audio library, acquiring feedback audio of animals, and acquiring close-range images according to the feedback audio, wherein the steps comprise:
when the motion end moves to a sampling point, collecting environmental audio;
inquiring at least one recorded audio of animals in the culture unit corresponding to the sampling point in a preset audio library, and randomly playing the recorded audio;
acquiring feedback audio of an animal, and correcting the feedback audio according to the environmental audio to obtain the animal audio;
inputting the animal audio into a trained identification model to generate an animal stable value, and collecting a close-range image when the stable value is smaller than a preset stable threshold value.
The technical scheme of the invention also provides an artificial intelligence-based inspection method and an artificial intelligence-based inspection system for the livestock and poultry pig farm, wherein the system comprises the following steps:
the port setting module is used for acquiring a culture model containing area labels, and determining a patrol path of the moving end and the installation position of the monitoring end based on the culture model; the regional tag comprises a passing region and a breeding unit;
the unit marking module is used for receiving the image information acquired by the monitoring end and marking the culture unit according to the image information and the installation position of the monitoring end;
the sampling point determining module is used for counting marked cultivation units, determining sampling points based on the position relation between each cultivation unit and the inspection path, and sending the inspection path containing the sampling points to the moving end;
the near-view image acquisition module is used for inquiring and playing audio information in a preset audio library when the motion end moves to a sampling point, acquiring feedback audio of animals and acquiring near-view images according to the feedback audio;
the audio library comprises a tag item and an audio item, wherein the tag item is used for representing the position information of the culture unit; the audio item is the sound of the animal in the pre-collected breeding unit.
As a further scheme of the invention: the port setting module includes:
the system comprises a passing area determining unit, a drawing database and a drawing database, wherein the passing area determining unit is used for establishing a connecting channel with the drawing database, reading a cultivation model containing a cultivation unit and determining a passing area according to the cultivation unit;
the culture state inquiry unit is used for establishing a connection channel with the culture record database and inquiring the culture state of each culture unit; the culture state is used for representing the physical sign parameters of the cultured animals, and the physical sign parameters at least comprise age and vaccination information;
the position determining unit is used for determining the stability level of each animal based on the physical sign parameters and determining the installation position of the monitoring end according to the stability level;
and the path determining unit is used for determining the routing inspection path of the moving end according to the traffic zone.
As a further scheme of the invention: the unit marking module includes:
the image ordering unit is used for receiving the image information containing the time information acquired by the monitoring terminal in real time, and ordering the image information according to the time information to obtain an image group;
the image screening unit is used for sequentially comparing adjacent images in the image group, calculating the contact ratio and screening the image group according to the contact ratio;
the contour analysis unit is used for carrying out contour recognition on the screened image group to obtain contours of all animals and obtain center points of the contours of all animals;
the execution unit is used for determining the motion trail of each animal according to the central point and marking the corresponding culture unit according to the motion trail;
the rule for screening the image group according to the contact ratio is that when the contact ratio is the same, deleting the later image; when the overlap ratio is different, both images remain.
Compared with the prior art, the invention has the beneficial effects that: according to the invention, the animal image is acquired through the traditional monitoring camera, the animal image is intelligently identified, so that the inspection instruction of the moving end is determined, the moving end interacts with the animal in the inspection process through audio information, whether the close-range image is acquired or not is determined according to feedback audio of the animal, the intelligent identification process is additionally arranged on the original monitoring architecture, and the working pressure of staff is greatly reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will briefly introduce the drawings that are needed in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the present invention.
Fig. 1 is a flow chart of an artificial intelligence-based inspection method for livestock and poultry pig farms.
Fig. 2 is a first sub-flow block diagram of an artificial intelligence based inspection method for livestock and poultry pig farms.
Fig. 3 is a second sub-flow block diagram of the livestock and poultry pig farm inspection method based on artificial intelligence.
Fig. 4 is a third sub-flowchart block diagram of the inspection method of the livestock and poultry pig farm based on artificial intelligence.
Fig. 5 is a fourth sub-flowchart block diagram of the inspection method of the livestock and poultry pig farm based on artificial intelligence.
Fig. 6 is a block diagram of the constitution of the inspection system of the livestock and poultry pig farm based on artificial intelligence.
