Intelligent patrol monitoring method and device for boar behaviors

Abstract

The invention provides an intelligent patrol monitoring method and device for boar behaviors, and the method comprises the steps: generating an identity label for a boar according to the identity of the to-be-monitored boar; deploying the identity label at a designated position of a pigsty where the boar is located in the pigsty; collecting an electronic map of a working way in the hog house; a destination coordinate point of the monitoring system hardware device during automatic patrol is set in the electronic map, the monitoring system hardware device starts automatic navigation according to the position of the monitoring system hardware device and the electronic map, automatic patrol of the monitoring system is started, and the monitoring system hardware device performs automatic patrol according to the destination coordinate point. And carrying out automatic path planning and mobile patrol on a working way in the hog house. The problems that the installation cost investment is large, resource waste is caused due to the fact that multiple pig house scenes cannot be shared, and the identity of the breeding pigs cannot be accurately positioned due to the fact that the identity information of the monitored pigs is not associated are solved.

Description

Technical Field

[0001] The present invention relates to the field of agricultural information technology, and in particular to a method, device, electronic equipment, computer-readable storage medium and computer program product for intelligent inspection and monitoring of breeding pig behavior. Background Art

[0002] In the field of livestock breeding and farming technology, the selection, cultivation, elimination and renewal of breeding pigs are the most critical links in the breeding process. There is a potential correlation between many behaviors of pigs and their genes. If the genetic effects of behavior are taken into account, the heritability of pig growth performance will be improved. Therefore, identifying the behavior of breeding pigs helps to screen good pigs and improve breeding efficiency. Good pigs, for example, are pigs with strong growth performance, including not only high lactation / feeding ability of sows, but also high social ability of piglets, that is, low aggressiveness. The earliest behavioral monitoring of pigs was that the breeders directly monitored their behavior. This method increased the time of contact between humans and animals, had the risk of zoonotic diseases, and was also affected by the subjective experience of the breeders. With the expansion of intensive farming, manual monitoring methods are bound to be difficult to meet the large number of pig behavior monitoring needs. Therefore, it is necessary to use an automated monitoring system to grasp the behavior, genetic habits, etc. of breeding pigs in a high-throughput and timely manner, reduce contact between humans and pigs, and reduce the probability of infectious diseases. This is of great significance to the selection and management of breeding pigs and improving the production efficiency of breeding pig farms.

[0003] The existing automatic monitoring methods of pig behavior mainly use fixed-position monitoring of a single pen or fixed track inspections in a single pig house to monitor pig behavior. However, these methods have some shortcomings. The fixed-position method cannot realize the behavior monitoring of breeding pigs in different pig pens. It is necessary to install monitoring equipment in each pen in the house, which has a large investment cost and low equipment resource utilization. The fixed track method requires laying the guide rail before use, which has a high installation cost and can only be used in a single pig house, and cannot be shared between different pig houses. It can be seen that the existing solutions are difficult to meet the needs of breeding pig behavior monitoring in different pig house scenarios in different growth cycles, and have poor versatility and flexibility. In addition, the existing pig behavior monitoring methods and systems often do not associate the identity information of the monitored pigs. They only identify the behavior category of the pigs in the image, but do not associate the identity information of the pigs that have the behavior, such as which pen, which pen, who the parents are, and other information required for the pig genetic breeding process. As a result, although the system monitors certain behaviors of the pigs, it cannot accurately locate the identity of the breeding pigs, and thus prevents managers from making targeted genetic breeding management measures in a timely manner. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for intelligent patrol and monitoring of breeding pig behavior, so as to solve the problems raised in the above background technology, such as high installation cost, inability to share multiple pig house scenes resulting in waste of resources, and inability to accurately locate the breeding pig's identity due to the lack of association with the identity information of the monitored pig.

[0005] In order to achieve the above technical effects, the present invention includes the following core invention points:

[0006] 1. The existing technology lacks an overall solution of "mobile patrol monitoring + behavior recognition + identity recognition in the breeding scene of sows". In view of this situation, the present invention provides an overall solution of mobile patrol combined with behavior monitoring;

[0007] 2. Different from the existing behavior recognition methods based on pure video classification and the behavior recognition methods based on single-frame image detection, the present invention applies the deep learning video action detection model (breeding pig behavior recognition model) to the real-time recognition of breeding pig behavior in multi-pig house scenes for the first time. The action detection model can not only detect the behavior of breeding pigs in the video (that is, these pig breeding-related behaviors can be automatically framed out and presented in the video), but also the model infers the specific category of the detected breeding pig behavior by comprehensively considering the continuous optical flow features of the pig behavior in the current video frame and the adjacent video frames. This comprehensive consideration of the current video frame and the adjacent video frame is better than the single-frame detection to identify the behavior category. In addition, it is worth mentioning that, unlike the prior art multi-stage recognition model (that is, first use a detection model to detect and frame the positions of all pigs in the picture, and then use another model to analyze the relationship between the two pigs in the picture, such as using the head-tail distance to determine whether there is a tail biting situation between the two pigs, etc.), the action detection model proposed by the present invention is an end-to-end single-stage recognition model, and only the action detection model can be used to complete the behavior detection and category recognition tasks in the video, so it is more efficient and has higher recognition accuracy. The action detection model of the present invention is improved on the basis of the original YOLOv2, making it more suitable for deployment on edge computing terminals, thereby realizing accurate and efficient pig behavior inspection and monitoring.

[0008] In view of the shortcomings of existing technologies, such as Figure 7 As shown, the present invention proposes a method for intelligent inspection and monitoring of breeding pig behavior, which includes:

[0009] The identification construction step generates an identification for the breeding pig according to the identity of the breeding pig to be monitored;

[0010] The identification deployment step is to deploy the identification tag at a designated location of the pen where the pig is located in the pig house;

[0011] In-house mapping step, collect the electronic map of the working path in the pig house;

[0012] The automatic inspection step includes setting the destination coordinate point of the monitoring system hardware device during the automatic inspection in the electronic map, starting the automatic navigation of the monitoring system hardware device according to its own position and the electronic map, and starting the automatic inspection of the monitoring system. The monitoring system hardware device performs automatic path planning and mobile inspection on the working path in the pig house according to the destination coordinate point;

[0013] Among them, the mobile inspection in the automatic inspection step includes:

[0014] During the mobile patrol process, the camera of the monitoring system hardware device is turned on to collect the phenotypic video of the breeding pig in the pens along the road, and scans and identifies the identity tag to obtain the breeding pig identity information; the breeding pig behavior recognition model reads the pig phenotypic video and identifies the breeding pig behavior information including the location and behavior category of the behavior; the breeding pig behavior information and its breeding pig identity information are associated as the pig behavior patrol monitoring result.