Fig. 7 is a block diagram of the port setting module in the inspection system of the livestock and poultry pig farm based on artificial intelligence.
Fig. 8 is a block diagram of the constitution of the unit marking module in the inspection system of the livestock and poultry pig farm based on artificial intelligence.
Detailed Description
In order to make the technical problems, technical schemes and beneficial effects to be solved more clear, the invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Example 1
Fig. 1 is a flow chart of an artificial intelligence-based livestock and poultry pig farm inspection method, and in an embodiment of the invention, the method comprises steps S100 to S400:
step S100: acquiring a culture model containing regional labels, and determining a patrol path of a moving end and the installation position of a monitoring end based on the culture model; the regional tag comprises a passing region and a breeding unit;
the cultivation model is a building model of a cultivation area, and comprises cultivation units, wherein a passing area is arranged between the cultivation units for workers to pass through; the inspection process of the moving end also occurs in the passing area; for example, in existing centralized intelligent pig farming areas, each pig has a separate compartment, each compartment corresponding to a farming unit;
step S200: receiving image information acquired by a monitoring end, and marking a culture unit according to the image information and the installation position of the monitoring end;
the monitoring end can be a common camera, acquires image information according to the camera, performs content identification on the image information, can judge the state of the animal, and marks a corresponding culture unit according to the state of the animal;
step S300: the statistical marked cultivation units determine sampling points based on the position relation between the cultivation units and the inspection path, and the inspection path containing the sampling points is sent to the moving end;
determining a sampling point in the inspection path according to the position information of the culture unit, and further detecting animals at the near position when the moving end moves to the sampling point;
step S400: when the motion end moves to a sampling point, inquiring and playing audio information in a preset audio library, acquiring feedback audio of animals, and acquiring a close-range image according to the feedback audio;
when the motion end moves to the sampling point, sounds which are usually emitted by the corresponding animal, such as humming sounds emitted during eating, or sounds which are monitored in a comfortable state, are emitted, the motion end simulates the sounds, feedback audio of the animal can be obtained, and whether the animal is abnormal or not can be judged according to the feedback audio of the animal; for example, if the animal sounds a little but not when hearing sounds that are liable to cause stress, it can be recognized that the animal is indeed in an abnormal state, at this time, a close-range image needs to be acquired, and then the close-range image is sent to the artificial end for evaluation by the human.
The audio library comprises a tag item and an audio item, wherein the tag item is used for representing the position information of the culture unit; the audio item is the sound of animals in the pre-collected culture unit; the audio library is preset data.
Fig. 2 is a first sub-flowchart of an artificial intelligence-based inspection method for a livestock and poultry pig farm, wherein the step of obtaining an inspection model containing a regional tag and determining an inspection path of a moving end and an installation position of a monitoring end based on the inspection model includes steps S101 to S104:
step S101: establishing a connection channel with a drawing database, reading a cultivation model containing cultivation units, and determining a passing area according to the cultivation units;
step S102: establishing a connection channel with a culture record database, and inquiring the culture state of each culture unit; the culture state is used for representing the physical sign parameters of the cultured animals, and the physical sign parameters at least comprise age and vaccination information;
step S103: determining the stability level of each animal based on the physical sign parameters, and determining the installation position of the monitoring end according to the stability level;
step S104: and determining the routing inspection path of the moving end according to the passing area.
In the above description, it should be noted that the physical parameters may also have other parameters, such as body fat rate, which are easily obtained in the existing devices, and many devices for evaluating physical functions in the common gymnasium, and the accuracy of the measured results is up to 95%; it is possible for enterprises that can intelligently breed animals in batches to be equipped with such devices.
Further, the step of determining the stability level of each animal based on the sign parameter and determining the installation position of the monitoring end according to the stability level includes:
inquiring the stability level of each animal in a preset level table based on the sign parameters;
determining monitoring definition according to the stability level, and determining the height of a monitoring end according to the monitoring definition;
determining a monitoring range of the monitoring end according to the height of the monitoring end and the wide angle of the monitoring end, and determining the installation position of the monitoring end according to the monitoring range;
wherein the union of the monitoring ranges is greater than the union of the culture units.
The stability level of each animal can be determined according to the physical sign parameters, the stability level is colloquially understood to be whether the animal is good or not, and for the animal with good body, the height of the monitoring end can be set higher, so that one monitoring end can monitor a plurality of culture units at the same time; for animals with weaker physical quality, the height of the monitoring end needs to be set lower, and the acquired image information has higher definition on the premise of fixed performance of the monitoring end; the installation requirement is that the sum of the monitoring ranges is larger than the sum of the corresponding areas of the culture units, and the areas where repeated monitoring can occur cannot occur.
Fig. 3 is a second sub-flowchart of the inspection method for the livestock and poultry pig farm based on artificial intelligence, wherein the steps of receiving the image information acquired by the monitoring terminal and marking the cultivation unit according to the image information and the installation position of the monitoring terminal include steps S201 to S204:
step S201: receiving image information containing time information obtained in real time by a monitoring terminal, and sequencing the image information according to the time information to obtain an image group;
in the data transmission process, the monitoring end acquires the image information and then transmits the image information to the system, and in the process, the arrangement sequence of the image information can be changed due to the influence of transmission parameters, so that the system needs to sort when receiving the image information;
step S202: sequentially comparing adjacent images in the image group, calculating the coincidence ratio, and screening the image group according to the coincidence ratio;
the image information in the image group is repeated in many cases, for example, the image information is almost repeated when an animal sleeps, and only one image is needed to be reserved, so that the contact ratio needs to be calculated, and then the images in the image group are subjected to reject screening according to the contact ratio; the coincidence degree calculating process is an image comparing process;
step S203: performing contour recognition on the screened image group to obtain animal contours and obtain center points of the animal contours;
performing contour recognition on each image in the image group, and positioning animal contours; if the monitoring end has a temperature acquisition function, the process is simpler; processing the animal outline to determine a center point of the animal outline; the center point is a superior concept that is determined by a preset rule, for example, a side having the longest distance is determined in a certain direction, and then a center point is determined on the side according to the total pixel point and the side length.
Step S204: determining the motion trail of each animal according to the center point, and marking the corresponding culture unit according to the motion trail;
analyzing the change process of the center point, combining the related time information, determining the motion trail, judging whether the animal has a problem according to the motion trail, and marking the animal if the animal has the problem.
The rule for screening the image group according to the contact ratio is that when the contact ratio is the same, deleting the later image; when the overlap ratio is different, both images remain.
As a preferred embodiment of the present invention, the step of determining the motion trail of each animal according to the center point, and marking the corresponding breeding unit according to the motion trail includes:
sequentially reading the center points of animal outlines of adjacent images in the screened image group, and determining animal displacement according to a preset scale;
calculating animal speed and acceleration according to the animal displacement and the time information of the adjacent images, and generating a motion parameter table in a preset time period; the motion parameter table comprises a time item;
inputting the motion parameter table into a trained comparison model containing a historical motion parameter table, and calculating an abnormal value;
when the abnormal value reaches a preset abnormal threshold value, marking a culture unit corresponding to the animal outline; and when the abnormal value is smaller than a preset abnormal threshold value, updating the historical motion parameter table according to the motion parameter table.
The specific parameters of the motion trail are described, and the core principle is that animal displacement is determined according to a central point, speed and acceleration are determined according to object displacement and time information, the speed and the acceleration are compared with historical information, and whether the animal has a problem is determined according to a comparison result.
Specifically, the comparison process may include total time of day of activity, total distance of activity, extreme value of speed, extreme value of acceleration, speed distribution, etc., and these parameters may be set freely by the staff, and are not limited in particular.
Fig. 4 is a third sub-flowchart of the inspection method for the livestock and poultry pig farm based on artificial intelligence, wherein the step of determining sampling points by the statistically marked cultivation units based on the positional relationship between each cultivation unit and the inspection path and transmitting the inspection path containing the sampling points to the moving end comprises steps S301 to S303:
step S301: the statistical marked cultivation unit inquires a corresponding passage section of the cultivation unit in the passage area;
step S302: determining image acquisition points in the traffic segment according to a preset positioning rule;
step S303: marking sampling points corresponding to the image acquisition points in the inspection path, and sending the inspection path containing the sampling points to a moving end.