[0015] The intelligent inspection and monitoring method for breeding pig behavior, wherein the monitoring system hardware device comprises:

[0016] Wheeled mobile chassis, retractable bracket, at least one camera;

[0017] The four-wheel mobile chassis adopts the design of front drive wheels and rear universal wheels;

[0018] The retractable bracket is installed in the center above the four-wheel mobile chassis and can be freely adjusted vertically according to the height of the pig pen.

[0019] The camera is installed on a telescopic bracket; the camera is electrically connected to the data processing module of the monitoring system hardware device to transmit the phenotypic video and the identity identification collected by the camera to the data processing module for processing, so as to identify the behavior information and identity information of the breeding pig.

[0020] The intelligent inspection and monitoring method for breeding pig behavior, wherein the breeding pig behavior information includes: suckling behavior, feeding behavior, and piglet attack behavior;

[0021] The intelligent inspection and monitoring method for breeding pig behavior also includes: a breeding step, according to the inspection and monitoring results, selecting multiple breeding pigs of each breed that meet the breeding requirements, and hybridizing multiple breeding pigs of one breed with multiple breeding pigs of another breed to select breeds carrying excellent genes.

[0022] The method for intelligent inspection and monitoring of sow behavior, wherein the process of constructing the sow behavior recognition model comprises:

[0023] The ELAN module in the 2D backbone network of YOWOv2 is replaced by the GELAN module from YOLOv9, and convolution RepConv is used in the GELAN module to obtain an intermediate model, so that the network can improve the feature representation learning ability without increasing the model calculation amount;

[0024] The 2D convolution of FasterNet in the 3D backbone network of the intermediate model is converted into 3D convolution, and the 2D pooling is converted into 3D pooling to obtain the pig behavior recognition model.

[0025] The present invention also provides a pig behavior intelligent patrol monitoring device, which includes:

[0026] The identification building module generates an identification for the breeding pig according to the identity of the breeding pig to be monitored;

[0027] The identification deployment module deploys the identity identification at a designated location of the pen where the pig is located in the pig house;

[0028] The in-house mapping module collects the electronic map of the working path in the pig house;

[0029] The automatic inspection module sets the destination coordinate point of the monitoring system hardware device during automatic inspection in the electronic map, starts automatic navigation of the monitoring system hardware device according to its own position and the electronic map, and starts the automatic inspection of the monitoring system. The monitoring system hardware device performs automatic path planning and mobile inspection on the working path in the pig house according to the destination coordinate point;

[0030] Among them, the mobile inspection in the automatic inspection module includes:

[0031] During the mobile patrol process, the camera of the monitoring system hardware device is turned on to collect the phenotypic video of the breeding pig in the pens along the road, and scans and identifies the identity tag to obtain the breeding pig identity information; the breeding pig behavior recognition model reads the pig phenotypic video and identifies the breeding pig behavior information including the location and behavior category of the behavior; the breeding pig behavior information and its breeding pig identity information are associated as the pig behavior patrol monitoring result.

[0032] The intelligent inspection and monitoring device for breeding pig behavior, wherein the monitoring system hardware device comprises:

[0033] Wheeled mobile chassis, retractable bracket, at least one camera;

[0034] The four-wheel mobile chassis adopts the design of front drive wheels and rear universal wheels;

[0035] The retractable bracket is installed in the center above the four-wheel mobile chassis and can be freely adjusted vertically according to the height of the pig pen.

[0036] The camera is installed on a telescopic bracket; the camera is electrically connected to the data processing module of the monitoring system hardware device to transmit the phenotypic video and the identity identification collected by the camera to the data processing module for processing, so as to identify the behavior information and identity information of the breeding pig.

[0037] The intelligent inspection and monitoring device for breeding pig behavior, wherein the breeding pig behavior information includes: suckling behavior, feeding behavior, and piglet attack behavior;

[0038] The intelligent inspection and monitoring device for breeding pig behavior also includes: a breeding module, which selects multiple breeding pigs of various breeds that meet breeding requirements according to the inspection and monitoring results, and crossbreeds multiple breeding pigs of one breed with multiple breeding pigs of another breed to breed breeds carrying excellent genes;

[0039] The construction process of the pig behavior recognition model includes:

[0040] The ELAN module in the 2D backbone network of YOWOv2 is replaced by the GELAN module from YOLOv9, and convolution RepConv is used in the GELAN module to obtain an intermediate model, so that the network can improve the feature representation learning ability without increasing the model calculation amount;

[0041] The 2D convolution of FasterNet in the 3D backbone network of the intermediate model is converted into 3D convolution, and the 2D pooling is converted into 3D pooling to obtain the pig behavior recognition model.

[0042] The present invention also proposes an electronic device, which includes the above-mentioned intelligent patrol and monitoring device for breeding pig behavior. The electronic device may be connected to an information display device, and the information display device is used to display the pig behavior patrol and monitoring results according to the display parameters and attributes set by the user or through an artificial intelligence model.

[0043] The present invention also proposes a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for intelligent patrol and monitoring of breeding pig behavior are implemented.

[0044] The present invention also proposes a computer program product, including a computer program, wherein the computer program implements the steps of the method for intelligent patrol and monitoring of breeding pig behavior when executed by a processor.

[0045] It should be noted that the intelligent patrol and monitoring method and system for breeding pig behavior provided by the present invention can be applicable to multi-pig house scenarios, such as a farrowing house scenario with sows in limited pens, or a nursery house scenario with piglets raised in groups.

[0046] In summary, it can be seen from the above scheme that the advantages of the present invention are:

[0047] 1. For the first time, the YOWOv2 algorithm of deep learning video motion detection was applied to the real-time recognition of breeding pig behavior in multi-pig house scenarios. At the same time, the YOWOv2 algorithm was improved to make it more suitable for deployment on edge computing terminals, thereby achieving more accurate and efficient breeding pig behavior inspection and monitoring.

[0048] 2. By setting a special identity QR code on the front baffle of the pig pen, the QR code can be further identified to associate the behavior and identity information of the monitored breeding pigs, which will help managers make more targeted breeding management measures.

[0049] 3. The method and system for intelligent monitoring of the behavior of sows provided by the present invention adopt a four-wheel mobile chassis driving mode, which does not require the modification of adding guide rails on site, has better mobility and flexible deployment, can monitor the behavior of sows in different pens in a single pig house, and can also be shared among multiple pig houses at different stages, meeting the actual needs of monitoring the behavior of sows in multiple pig house scenarios with one system, avoiding waste of resources, and achieving the purpose of saving costs and increasing efficiency.