The passing section refers to a position where a close-up image of the culture unit can be acquired, and a connecting line between the image acquisition point and the center of the culture unit is generally perpendicular to the inspection path.
Fig. 5 is a fourth sub-flowchart of the inspection method for the livestock and poultry pig farm based on artificial intelligence, wherein when the moving end moves to the sampling point, audio information is queried and played in a preset audio library, feedback audio of animals is obtained, and steps of collecting close-range images according to the feedback audio include steps S401 to S404:
step S401: when the motion end moves to a sampling point, collecting environmental audio;
step S402: inquiring at least one recorded audio of animals in the culture unit corresponding to the sampling point in a preset audio library, and randomly playing the recorded audio;
step S403: acquiring feedback audio of an animal, and correcting the feedback audio according to the environmental audio to obtain the animal audio;
step S404: inputting the animal audio into a trained identification model to generate an animal stable value, and collecting a close-range image when the stable value is smaller than a preset stable threshold value.
The above-mentioned contents specifically describe the analysis process of the audio information, firstly, environmental audio needs to be collected, and the environmental audio is taken as a default value; then, inquiring recorded audio in a preset audio library, and playing, wherein the number of the recorded audio is generally not unique, so that animals are prevented from getting used to the sound, and unreal feedback is further made; finally, obtaining feedback audio of the animal, and eliminating environmental audio from the feedback audio to obtain the animal audio; identifying the animal audio, namely determining a stable value of the animal, wherein the stable value is not the same as the stable level, and acquiring a close-range image when the stable value is smaller than a preset stable threshold; and the close-range image is delivered to a manual end for manual detection.
Example 2
Fig. 6 is a block diagram of a constituent structure of an artificial intelligence-based inspection system for a livestock and poultry farm, in which the system 10 includes:
the port setting module 11 is used for acquiring a culture model containing area labels, and determining a patrol path of the moving end and the installation position of the monitoring end based on the culture model; the regional tag comprises a passing region and a breeding unit;
the unit marking module 12 is used for receiving the image information acquired by the monitoring end and marking the culture unit according to the image information and the installation position of the monitoring end;
the sampling point determining module 13 is used for counting marked cultivation units, determining sampling points based on the position relation between each cultivation unit and the inspection path, and sending the inspection path containing the sampling points to the moving end;
the close-range image acquisition module 14 is used for inquiring and playing audio information in a preset audio library when the motion end moves to a sampling point, acquiring feedback audio of animals, and acquiring close-range images according to the feedback audio;
the audio library comprises a tag item and an audio item, wherein the tag item is used for representing the position information of the culture unit; the audio item is the sound of the animal in the pre-collected breeding unit.
Fig. 7 is a block diagram of the constitution of a port setting module 11 in the inspection system of the livestock and poultry pig farm based on artificial intelligence, wherein the port setting module 11 comprises:
a passing area determining unit 111, configured to establish a connection channel with a drawing database, read a cultivation model including a cultivation unit, and determine a passing area according to the cultivation unit;
a cultivation state query unit 112, configured to establish a connection channel with the cultivation record database, and query the cultivation state of each cultivation unit; the culture state is used for representing the physical sign parameters of the cultured animals, and the physical sign parameters at least comprise age and vaccination information;
a position determining unit 113, configured to determine a stability level of each animal based on the sign parameter, and determine an installation position of the monitoring end according to the stability level;
the path determining unit 114 is configured to determine a routing inspection path of the moving end according to the traffic zone.