[0050] 4. The present invention adopts a camera with an anti-shake pan / tilt, which can greatly reduce the image ghosting problem caused by the bumping and shaking of the camera when the four-wheel mobile chassis moves automatically, thereby enhancing the stability of the breeding pig behavior monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of the intelligent inspection and monitoring method for breeding pig behavior provided by the present invention;

[0052] Figure 2 It is a schematic diagram of the module structure of the hardware device of the monitoring system of the present invention;

[0053] Figure 3 It is a schematic diagram of the composition of the hardware modules in the warehouse of the monitoring system hardware device module provided by the present invention;

[0054] Figure 4 It is a structural schematic diagram of a system control and data processing module provided by the present invention;

[0055] Figure 5 This is a schematic diagram of a client submodule software interface provided by the present invention;

[0056] Figure 6 This is a schematic diagram of an improved YOWOv2 network structure provided by the present invention;

[0057] Figure 7 is a flow chart of the method of the present invention;

[0058] Figure 8 It is a module diagram of the device of the present invention;

[0059] Fig. 9 This is a schematic diagram of the structure of a first electronic device of the present invention;

[0060] Fig.10 This is a schematic diagram of the application environment structure of the first electronic device of the present invention;

[0061] Fig.11 It is a schematic diagram of the structure of a second electronic device of the present invention.

[0062] Reference numerals:

[0063] A-First electronic device;

[0064] B- Intelligent patrol and monitoring device for breeding pig behavior;

[0065] C-data acquisition equipment;

[0066] D-information display device;

[0067] 1000 - second electronic device;

[0068] Ⅰ-computational unit;

[0069] II-ROM;

[0070] III-RAM;

[0071] IV-bus;

[0072] V-interface;

[0073] VI - input unit;

[0074] VII-output unit;

[0075] VIII- Storage medium;

[0076] Ⅸ-Communication unit. DETAILED DESCRIPTION

[0077] It should be noted that, in this application, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0078] Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0079] The processor described in the present invention is the control center of the electronic device, which can be a processor or a general term for multiple processing elements. For example, it can be one or more central processing units (CPUs), or application specific integrated circuits (ASICs), or one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (FPGAs).

[0080] Optionally, the processor can perform various functions of the electronic device by running or executing a software program stored in the memory, and calling data stored in the memory.

[0081] In a specific implementation, as an embodiment, the processor may include one or more CPUs. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions). Electronic devices may include: servers, desktop computers, laptops, smart phones, tablet computers, embedded computers, etc., wherein the embedded computers include vehicles and robots, etc.

[0082] The memory is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0083] It should be noted that the structure of the electronic device shown in the drawings of the present invention does not constitute a limitation thereto, and the actual knowledge structure recognition device may include more or fewer components than shown in the drawings, or a combination of certain components, or a different arrangement of components.

[0084] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0085] It should also be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0086] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0087] It should also be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0088] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0089] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0091] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0092] In order to achieve the above technical effects, in a first aspect, the present invention provides a method for intelligent inspection and monitoring of breeding pig behavior, the implementation steps of which include:

[0093] Step S1, constructing a QR code label. To construct a QR code label, you first need to generate a QR code electronic image. You can choose to use the Python QR code generation library qrcode to generate a QR code electronic image with specific information. The generated QR code electronic image can be printed by a label printer to generate a QR code label. The information in the QR code label includes identity information such as the pig house number, the stall number, the pen number, and the parent identification number.

[0094] Step S2, deploy the QR code logo. The generated QR code logo needs to be deployed to each pen in the pig house. Specifically, for the farrowing house scene, the QR code logo is fixed on the front baffle of each limit fence, at the junction of the two limit fences, so that when the system moves to this position, the first camera is at the best shooting angle, and the height of the QR code logo from the ground is in the range of 0.65 meters to 0.75 meters; for the nursery scene, the QR code logo is fixed in the middle of the front baffle of each piglet pen, so that when the system moves to this position, the first camera is at the best shooting angle, and the height of the QR code logo from the ground is in the range of 0.65 meters to 0.75 meters.

[0095] Step S3, build a map inside the pig house. The map built inside the pig house is mainly used to obtain the road condition information inside the pig house before the system automatically navigates, so that the monitoring system can automatically plan navigation and movement based on the built electronic map. The main steps of building a map inside the pig house include: the staff controls the monitoring system hardware device module through a handheld remote control to traverse the working path of the entire area of ​​the pig house in advance, and uses laser radar, RGB-D camera and SLAM technology to traverse and scan and collect the road condition information of the pig house, forming a three-dimensional electronic map of the pig house scene, and storing it in the system control and data processing module for subsequent automatic navigation inspections.

[0096] Step S4, start automatic inspection. Set the destination coordinate point that the monitoring system hardware device module needs to go to during automatic inspection in the constructed three-dimensional map. The destination coordinate point can be set in the three-dimensional map using rviz software; place the monitoring system hardware device module at the navigation starting point, execute the start automatic navigation ROS command, and start the automatic inspection of the monitoring system. The monitoring system hardware device module will perform automatic path planning and mobile inspection on the pig house work path according to the navigation destination coordinate point.

[0097] Step S5, identification of sow behavior and identity information. During the automatic inspection process of the monitoring system hardware device module, the first camera located at the upper part of the monitoring system hardware device module is turned on and collects phenotypic video information of sows in the pig limiting fences or pig pens along the way, but the visual range of the first camera may be difficult to cover the QR code logo on the pig pen baffle. At the same time, the second camera located at the lower part of the monitoring system hardware device module will scan the QR code logo fixed on the pig limiting fence or pig pen baffle along the way; the sow behavior recognition model reads the pig phenotypic video information collected by the first camera and identifies the sow behavior; the sow behavior includes the location where the behavior occurs and the behavior category. Through reasoning, the model can automatically frame the location of specific sow behavior in the video and identify the behavior category of the pig in the frame. These specific sow behaviors are behaviors of concern in pig breeding, such as sow lactation behavior, feeding behavior, piglet attack behavior, etc. These behaviors will be passed on to the next generation through genes. The more times a sow breastfeeds, the healthier the next generation will be. The more sows eat, the more milk they have, which can feed the healthier next generation. On the contrary, if the piglets are more aggressive, on the one hand, a lot of activity will cause them to lose weight and have low growth performance. On the other hand, aggression will cause harm to other pigs and affect the growth of other pigs. Therefore, it is necessary to identify pigs with aggressive behaviors and eliminate them. At the same time, after the second camera scans the QR code, it further identifies the pig identification information in the QR code, such as the pig house number, sow stall number, piglet pen number, parent identification number, etc.

[0098] Step S6, forming a monitoring behavior record. The behavior of the sow and its identity information are associated to form a complete sow behavior record with identity information, and the record content at least includes the pig house number, sow stall number or piglet pen number, parent identity number, behavior recognition result image and category label. When breeding pigs, multiple samples of one variety are hybridized with multiple samples of another variety to select varieties carrying excellent genes. However, this hybridization cannot be completed in one generation, and it usually takes several generations of breeding to obtain an excellent variety. By identifying the identity information of the pig's parents, we can further understand the individual information of the pig's parents, such as the breed, growth performance, reproductive performance, characteristics and habits during the growth process, and can judge whether the pig has recessive gene expression through the parent information. Thus, a comprehensive breeding strategy for pigs is given.

[0099] Step S7, information transmission and visualization. Use simplejpeg and imagezmq technology to compress and package the behavior recognition result image and the sow behavior record with identity information, and send it to the monitoring system client submodule for real-time visual feedback via wireless transmission (the wireless network can be the pig house's own WIFI network or a local area network built by a bridge).

[0100] Step S8, monitoring task completion and restart. When the monitoring system completes a round of patrol monitoring tasks, it automatically returns to the initial starting point to charge and starts the countdown. When the countdown ends, the next round of patrol monitoring begins.

[0101] According to the present invention, a method for intelligent patrol and monitoring of breeding pig behavior is provided, wherein the coordinate points of the destinations that the monitoring system hardware device modules need to go to during automatic patrol are set in the constructed three-dimensional map, and the method also includes: predetermining the patrol mode in the pig house based on the electronic map, and the patrol mode includes: local patrol in the house and full patrol in the house; the local area patrol in the house is to patrol at least one area of ​​interest in the pig house; the full house patrol is to traverse and patrol the pig house as a whole.

[0102] In the second aspect, the present invention also provides an intelligent patrol monitoring system for breeding pig behavior, including: a monitoring system hardware device module, used to perform patrol movement within the house, information collection and data transmission; a system control and data processing module, used to process the collected information and visualize the results, and the processed information includes video images, behavior recognition result images and category label information, identity information obtained by recognizing QR codes, etc.

[0103] According to a breeding pig behavior intelligent patrol monitoring system provided by the present invention, the monitoring system hardware device module mainly comprises a four-wheel mobile chassis, a retractable bracket, a first camera, and a second camera.

[0104] The four-wheel mobile chassis adopts a design of front drive wheels and rear universal wheels. Compared with crawler, suspension, magnetic strip and other modes, it has the characteristics of light weight and flexible steering, and does not require the installation of additional auxiliary tracks.

[0105] The retractable bracket is installed in the center above the four-wheel mobile chassis and can be freely adjusted vertically according to the height of the pig house pen to ensure that the camera is at the optimal observation height.

[0106] The first camera and the second camera are installed on a telescopic bracket and can move with the four-wheel mobile chassis. Specifically, the first camera is arranged on the top of the telescopic bracket and the second camera is arranged on the bottom. The first camera is an unmanned vehicle airborne camera with an anti-shake gimbal. The camera monitoring height is adjusted by vertical lifting and lowering, and the camera gimbal is used to adjust the angle of the camera for observing the behavior of the breeding pigs so as to achieve the best collection effect. The second camera is an unmanned vehicle airborne camera with an anti-shake gimbal. The camera gimbal is used to adjust the angle of the camera for observing the QR code logo so as to achieve the best collection effect.

[0107] It should be noted that the anti-shake gimbal can greatly reduce the image ghosting problem caused by the camera's bumps and shakes when the four-wheel mobile chassis moves automatically.

[0108] The first camera and the second camera are electrically connected to the system control and data processing module to transmit the image data collected by the first camera and the second camera to the system control and data processing module for processing to identify the behavior and identity information of the breeding pigs.

[0109] The present invention provides a monitoring system hardware device module, which also includes an in-warehouse hardware module, which is deployed in a four-wheel mobile chassis.

[0110] The hardware modules in the warehouse include sensor modules, control modules, drive modules, and power modules. The sensor module is mainly composed of IMU, Lidar, and RGB-D camera, which are used to monitor the movement and posture data of the system hardware device module and the acquisition of road condition information; the control module is mainly composed of STM32, Jetson Nano, and Jetson AGX, which are used for electrical control, data processing, and graphics calculation; the drive module is mainly composed of motor drivers and motors with encoders, which are used to monitor the movement control of the system hardware device module; the power module is mainly equipped with lithium batteries to power each module.

[0111] According to the present invention, a sow behavior intelligent patrol monitoring system, a system control and data processing module, includes a navigation control submodule, a sow behavior recognition model, an identity information recognition submodule, an information receiving and sending submodule, and a monitoring system client submodule.

[0112] The navigation control submodule is used to guide the monitoring system hardware device module to complete the automatic planning movement during the pig house inspection;

[0113] The sow behavior recognition model is used to complete the real-time recognition of sow behaviors, such as standing, sitting, lying, suckling, feeding, etc. of sows in the farrowing house and aggressive behaviors such as biting ears, biting tails, climbing, and arching belly of piglets, and feedback the recognition results, including images of the behavior recognition results and behavior category labels;

[0114] The identity information recognition submodule is used to read the collected two-dimensional code image, identify the sow identity information therein (such as pig house number, stall number or pen number, parent identification number, etc.), and feed back the recognition result;

[0115] The information transceiver submodule is used to send the sow behavior recognition result and identity recognition result to the monitoring system client submodule, and at the same time receive the command sent by the monitoring system client submodule;

[0116] The monitoring system client submodule is used to receive the breeding pig behavior recognition results and identity information sent by the information transceiver submodule, and perform visual presentation and local automatic storage after data decoding, which is convenient for managers to view and trace, and can also send control commands to the information transceiver submodule.

[0117] Specifically, the steps of establishing the breeding pig behavior recognition model are as follows:

[0118] Step 1, data set preparation;

[0119] Step 2, standardize and annotate the data set. The data set is produced according to the UCF101-24 data set specification. Each video in the data set is extracted frame by frame, and the image frames from the same video are stored in the same folder. 1) For the behavior of sows in the farrowing house, the label making software is used to select and annotate the sow behavior in each frame of the image. When annotating, the whole pig needs to be selected and the behavior category is given. The behavior category includes but is not limited to the following items: standing, sitting, lying, suckling, and eating. 2) For the behavior of piglets in the nursery house, the label making software is used to select and annotate the aggressive behavior in each frame of the image. Since the aggressive behavior is an interactive behavior between two pigs, the piglets that carry out the attack and the piglets that are attacked need to be included in the annotation box, and the category of the aggressive behavior is given. The behavior category includes but is not limited to the following items: biting the ear, biting the tail, climbing, and arching the belly;

[0120] Step 3: Dataset division: randomly assign the labeled data to training set, test set and validation set;

[0121] Step 4: Model training to obtain the behavior recognition model of breeding pigs;

[0122] Step 5: Model deployment: deploy the breeding pig behavior recognition model to Jetson AGX.