Fig. 8 is a block diagram of the constitution of a unit marking module 12 in an inspection system of a livestock and poultry pig farm based on artificial intelligence, wherein the unit marking module 12 comprises:
an image sorting unit 121, configured to receive image information containing time information acquired in real time by a monitoring end, sort the image information according to the time information, and obtain an image group;
an image screening unit 122, configured to sequentially compare adjacent images in the image group, calculate an overlap ratio, and screen the image group according to the overlap ratio;
the contour analysis unit 123 is configured to perform contour recognition on the screened image set to obtain contours of the animals and obtain center points of the contours of the animals;
an execution unit 124, configured to determine a motion trail of each animal according to the center point, and mark a corresponding cultivation unit according to the motion trail;
the rule for screening the image group according to the contact ratio is that when the contact ratio is the same, deleting the later image; when the overlap ratio is different, both images remain.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (2)
1. An artificial intelligence-based inspection method for livestock and poultry pig farms is characterized by comprising the following steps of:
acquiring a culture model containing regional labels, and determining a patrol path of a moving end and the installation position of a monitoring end based on the culture model; the regional tag comprises a passing region and a breeding unit;
receiving image information acquired by a monitoring end, and marking a culture unit according to the image information and the installation position of the monitoring end;
the statistical marked cultivation units determine sampling points based on the position relation between the cultivation units and the inspection path, and the inspection path containing the sampling points is sent to the moving end;
when the motion end moves to a sampling point, inquiring and playing audio information in a preset audio library, acquiring feedback audio of animals, and acquiring a close-range image according to the feedback audio;
the audio library comprises a tag item and an audio item, wherein the tag item is used for representing the position information of the culture unit; the audio item is the sound of animals in the pre-collected culture unit;
the step of obtaining the culture model containing the regional tag and determining the patrol path of the moving end and the installation position of the monitoring end based on the culture model comprises the following steps:
establishing a connection channel with a drawing database, reading a cultivation model containing cultivation units, and determining a passing area according to the cultivation units;
establishing a connection channel with a culture record database, and inquiring the culture state of each culture unit; the culture state is used for representing the physical sign parameters of the cultured animals, and the physical sign parameters at least comprise age and vaccination information;
determining the stability level of each animal based on the physical sign parameters, and determining the installation position of the monitoring end according to the stability level;
determining a routing inspection path of the moving end according to the passing area;
the step of determining sampling points by the culture units of the statistical marks based on the position relation between each culture unit and the inspection path and sending the inspection path containing the sampling points to the moving end comprises the following steps:
the statistical marked cultivation unit inquires a corresponding passage section of the cultivation unit in the passage area;
determining image acquisition points in the traffic segment according to a preset positioning rule;
marking sampling points corresponding to the image acquisition points in the inspection path, and sending the inspection path containing the sampling points to a moving end;
when the motion end moves to a sampling point, inquiring and playing audio information in a preset audio library, acquiring feedback audio of animals, and acquiring close-range images according to the feedback audio, wherein the steps comprise:
when the motion end moves to a sampling point, collecting environmental audio;
inquiring at least one recorded audio of animals in the culture unit corresponding to the sampling point in a preset audio library, and randomly playing the recorded audio;
acquiring feedback audio of an animal, and correcting the feedback audio according to the environmental audio to obtain the animal audio;
inputting the animal audio into a trained identification model to generate an animal stable value, and collecting a close-range image when the stable value is smaller than a preset stable threshold value;
the step of determining the stability level of each animal based on the physical sign parameters and determining the installation position of the monitoring end according to the stability level comprises the following steps:
inquiring the stability level of each animal in a preset level table based on the sign parameters;
determining monitoring definition according to the stability level, and determining the height of a monitoring end according to the monitoring definition;
determining a monitoring range of the monitoring end according to the height of the monitoring end and the wide angle of the monitoring end, and determining the installation position of the monitoring end according to the monitoring range;
wherein the union of the monitoring ranges is greater than the union of the culture units;
the step of receiving the image information acquired by the monitoring end and marking the culture unit according to the image information and the installation position of the monitoring end comprises the following steps:
receiving image information containing time information obtained in real time by a monitoring terminal, and sequencing the image information according to the time information to obtain an image group;
sequentially comparing adjacent images in the image group, calculating the coincidence ratio, and screening the image group according to the coincidence ratio;
performing contour recognition on the screened image group to obtain animal contours and obtain center points of the animal contours;
determining the motion trail of each animal according to the center point, and marking the corresponding culture unit according to the motion trail;
the rule for screening the image group according to the contact ratio is that when the contact ratio is the same, deleting the later image; when the coincidence ratio is different, both images are reserved;
the step of determining the motion trail of each animal according to the center point and marking the corresponding breeding unit according to the motion trail comprises the following steps:
sequentially reading the center points of animal outlines of adjacent images in the screened image group, and determining animal displacement according to a preset scale;
calculating animal speed and acceleration according to the animal displacement and the time information of the adjacent images, and generating a motion parameter table in a preset time period; the motion parameter table comprises a time item;
inputting the motion parameter table into a trained comparison model containing a historical motion parameter table, and calculating an abnormal value;
when the abnormal value reaches a preset abnormal threshold value, marking a culture unit corresponding to the animal outline; and when the abnormal value is smaller than a preset abnormal threshold value, updating the historical motion parameter table according to the motion parameter table.