[0123] Specifically, the specific steps of the pig behavior recognition model training include:

[0124] S4-1. Set network training hyperparameters, such as Epoch, Batchsize, initial learning rate, etc.

[0125] S5-2, put the prepared data set into the YOWOv2 network with the set network training hyperparameters to train the model;

[0126] S5-3. After the training is completed, based on the comprehensive evaluation of Precision and Recall, the weighted model with the best recognition effect is selected as the final breeding pig behavior recognition model.

[0127] Preferably, the specific steps of the breeding pig behavior recognition model training also include:

[0128] S5-4. Optimization of the pig behavior recognition model, including:

[0129] A. In the 2D backbone network of YOWOv2, the ELAN module in the original network is replaced with the GELAN module from YOLOv9, and RepConv is used in the GELAN module, so that the network can improve the feature representation learning ability without increasing the model calculation amount.

[0130] B. In the 3D backbone network of YOWOv2, FasterNet is improved to 3DFasterNet, and the original 3D backbone network of YOWOv2, 3DShuffleNetsv2, is replaced. The method of improving FasterNet to 3DFasterNet is to convert 2D convolution into 3D convolution and 2D pooling into 3D pooling. Specifically, in the pytorch framework, nn.Conv2d in Partial_conv is replaced with nn.Conv3d, and nn.AdaptiveAvgPool2d is replaced with nn.AdaptiveAvgPool3d. The improved network can extract spatial features more efficiently by reducing redundant calculations and memory accesses, which helps to achieve higher operating speeds on edge computing terminals without affecting the accuracy of visual tasks.

[0131] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-described intelligent patrol and monitoring methods for breeding pig behavior are implemented.

[0132] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described methods for intelligent patrol and monitoring of breeding pig behavior.

[0133] In order to make the above features and effects of the present invention more clearly and understandably described, embodiments are given below and described in detail with reference to the accompanying drawings. This specification discloses one or more embodiments that include the features of the present invention. The disclosed embodiments are only for illustration. The scope of protection of the present invention is not limited to the disclosed embodiments, and the present invention is defined by the attached claims.

[0134] Combine the following Figure 1-Figure 6 The intelligent patrol monitoring method and system for breeding pig behavior provided by the embodiments of the present invention are described. Figure 1 The figure is a flow chart of the method for intelligent inspection and monitoring of pig behavior provided by the present invention, including but not limited to the following steps:

[0135] Step S1, construct a QR code identifier

[0136] To build a QR code label, you first need to generate a QR code electronic image. You can choose to use the Python QR code generation library qrcode to generate a QR code electronic image with specific information. The generated QR code electronic image can be printed by a label printer to generate a QR code label. The information in the QR code label includes identity information such as the pig house number, stall number, pen number, and parent identification number.

[0137] Step S2: deploy QR code identification

[0138] The generated QR code logo needs to be deployed to each pen in the pig house. Specifically, for the farrowing house scenario, the QR code logo is fixed on the front baffle of each limit fence, at the junction of the two limit fences, so that the first camera is at the best shooting angle when the system moves to this position, and the height of the QR code logo is between 0.65 meters and 0.75 meters from the ground; for the nursery house scenario, the QR code logo is fixed in the middle of the front baffle of each piglet pen, so that the first camera is at the best shooting angle when the system moves to this position, and the height of the QR code logo is between 0.65 meters and 0.75 meters from the ground.

[0139] Step S3: Build a map in the house

[0140] The in-house map is mainly used to obtain the road condition information in the pig house before the system automatically navigates, so that the monitoring system can automatically plan navigation and movement based on the built electronic map. The main steps of in-house map building include: the staff controls the monitoring system hardware device module through the handheld remote control to traverse the working path of the entire pig house in advance, and uses laser radar, RGB-D camera and SLAM technology to traverse and scan and collect the road condition information of the pig house, forming a three-dimensional electronic map of the pig house scene, and storing it in the system control and data processing module for subsequent automatic navigation inspections.

[0141] Step S4, start automatic inspection

[0142] In the constructed three-dimensional map, set the destination coordinate point that the monitoring system hardware device module needs to go to during automatic patrol. The rviz software can be used to set the destination coordinate point in the three-dimensional map; place the monitoring system hardware device module at the navigation starting point, execute the start automatic navigation ROS command, and start the automatic patrol of the monitoring system. The monitoring system hardware device module will perform automatic path planning and mobile patrol on the pig house working path according to the navigation destination coordinate point.

[0143] Step S5, identification of pig behavior and identity information

[0144] During the automatic patrol process of the monitoring system hardware device module, the first camera is turned on to collect phenotypic video information of the breeding pigs in the pig confinement pens or pig pens along the way, and the second camera will scan the QR code logo fixed on the pig confinement pens or pig pen baffles along the way; the breeding pig behavior recognition model reads the pig phenotypic video information collected by the first camera and identifies the breeding pig behavior (including the location and category of the behavior), and after the second camera scans the QR code, it further identifies the breeding pig identity information in the QR code, such as the pig house number, sow confinement pens number, piglet pen number, parent identity number, etc.

[0145] Step S6: Forming monitoring behavior records

[0146] The behavior of the breeding pigs and their identity information are associated to form a complete breeding pig behavior record with identity information. The record content at least includes the pig house number, sow stall number or piglet pen number, parent identification number, behavior recognition result image and category label.

[0147] Step S7, information transmission and visualization

[0148] Use simplejpeg and imagezmq technology to compress and package the behavior recognition result images and the pig behavior records with identity information, and send them to the monitoring system client submodule for real-time visual feedback through wireless transmission (the wireless network can be the pig house's own WIFI network or a local area network built by a bridge);

[0149] Step S8, monitoring task completion and restart

[0150] When the monitoring system completes a round of patrol monitoring tasks, it automatically returns to the initial starting point to charge and begins the countdown. When the countdown ends, the next round of patrol monitoring begins.

[0151] Figure 2 It is a structural schematic diagram of a monitoring system hardware device module provided by the present invention, which mainly includes: a four-wheel mobile chassis 101, a retractable bracket 102, a first camera 103, and a second camera 104.

[0152] The four-wheel mobile chassis 101 adopts the design of front-wheel drive wheels and rear-wheel universal wheels. Compared with crawler-type, suspension-type, magnetic strip-type and other modes, it has the characteristics of lightness and flexible steering, and does not require the installation of additional auxiliary tracks.