2. An artificial intelligence-based livestock and poultry pig farm inspection system, characterized in that the system comprises:
the port setting module is used for acquiring a culture model containing area labels, and determining a patrol path of the moving end and the installation position of the monitoring end based on the culture model; the regional tag comprises a passing region and a breeding unit;
the unit marking module is used for receiving the image information acquired by the monitoring end and marking the culture unit according to the image information and the installation position of the monitoring end;
the sampling point determining module is used for counting marked cultivation units, determining sampling points based on the position relation between each cultivation unit and the inspection path, and sending the inspection path containing the sampling points to the moving end;
the near-view image acquisition module is used for inquiring and playing audio information in a preset audio library when the motion end moves to a sampling point, acquiring feedback audio of animals and acquiring near-view images according to the feedback audio;
the audio library comprises a tag item and an audio item, wherein the tag item is used for representing the position information of the culture unit; the audio item is the sound of animals in the pre-collected culture unit;
the port setting module includes:
the system comprises a passing area determining unit, a drawing database and a drawing database, wherein the passing area determining unit is used for establishing a connecting channel with the drawing database, reading a cultivation model containing a cultivation unit and determining a passing area according to the cultivation unit;
the culture state inquiry unit is used for establishing a connection channel with the culture record database and inquiring the culture state of each culture unit; the culture state is used for representing the physical sign parameters of the cultured animals, and the physical sign parameters at least comprise age and vaccination information;
the position determining unit is used for determining the stability level of each animal based on the physical sign parameters and determining the installation position of the monitoring end according to the stability level;
the route determining unit is used for determining a routing inspection route of the moving end according to the passing area;
the statistic marked cultivation unit determines sampling points based on the position relation between each cultivation unit and the inspection path, and sends the inspection path containing the sampling points to the moving end, wherein the content comprises the following components:
the statistical marked cultivation unit inquires a corresponding passage section of the cultivation unit in the passage area;
determining image acquisition points in the traffic segment according to a preset positioning rule;
marking sampling points corresponding to the image acquisition points in the inspection path, and sending the inspection path containing the sampling points to a moving end;
when the motion end moves to a sampling point, inquiring and playing audio information in a preset audio library, acquiring feedback audio of animals, and acquiring content of close-range images according to the feedback audio comprises the following steps:
when the motion end moves to a sampling point, collecting environmental audio;
inquiring at least one recorded audio of animals in the culture unit corresponding to the sampling point in a preset audio library, and randomly playing the recorded audio;
acquiring feedback audio of an animal, and correcting the feedback audio according to the environmental audio to obtain the animal audio;
inputting the animal audio into a trained identification model to generate an animal stable value, and collecting a close-range image when the stable value is smaller than a preset stable threshold value;
the determining the stability level of each animal based on the sign parameters, and determining the installation position of the monitoring end according to the stability level comprises the following steps:
inquiring the stability level of each animal in a preset level table based on the sign parameters;
determining monitoring definition according to the stability level, and determining the height of a monitoring end according to the monitoring definition;
determining a monitoring range of the monitoring end according to the height of the monitoring end and the wide angle of the monitoring end, and determining the installation position of the monitoring end according to the monitoring range;
wherein the union of the monitoring ranges is greater than the union of the culture units;
the unit marking module includes:
the image ordering unit is used for receiving the image information containing the time information acquired by the monitoring terminal in real time, and ordering the image information according to the time information to obtain an image group;
the image screening unit is used for sequentially comparing adjacent images in the image group, calculating the contact ratio and screening the image group according to the contact ratio;
the contour analysis unit is used for carrying out contour recognition on the screened image group to obtain contours of all animals and obtain center points of the contours of all animals;
the execution unit is used for determining the motion trail of each animal according to the central point and marking the corresponding culture unit according to the motion trail;
the rule for screening the image group according to the contact ratio is that when the contact ratio is the same, deleting the later image; when the coincidence ratio is different, both images are reserved;
the step of determining the motion trail of each animal according to the center point and marking the corresponding breeding unit according to the motion trail comprises the following steps:
sequentially reading the center points of animal outlines of adjacent images in the screened image group, and determining animal displacement according to a preset scale;
calculating animal speed and acceleration according to the animal displacement and the time information of the adjacent images, and generating a motion parameter table in a preset time period; the motion parameter table comprises a time item;
inputting the motion parameter table into a trained comparison model containing a historical motion parameter table, and calculating an abnormal value;
when the abnormal value reaches a preset abnormal threshold value, marking a culture unit corresponding to the animal outline; and when the abnormal value is smaller than a preset abnormal threshold value, updating the historical motion parameter table according to the motion parameter table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210923788.8A CN115250950B (en) | 2022-08-02 | 2022-08-02 | Method and system for inspecting livestock and poultry pig farm based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210923788.8A CN115250950B (en) | 2022-08-02 | 2022-08-02 | Method and system for inspecting livestock and poultry pig farm based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115250950A CN115250950A (en) | 2022-11-01 |
| CN115250950B true CN115250950B (en) | 2024-01-19 |
Family
ID=83746465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210923788.8A Active CN115250950B (en) | 2022-08-02 | 2022-08-02 | Method and system for inspecting livestock and poultry pig farm based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115250950B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116074479A (en) * | 2023-03-03 | 2023-05-05 | 山东交通学院 | Image analysis-based passenger monitoring system, method, equipment and storage medium |
| CN116151779B (en) * | 2023-04-23 | 2023-07-18 | 深圳奥雅设计股份有限公司 | Plant view monitoring and management system |
| CN117911190B (en) * | 2024-01-23 | 2024-09-24 | 内蒙古工业大学 | Pasture animal management method based on electronic fence and image analysis |
| CN118446830B (en) * | 2024-07-05 | 2024-09-24 | 厦门农芯数字科技有限公司 | Intelligent pig farm dialogue system and method with long-term and short-term memory cooperation |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012079107A2 (en) * | 2010-12-15 | 2012-06-21 | Mkw Electronics Gmbh | Method for displaying a piece of information associated with an animal |
| CN103995534A (en) * | 2014-05-13 | 2014-08-20 | 北京艾科尔明国际科技发展有限公司 | Mobile path control system and control method for patrolling robot |
| CN109255297A (en) * | 2018-08-06 | 2019-01-22 | 百度在线网络技术（北京）有限公司 | animal state monitoring method, terminal device, storage medium and electronic equipment |
| CN109460029A (en) * | 2018-11-29 | 2019-03-12 | 华南农业大学 | Livestock and poultry cultivation place inspection mobile platform and its control method |
| CN109922108A (en) * | 2017-12-13 | 2019-06-21 | 南京开天眼无人机科技有限公司 | A kind of herd management system based on unmanned plane |
| CN110265041A (en) * | 2019-07-01 | 2019-09-20 | 河南牧业经济学院 | A kind of method and system for the song behavior collected, analyze pig |
| CN112000051A (en) * | 2020-09-14 | 2020-11-27 | 江西省农业科学院农业经济与信息研究所 | Livestock breeding management system based on Internet of things |
| CN112598701A (en) * | 2020-11-06 | 2021-04-02 | 太原理工大学 | Automatic tracking and monitoring video acquisition system and method for farm targets |
| KR102315991B1 (en) * | 2020-12-31 | 2021-10-21 | 한국축산데이터 주식회사 농업회사법인 | Device and system for managing livestck remotly using artificial intelligence |