[0153] The four-wheeled mobile chassis 101 is equipped with a positioning device, which is composed of a laser radar and an RGB-D camera. It can obtain the position information and road condition information of the four-wheeled mobile chassis 101 in the house, and three-dimensional mapping of the pig house road can be achieved through this information.

[0154] The retractable bracket 102 is installed in the center above the four-wheel mobile chassis 101, and is mainly used to support the first camera 103 and the second camera 104. The retractable bracket 102 can adjust the monitoring height of the camera by vertical lifting. The retractable bracket 102 can move with the four-wheel mobile chassis 101.

[0155] The first camera 103 is arranged at the top of the retractable bracket 102, and the second camera 104 is arranged at the bottom of the retractable bracket 102. The first camera 103 is a drone camera with an anti-shake gimbal, and the camera observation angle is adjusted by the camera gimbal to achieve the best acquisition effect of the collected picture; the second camera is a drone camera with an anti-shake gimbal, and the camera observation angle is adjusted by the camera gimbal to achieve the best acquisition effect of the collected picture.

[0156] It should be noted that the anti-shake gimbal can largely reduce the image ghosting problem caused by the bumping and shaking of the camera when the four-wheel mobile chassis 101 moves automatically. The first camera 103 and the second camera 104 are electrically connected to the system control and data processing module to transmit the data collected by the first camera 103 and the second camera 104 to the system control and data processing module, thereby identifying the behavior and identity information of the sow.

[0157] The present invention provides a monitoring system hardware device module, which also includes an in-warehouse hardware module, which is deployed in a four-wheel mobile chassis 101.

[0158] Figure 3 The present invention provides a schematic diagram of the hardware module composition in the warehouse of a monitoring system hardware device module, including a sensor module, a control module, a drive module, and a power module. Among them, the sensor module is mainly composed of IMU, Lidar and RGB-D camera, which is used to monitor the movement and posture data of the system hardware device module and the acquisition of road condition information; the control module is mainly composed of STM32, Jetson Nano, Jetson AGX, which is used for electrical control, data processing and graphics calculation; the drive module is mainly composed of a motor driver and a motor with an encoder, which is used for the movement control of the monitoring system hardware device module; the power module is mainly equipped with a lithium battery to power each module.

[0159] Figure 4 It is a schematic diagram of the composition of the system control and data processing module, including the navigation control submodule, the sow behavior recognition model, the identity information recognition submodule, the information receiving and sending submodule, and the monitoring system client submodule.

[0160] The navigation control submodule is used to guide the monitoring system hardware device module to complete the automatic planning and movement during the pig house inspection. The navigation control submodule also includes an electronic map, which can be generated by the above method. The electronic map generated by the pig house mapping is stored in advance, and the monitoring system hardware device module can be guided to walk in the pig house according to the electronic map generated by the pig house mapping, and has multiple cruise modes such as global area cruise and local area cruise;

[0161] The breeding pig behavior recognition model is used to complete the real-time recognition of sow behaviors (such as standing, sitting, lying, suckling, eating, etc.) and piglet aggressive behaviors (such as biting ears, biting tails, climbing, arching belly, etc.) in the farrowing house, and feedback the recognition results;

[0162] The identity information recognition submodule is used to read the collected QR code image, identify the breeding pig identity information (such as pig house number, stall number or pen number, parent identification number, etc.), and feedback the recognition result.

[0163] The information transceiver submodule is used to send behavior recognition results and identity recognition results to the monitoring system client submodule, and receive commands sent by the monitoring system client submodule;

[0164] The monitoring system client submodule is used to receive the identification results and identity information sent by the information transceiver submodule, and visualize and automatically store the data after data decoding, which is convenient for managers to view and trace, and can also send control commands to the information transceiver submodule. Figure 5 As shown, it includes a behavior recognition window, an identity code scanning window, a behavior snapshot window and a record list window.

[0165] In the system control and data processing module, the steps for establishing the breeding pig behavior recognition model are as follows:

[0166] Step 1, data set preparation;

[0167] Step 2, standardize and annotate the data set. The data set is produced according to the UCF101-24 data set specification. Each video in the data set is extracted frame by frame, and the image frames from the same video are stored in the same folder. 1) For the behavior of sows in the farrowing house, the label making software is used to select and annotate the sow behavior in each frame of the image. When annotating, the whole pig needs to be selected and the behavior category is given. The behavior category includes but is not limited to the following items: standing, sitting, lying, suckling, and eating. 2) For the behavior of piglets in the nursery house, the label making software is used to select and annotate the aggressive behavior in each frame of the image. Since the aggressive behavior is an interactive behavior between two pigs, the piglets that carry out the attack and the piglets that are attacked need to be included in the annotation box, and the category of the aggressive behavior is given. The behavior category includes but is not limited to the following items: biting the ear, biting the tail, climbing, and arching the belly;

[0168] Step 3: Dataset division: randomly assign the labeled data to training set, test set and validation set;

[0169] Step 4: Model training to obtain the behavior recognition model of breeding pigs;

[0170] Step 5: Model deployment: deploy the breeding pig behavior recognition model to Jetson AGX.

[0171] Specifically, the specific steps of the pig behavior recognition model training include:

[0172] S4-1. Set network training hyperparameters, such as Epoch, Batchsize, initial learning rate, etc.

[0173] S5-2, put the prepared data set into the YOWOv2 network with the set network training hyperparameters to train the model;

[0174] S5-3. After the training is completed, based on the comprehensive evaluation of Precision and Recall, the weighted model with the best recognition effect is selected as the final breeding pig behavior recognition model.

[0175] Preferably, the specific steps of the breeding pig behavior recognition model training also include:

[0176] S5-4. Optimization of the pig behavior recognition model, including:

[0177] Preferably, the specific process of lightweighting the S1-4 model is:

[0178] A. In the 2D backbone network of YOWOv2, the ELAN module in the original network is replaced with the GELAN module from YOLOv9, and RepConv is used in the GELAN module, so that the network can improve the feature representation learning ability without increasing the model calculation amount.

[0179] B. In the 3D backbone network of YOWOv2, FasterNet is improved to 3DFasterNet, and the original 3D backbone network of YOWOv2, 3DShuffleNetsv2, is replaced. The method of improving FasterNet to 3DFasterNet is to convert 2D convolution into 3D convolution and 2D pooling into 3D pooling. Specifically, in the pytorch framework, nn.Conv2d in Partial_conv is replaced with nn.Conv3d, and nn.AdaptiveAvgPool2d is replaced with nn.AdaptiveAvgPool3d. The improved network can extract spatial features more efficiently by reducing redundant calculations and memory accesses, which helps to achieve higher operating speeds on edge computing terminals without affecting the accuracy of visual tasks.

[0180] The schematic diagram of the optimized improved YOWOv2 network structure is as follows Figure 6 shown.

[0181] In the system control and data processing module, the identity information recognition submodule can use opencv combined with pyzbar technology to complete the scanning and information recognition of the QR code. The implementation process is as follows:

[0182] The first step is to install the required libraries, such as opencv and pyzbar;

[0183] The second step is to import necessary libraries, such as opencv and pyzbar;

[0184] The third step is to control the program to open the camera;

[0185] Step 4: The control program reads and processes each frame of the image;

[0186] Step 5: Detect and parse the QR code in the image frame;

[0187] Step 6: Decode the QR code and output it in "utf-8" format.

[0188] The following is a system embodiment corresponding to the above method embodiment. This embodiment can be implemented in conjunction with the above embodiment. The relevant technical details mentioned in the above embodiment are still valid in this embodiment. In order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the above embodiment.

[0189] like Figure 8 As shown, the present invention also proposes a pig behavior intelligent patrol monitoring device, which includes:

[0190] The identification building module generates an identification for the breeding pig according to the identity of the breeding pig to be monitored;

[0191] The identification deployment module deploys the identity identification at a designated location of the pen where the pig is located in the pig house;

[0192] The in-house mapping module collects the electronic map of the working path in the pig house;

[0193] The automatic inspection module sets the destination coordinate point of the monitoring system hardware device during automatic inspection in the electronic map, starts automatic navigation of the monitoring system hardware device according to its own position and the electronic map, and starts the automatic inspection of the monitoring system. The monitoring system hardware device performs automatic path planning and mobile inspection on the working path in the pig house according to the destination coordinate point;

[0194] Among them, the mobile inspection in the automatic inspection module includes:

[0195] During the mobile patrol process, the camera of the monitoring system hardware device is turned on to collect the phenotypic video of the breeding pig in the pens along the road, and scans and identifies the identity tag to obtain the breeding pig identity information; the breeding pig behavior recognition model reads the pig phenotypic video and identifies the breeding pig behavior information including the location and behavior category of the behavior; the breeding pig behavior information and its breeding pig identity information are associated as the pig behavior patrol monitoring result.

[0196] The intelligent inspection and monitoring device for breeding pig behavior, wherein the monitoring system hardware device comprises:

[0197] Wheeled mobile chassis, retractable bracket, at least one camera;

[0198] The four-wheel mobile chassis adopts the design of front drive wheels and rear universal wheels;

[0199] The retractable bracket is installed in the center above the four-wheel mobile chassis and can be freely adjusted vertically according to the height of the pig pen.

[0200] The camera is installed on a telescopic bracket; the camera is electrically connected to the data processing module of the monitoring system hardware device to transmit the phenotypic video and the identity identification collected by the camera to the data processing module for processing, so as to identify the behavior information and identity information of the breeding pig.

[0201] The intelligent inspection and monitoring device for breeding pig behavior, wherein the breeding pig behavior information includes: suckling behavior, feeding behavior, and piglet attack behavior;

[0202] The intelligent inspection and monitoring device for breeding pig behavior also includes: a breeding module, which selects multiple breeding pigs of various breeds that meet breeding requirements according to the inspection and monitoring results, and crossbreeds multiple breeding pigs of one breed with multiple breeding pigs of another breed to breed breeds carrying excellent genes;

[0203] The construction process of the pig behavior recognition model includes:

[0204] The ELAN module in the 2D backbone network of YOWOv2 is replaced by the GELAN module from YOLOv9, and convolution RepConv is used in the GELAN module to obtain an intermediate model, so that the network can improve the feature representation learning ability without increasing the model calculation amount;

[0205] The 2D convolution of FasterNet in the 3D backbone network of the intermediate model is converted into 3D convolution, and the 2D pooling is converted into 3D pooling to obtain the pig behavior recognition model.

[0206] like Fig. 9 As shown, the present invention further proposes a first electronic device A in another embodiment, including the above-mentioned intelligent patrol and monitoring device for breeding pig behavior.

[0207] like Fig.10 As shown, the first electronic device A can also be connected to the data acquisition device C and the information display device D through a wired or wireless information transmission scheme, the data acquisition device C is used to collect videos, such as the phenotypic video described in the embodiment of the present invention, and the information display device D is used to display the pig behavior inspection and monitoring results obtained by the analysis of the present invention.

[0208] The information display device D can sort and process the data output by the first electronic device A based on the information display mechanism to improve the readability of the data output by the first electronic device A. The information display mechanism can be manually preset, for example, the data output by the first electronic device A is visually displayed, which can be based on the display parameters and / or attributes set by the user. The display parameters can be, for example, the display data range, and the display attributes can be, for example, the display font, color, whether to scroll and play, etc. The user is presented with the key information specified by the user, such as breastfeeding behavior, feeding behavior, piglet attack behavior, etc. The user can understand this information more timely without having to access the secondary page or scroll the page, saving the user's operation. Or the information display mechanism can be an artificial intelligence AI display model, which can learn the user's key information based on the user's previous usage habits, such as viewing time, number of clicks, number of edits, etc., and then automatically present rich and necessary key information to the user.

[0209] The present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a readable storage medium. When the computer program is executed by a processor, the computer can execute the intelligent patrol and monitoring methods for breeding pig behavior provided by the above methods.

[0210] The present invention also proposes a storage medium VIII in another embodiment for storing a computer program for executing the intelligent patrol monitoring method of the breeding pig behavior. It should be understood that the storage medium in the embodiment of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DRRAM).

[0211] Fig.11 A schematic block diagram of a second electronic device 1000 that can be used to implement an embodiment of the present invention is shown. The second electronic device 1000 electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The second electronic device 1000 can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present invention described and / or required herein. The second electronic device 1000 may be the same or different from the first electronic device A.

[0212] The second electronic device 1000 includes a computing unit I, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory II (ROM) or a computer program loaded from a storage medium VIII into a random access memory (RAM) III. In RAM III, various programs and data required for the operation of the second electronic device 1000 can also be stored. The computing unit I, ROM II, and RAM III are connected to each other via a bus IV. An input / output (I / O) interface V is also connected to the bus IV.

[0213] A plurality of components in the second electronic device 1000 are connected to the I / O interface V, including: an input unit VI, such as a keyboard, a mouse, etc.; an output unit VII, such as various types of displays, speakers, etc.; a storage medium VIII, such as a disk, an optical disk, etc.; and a communication unit IX, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit IX allows the second electronic device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0214] The computing unit I may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit I include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit I performs the various methods and processes described above, such as method steps S1-S4. For example, in some embodiments, the method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage medium VIII. In some embodiments, part or all of the computer program may be loaded and / or installed on the second electronic device 1000 via ROM II and / or communication unit IX. When the computer program is loaded into RAM III and executed by the computing unit I, one or more steps of the method described above may be performed. Alternatively, in other embodiments, the computing unit I may be configured to execute the method in any other appropriate manner (e.g., by means of firmware).

[0215] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

Claims

1. A method for intelligent inspection and monitoring of breeding pig behavior, characterized in that: include: The identification construction step generates an identification for the breeding pig according to the identity of the breeding pig to be monitored; The identification deployment step is to deploy the identification tag at a designated location of the pen where the pig is located in the pig house; In-house mapping step, collect the electronic map of the working path in the pig house; The automatic inspection step includes setting the destination coordinate point of the monitoring system hardware device during the automatic inspection in the electronic map, starting the automatic navigation of the monitoring system hardware device according to its own position and the electronic map, and starting the automatic inspection of the monitoring system. The monitoring system hardware device performs automatic path planning and mobile inspection on the working path in the pig house according to the destination coordinate point; Among them, the mobile inspection in the automatic inspection step includes: During the mobile patrol process, the camera of the monitoring system hardware device is turned on to collect the phenotypic video of the breeding pig in the pens along the road, and scans and identifies the identity tag to obtain the breeding pig identity information; the breeding pig behavior recognition model reads the pig phenotypic video and identifies the breeding pig behavior information including the location and behavior category of the behavior; the breeding pig behavior information and its breeding pig identity information are associated as the pig behavior patrol monitoring result.

2. The method for intelligent inspection and monitoring of breeding pig behavior as claimed in claim 1, characterized in that: The monitoring system hardware includes: Wheeled mobile chassis, retractable bracket, at least one camera; The wheeled mobile chassis adopts the design of front drive wheel and rear universal wheel; The retractable bracket is installed in the center above the wheeled mobile chassis and can be freely adjusted vertically according to the height of the pig pen; The camera is installed on a telescopic bracket; the camera is electrically connected to the data processing module of the monitoring system hardware device to transmit the phenotypic video and the identity identification collected by the camera to the data processing module for processing, so as to identify the behavior information and identity information of the breeding pig.

3. The method for intelligent inspection and monitoring of breeding pig behavior as claimed in claim 1, characterized in that: The behavior information of the breeding pigs includes: suckling behavior, feeding behavior, and piglet aggressive behavior; The intelligent inspection and monitoring method for breeding pig behavior also includes: a breeding step, according to the inspection and monitoring results, selecting multiple breeding pigs of each breed that meet the breeding requirements, and hybridizing multiple breeding pigs of one breed with multiple breeding pigs of another breed to select breeds carrying excellent genes.

4. The method for intelligent inspection and monitoring of breeding pig behavior according to claim 1, characterized in that: The construction process of the pig behavior recognition model includes: The ELAN module in the 2D backbone network of YOWOv2 is replaced by the GELAN module from YOLOv9, and convolution RepConv is used in the GELAN module to obtain an intermediate model, so that the network can improve the feature representation learning ability without increasing the model calculation amount; The 2D convolution of FasterNet in the 3D backbone network of the intermediate model is converted into 3D convolution, and the 2D pooling is converted into 3D pooling to obtain the pig behavior recognition model.

5. An intelligent inspection and monitoring device for breeding pig behavior, characterized in that: include: The identification building module generates an identification for the breeding pig according to the identity of the breeding pig to be monitored; The identification deployment module deploys the identity identification at a designated location of the pen where the pig is located in the pig house; The in-house mapping module collects the electronic map of the working path in the pig house; The automatic inspection module sets the destination coordinate point of the monitoring system hardware device during automatic inspection in the electronic map, starts automatic navigation of the monitoring system hardware device according to its own position and the electronic map, and starts the automatic inspection of the monitoring system. The monitoring system hardware device performs automatic path planning and mobile inspection on the working path in the pig house according to the destination coordinate point; Among them, the mobile inspection in the automatic inspection module includes: During the mobile patrol process, the camera of the monitoring system hardware device is turned on to collect the phenotypic video of the breeding pig in the pens along the road, and scans and identifies the identity tag to obtain the breeding pig identity information; the breeding pig behavior recognition model reads the pig phenotypic video and identifies the breeding pig behavior information including the location and behavior category of the behavior; the breeding pig behavior information and its breeding pig identity information are associated as the pig behavior patrol monitoring result.

6. The intelligent inspection and monitoring device for breeding pig behavior as claimed in claim 5, characterized in that: The monitoring system hardware includes: Wheeled mobile chassis, retractable bracket, at least one camera; The wheeled mobile chassis adopts the design of front drive wheel and rear universal wheel; The retractable bracket is installed in the center above the wheeled mobile chassis and can be freely adjusted vertically according to the height of the pig pen; The camera is installed on a telescopic bracket; the camera is electrically connected to the data processing module of the monitoring system hardware device to transmit the phenotypic video and the identity identification collected by the camera to the data processing module for processing, so as to identify the behavior information and identity information of the breeding pig.

7. The intelligent inspection and monitoring device for breeding pig behavior as claimed in claim 5, characterized in that: The behavior information of the breeding pigs includes: suckling behavior, feeding behavior, and piglet aggressive behavior; The intelligent inspection and monitoring device for breeding pig behavior also includes: a breeding module, which selects a plurality of breeding pigs of various breeds that meet the breeding requirements according to the inspection and monitoring results, and crossbreeds a plurality of breeding pigs of one breed with a plurality of breeding pigs of another breed to breed a breed carrying excellent genes; The construction process of the pig behavior recognition model includes: The ELAN module in the 2D backbone network of YOWOv2 is replaced by the GELAN module from YOLOv9, and convolution RepConv is used in the GELAN module to obtain an intermediate model, so that the network can improve the feature representation learning ability without increasing the model calculation amount; The 2D convolution of FasterNet in the 3D backbone network of the intermediate model is converted into 3D convolution, and the 2D pooling is converted into 3D pooling to obtain the pig behavior recognition model.

8. An electronic device, characterized in that: It comprises the intelligent patrol and monitoring device for breeding pig behavior as described in any one of claims 5-7, and the electronic device is connected to an information display device, and the information display device is used to display the pig behavior patrol and monitoring results according to the display parameters and attributes set by the user or through an artificial intelligence model.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for intelligent patrol and monitoring of breeding pig behavior described in any one of claims 1-4 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for intelligent patrol and monitoring of breeding pig behavior described in any one of claims 1-4 are implemented.

Images