| WO2021237144A1 (en) * | 2020-05-21 | 2021-11-25 | Board Of Trustees Of Michigan State University | Systems and methods for automatic and noninvasive livestock health analysis |
| CN114581855A (en) * | 2022-04-29 | 2022-06-03 | 深圳格隆汇信息科技有限公司 | Information collection method and system based on big data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11766028B2 (en) * | 2019-06-04 | 2023-09-26 | Aglogica Holdings Inc | Methods and systems for using sound data to analyze health condition and welfare states in collections of farm animals |
-
2022
- 2022-08-02 CN CN202210923788.8A patent/CN115250950B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012079107A2 (en) * | 2010-12-15 | 2012-06-21 | Mkw Electronics Gmbh | Method for displaying a piece of information associated with an animal |
| CN103995534A (en) * | 2014-05-13 | 2014-08-20 | 北京艾科尔明国际科技发展有限公司 | Mobile path control system and control method for patrolling robot |
| CN109922108A (en) * | 2017-12-13 | 2019-06-21 | 南京开天眼无人机科技有限公司 | A kind of herd management system based on unmanned plane |
| CN109255297A (en) * | 2018-08-06 | 2019-01-22 | 百度在线网络技术（北京）有限公司 | animal state monitoring method, terminal device, storage medium and electronic equipment |
| CN109460029A (en) * | 2018-11-29 | 2019-03-12 | 华南农业大学 | Livestock and poultry cultivation place inspection mobile platform and its control method |
| CN110265041A (en) * | 2019-07-01 | 2019-09-20 | 河南牧业经济学院 | A kind of method and system for the song behavior collected, analyze pig |
| WO2021237144A1 (en) * | 2020-05-21 | 2021-11-25 | Board Of Trustees Of Michigan State University | Systems and methods for automatic and noninvasive livestock health analysis |
| CN112000051A (en) * | 2020-09-14 | 2020-11-27 | 江西省农业科学院农业经济与信息研究所 | Livestock breeding management system based on Internet of things |
| CN112598701A (en) * | 2020-11-06 | 2021-04-02 | 太原理工大学 | Automatic tracking and monitoring video acquisition system and method for farm targets |
| KR102315991B1 (en) * | 2020-12-31 | 2021-10-21 | 한국축산데이터 주식회사 농업회사법인 | Device and system for managing livestck remotly using artificial intelligence |
| CN114581855A (en) * | 2022-04-29 | 2022-06-03 | 深圳格隆汇信息科技有限公司 | Information collection method and system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115250950A (en) | 2022-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115250950B (en) | Method and system for inspecting livestock and poultry pig farm based on artificial intelligence | |
| CN108037717B (en) | Intelligent pig farm monitoring system | |
| CN107711576B (en) | Sow oestrus identification method and system | |
| CN111766827A (en) | Modern pig breeding base real-time monitoring management system based on big data | |
| CN107667904A (en) | Live pig big data system based on technology of Internet of things | |
| CN105091938A (en) | Poultry health monitoring method and system | |
| Feiyang et al. | Monitoring behavior of poultry based on RFID radio frequency network | |
| CN111797831A (en) | BIM and artificial intelligence based parallel abnormality detection method for poultry feeding | |
| CN107771706B (en) | Sow oestrus detection method and system | |
| CN110991222A (en) | Object state monitoring and sow oestrus monitoring method, device and system | |
| CN116627082A (en) | Intelligent management system for live pig breeding suitable for stock quantity checking | |
| KR20100011363A (en) | System and method for servicing suit feed in a farm using wire and wireless telecommunication network | |
| CN112325949A (en) | Breeding environment monitoring system | |
| CN106614071B (en) | Sow antenatal behavior monitoring system and method | |
| CN115468598A (en) | Intelligent monitoring method and system for pigsty environment | |
| CN115359050B (en) | Method and device for detecting abnormal feed intake of livestock | |
| CN113516139A (en) | Data processing method, device, equipment and storage medium | |
| CN110779575A (en) | Animal husbandry monitoring system based on big data platform | |
| CN116110586A (en) | Elephant health management system based on YOLOv5 and SlowFast | |
| CN115968810A (en) | Identification method and identification system for sick and epidemic pigs | |
| TWI784740B (en) | Poultry health monitoring system and method thereof | |
| CN113408334B (en) | Crayfish full-chain data acquisition and intelligent detection method and device | |
| CN114022302A (en) | Pregnant goat intelligent management system and management method based on video monitoring | |
| Bastiaansen et al. | Continuous real-time cow identification by reading ear tags from live-stream video | |
| Chan et al. | Comparison of manual, machine learning, and hybrid methods for video annotation to extract parental care data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |