Low-yield chicken coop distinguishing system and method, electronic equipment and storage medium

Abstract

The invention provides a low-yield coop discrimination system and method, electronic equipment and a storage medium, and relates to the technical field of image processing, and the system comprises a visual system, chassis equipment and power supply equipment; the power supply device is electrically connected with the chassis device and the visual device. The visual system is arranged on the chassis device. Wherein the chassis equipment comprises a controller, a navigation module, a remote control module and a driving module, the vision system comprises a fixed bracket, at least one camera, light supplementing equipment and an edge calculator, the camera is used for collecting a coop image of a coop to be detected, and the edge calculator is used for automatically dividing and counting the coop image; and obtaining the number of chickens and the number of eggs of the to-be-detected hencoop, and transmitting the number of chickens and the number of eggs of the to-be-detected hencoop to a cloud server, so that the cloud server determines whether the to-be-detected hencoop is a low-yield hencoop according to the number of chickens and the number of eggs of the to-be-detected hencoop in a preset time period.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a low-yield chicken cage identification system, method, electronic equipment and storage medium. Background Art

[0002] Eggs are an important source of high-quality protein for humans. In the daily management of large-scale chicken farms, in order to improve production efficiency, it is necessary to regularly evaluate the egg production and health of the chickens. Timely identification and elimination of low-yielding chickens is the key to achieving efficient production. However, the current identification of low-yielding chickens mainly relies on manual monitoring, that is, judging the production capacity of chickens by recording and observing the egg production of each chicken cage.

[0003] Therefore, how to identify low-laying hens more efficiently has become a problem that needs to be solved urgently in the industry. Summary of the invention

[0004] The present invention provides a low-production chicken cage identification system, method, electronic equipment and storage medium, which are used to solve the problem of how to more efficiently identify low-production laying hens in the prior art.

[0005] The present invention provides a low-yield chicken cage identification system, comprising: a visual system, a chassis device and a power supply device; the power supply device is electrically connected to the chassis device and the visual device, and the visual system is arranged on the chassis device; Wherein, the chassis equipment includes: a controller, a navigation module, a remote control module and a driving module, and the controller is respectively connected to the navigation module, the remote control module, the driving module and the visual system for communication; The controller is used to generate a driving control instruction according to the information transmitted by the navigation module and the remote control module, and the driving module is used to control the movement of the chassis device according to the driving control instruction; The visual system includes: a fixed bracket, at least one camera, a fill light device and an edge calculator, wherein the fixed bracket is fixed above the chassis device, and the camera, the fill light device and the edge calculator are all fixed on the fixed bracket; the camera is used to collect the image of the chicken cage to be tested, and the edge calculator is used to automatically count the image of the chicken cage to obtain the number of chickens and eggs in the chicken cage to be tested, and transmit the number of chickens and eggs in the chicken cage to be tested to the cloud server, so that the cloud server can determine whether the chicken cage to be tested is a low-yield chicken cage based on the number of chickens and eggs in the chicken cage to be tested within a preset time period.

[0006] According to a low-yield chicken cage identification system provided by the present invention, the navigation module includes: a geomagnetic sensor, a site sensor and an obstacle avoidance sensor; The geomagnetic sensor is arranged at the bottom of the chassis device, and is used to detect the preset magnetic track signal in real time when in working state, and the controller outputs the speed of the driving wheel after calculation and processing to keep it in the track state; The site sensor locates the equipment through the detected site position, so as to realize the turning and U-turn of the chassis equipment; Wherein, the obstacle avoidance sensor is used to implement emergency braking of the chassis equipment when an obstacle is detected.

[0007] According to a low-yield chicken cage identification system provided by the present invention, the remote control module includes: a receiver and a controller; The receiver is used to receive control instructions and implement manual control of the chassis equipment according to the controller.

[0008] According to a low-yield chicken cage identification system provided by the present invention, the driving module specifically includes: a servo motor, a driver, a reducer, a driving wheel and a driven wheel; the driving wheels are respectively arranged on the left and right sides of the chassis of the chassis equipment, and the driven wheels are arranged on the front and rear sides of the chassis of the chassis equipment.

[0009] The driver is used to drive the communication connection with the controller, adjust the speed of the servo motor by proportional-integral-differential control, and increase the torque through the reducer to drive the driving wheel to rotate; The driven wheel utilizes a universal structure to assist the device in turning and adjusting its posture.

[0010] According to a low-yield chicken cage identification system provided by the present invention, the visual system specifically includes: a first camera, a second camera, a third camera, a fill light device, an edge calculator and a data display screen; The setting position of the first camera corresponds to the height of the first layer of chicken cages; the setting position of the second camera corresponds to the height of the second layer of chicken cages; the setting position of the third camera corresponds to the height of the third layer of chicken cages; Wherein, the fill light is used to fill light for the shooting area of ​​each camera; Among them, the data display screen is used to display the chicken cage images collected by each camera.

[0011] According to a low-yield chicken cage identification system provided by the present invention, the edge calculator is specifically used for: The YOLOv8n-ML-DE model is used to identify the chickens, eggs and the QR code of the chicken cage to be tested in the chicken cage image, and the number of the chicken cage to be tested is determined according to the QR code of the chicken cage; According to the coordinate information of the chicken cage QR code, the counting region of interest corresponding to the chicken cage to be tested is determined, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The serial number of the chicken cage to be tested and the cage count are uploaded to the cloud server together, so that the cloud server can determine whether the chicken cage to be tested is a low-production chicken cage based on the number of chickens and eggs in the chicken cage to be tested within a preset time period.

[0012] According to a low-yield chicken cage identification system provided by the present invention, the YOLOv8n-ML-DE model specifically refers to: Replace the Ciou loss function of the YOLOv8n model with Wiou(v2), add the MLCA attention mechanism to the lower layer of the spatial pooling pyramid SPPF of the backbone part, and replace the Detect of the Head with Efficient_Detect.

[0013] The present invention also provides a low-yield chicken cage identification method based on the low-yield chicken cage identification system, comprising: The chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instructions or the preset inspection path; The camera in the visual system is used to collect images of the chicken cages to be inspected. The edge calculator in the visual system uses the YOLOv8n-ML-DE model to identify the chickens, eggs, and the QR code of the chicken cages to be inspected in the images of the chicken cages, and determines the number of the chicken cages to be inspected according to the QR code of the chicken cages. The edge calculator determines the counting region of interest corresponding to the chicken cage to be tested according to the coordinate information of the chicken cage QR code, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The wireless transmission device in the visual system uploads the serial number of the chicken cage to be tested and the cage count to the cloud server, so that the cloud server determines whether the chicken cage to be tested is a low-yield chicken cage according to the cage count of the chicken cage to be tested.

[0014] The present invention also 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 computer program, any of the above-mentioned methods for identifying low-yield chicken cages is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for identifying a low-yield chicken cage as described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for identifying a low-yield chicken cage as described above is implemented.

[0017] The low-yield chicken cage identification system, method, electronic device and storage medium provided by the present invention, the chassis equipment includes a controller, a drive module, a navigation module and a remote control module, which is suitable for the multi-layer inspection needs of three-dimensional cage farms. The controller realizes precise control of the motor by communicating with the servo driver; the drive module uses PID regulation and a reducer to increase the torque to achieve efficient and stable movement of the chassis; the geomagnetic, station and obstacle avoidance sensors in the navigation module ensure that the equipment travels stably along the track and turns safely; the remote control module supports flexible switching between automatic navigation and manual control. The visual system integrates data acquisition, processing and real-time display functions. The edge calculator has deep learning acceleration capabilities and supports real-time processing and model reasoning of multi-channel video data; multiple cameras are used to collect three-layer chicken cage images, and the data is wirelessly transmitted to the cloud via WiFi. The fill light device ensures the clarity of the image in a low-light environment, and the high-definition touch screen displays the recognition results and status information, which is convenient for users to view and manage in real time. In addition, an improved YOLOv8n-ML-DE model is proposed. By replacing the Ciou loss function with Wiou (v2) to improve the bounding box accuracy, the MLCA attention mechanism is added to enhance the small target detection capability, and the lightweight Efficient_Detect module is used to reduce the computational complexity to ensure that the model runs efficiently on edge devices. At the same time, a three-layer chicken cage counting algorithm is designed. By analyzing the egg production rate data for 10 consecutive days, low-yield chicken cages are automatically identified, significantly reducing the need for manual operation. It provides strong support for the automatic identification and cage counting of low-yield chickens, meeting the efficient management needs of large-scale farms. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a schematic diagram of the system architecture of the low-yield chicken cage identification system provided by the present invention; Figure 2 A schematic diagram of the overall structure of the low-yield chicken cage identification system provided by the present invention; Figure 3 A side view of the low-yield chicken cage identification system provided by the present invention; Figure 4 A bottom view of the low-yield chicken cage identification system provided by the present invention; Figure 5 A schematic diagram of the inspection provided by the present invention; Figure 6A schematic diagram of the YOLOv8n-ML-DE network framework structure provided by the present invention; Figure 7 Schematic diagram of the process flow of the low-yield chicken cage identification method provided by the present invention Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Figure 1 Schematic diagram of the system architecture of the low-yield chicken cage identification system provided by the present invention. Figure 1 As shown, it includes: a visual system 11, a chassis device 12 and a power supply device 13; the power supply device 13 is electrically connected to the chassis device and the visual device, and the visual system is arranged on the chassis device; The chassis device 12 includes: a controller, a navigation module, a remote control module and a driving module, and the controller is respectively connected to the navigation module, the remote control module, the driving module and the visual system for communication; The controller is used to generate a driving control instruction according to the information transmitted by the navigation module and the remote control module, and the driving module is used to control the movement of the chassis device according to the driving control instruction; Among them, the visual system 11 includes: a fixed bracket, at least one camera, a fill light device and an edge calculator, the fixed bracket is fixed above the chassis device, and the camera, the fill light device and the edge calculator are all fixed on the fixed bracket; the camera is used to collect the chicken cage image of the chicken cage to be tested, and the edge calculator is used to automatically count the chicken cage image to obtain the number of chickens and eggs in the chicken cage to be tested, and transmit the number of chickens and eggs in the chicken cage to be tested to the cloud server, so that the cloud server can determine whether the chicken cage to be tested is a low-yield chicken cage based on the number of chickens and eggs in the chicken cage to be tested within a preset time period.

[0022] In the present invention, the visual system is responsible for image acquisition and processing, including a camera, a fill light device and an edge calculator. The chassis device provides mobility, including a controller, a navigation module, a remote control module and a drive module. The power supply device provides power for the chassis device and the visual system.

[0023] In the present invention, the controller is the center of the system, responsible for receiving information from the navigation module and the remote control module, and generating driving control instructions according to the information.

[0024] The navigation module contains sensors and algorithms to locate the position of the chassis equipment and plan the moving path. The remote control module allows the operator to remotely control the chassis equipment when necessary, providing the ability for manual intervention. The drive module: performs specific movement operations according to the instructions of the controller, including hardware such as motors and drives.

[0025] In the present invention, a fixed bracket is installed above the chassis device and is used to fix the camera, the fill light device and the edge calculator. The camera is used to collect images of the chicken cage to be tested and is the data source of the visual system. The fill light device is used to provide auxiliary lighting when the lighting conditions are not ideal to ensure that the camera can clearly collect images. The edge calculator is used to process and analyze the images collected by the camera to achieve automatic cage counting.

[0026] In the present invention, the camera collects the image of the chicken cage, and the fill light device provides lighting as needed. The edge computer receives the image collected by the camera and uses the automatic cage counting algorithm to process the image to identify the chickens and eggs.

[0027] The edge computer transmits the identification results (number of chickens and eggs) to the cloud server. The cloud server analyzes the production performance of each chicken cage based on the data within the preset time period and identifies the low-yield chicken cages. The controller generates drive control instructions based on the location information provided by the navigation module and the instructions of the remote control module. The drive module executes these instructions and controls the chassis equipment to move to the next chicken cage to be tested.

[0028] Figure 2 This is a schematic diagram of the overall structure of the low-yield chicken cage identification system provided by the present invention. Figure 3 A side view of the low-yield chicken cage identification system provided by the present invention, Figure 4 The bottom bottom view of the low-yield chicken cage identification system provided by the present invention is as follows: Figure 2-4 As shown, it includes: a camera 1; a fill light device 2; a control panel 3; a controller 4; a motor driver 5; a power supply 6; a transformer 7; a data display screen 8; a control switch 9; a wireless network card 10; an infrared sensor 11; a wheel drum motor 12; an edge calculator 13; a geomagnetic sensor 14; and a site sensor 15.

[0029] The equipment uses geomagnetic navigation and positioning technology, wheeled chassis movement, laser radar obstacle avoidance, and three cameras with adjustable height and angle are used to identify the data information of the upper, middle and lower layers of chicken cages. The data display screen displays real-time identification information and uploads it to the cloud platform via Wifi. The chassis equipment radius is 25cm and the total height is 186.6cm. The equipment is mainly composed of chassis structure and visual system.

[0030] The chassis equipment consists of a controller, a drive module, a navigation module and a remote control module. The core controller uses STM32F407 as the core chip, and performs processing and output control by collecting data information from each module. The drive module includes a servo motor, a drive, a reducer and a master / slave wheel. The drive communicates and interacts with the controller through CAN, uses PID to adjust the motor speed, and increases the torque through the reducer, and finally achieves the target speed of the chassis equipment moving efficiently and stably. The power is transmitted by two active wheels located on the left and right sides of the chassis, and steering is achieved through the differential movement of the motor; the two driven wheels located in front and behind the chassis use a universal structure to assist the chassis equipment in turning and adjusting its posture. The navigation module includes a geomagnetic sensor, a station sensor and an obstacle avoidance sensor. Among them, the geomagnetic sensor is installed at the bottom of the chassis equipment. When in working state, it detects the preset magnetic track signal in real time. After the controller calculates and processes, it outputs the speed of the active wheel and keeps it in the track state. The station sensor can deduce the chicken house position where the chassis equipment is located through the detected station position, so as to realize the turning and U-turn of the chassis equipment. The obstacle avoidance sensor realizes emergency braking to ensure personnel safety. The remote control module consists of a remote controller and a receiver. The control switch can switch between automatic navigation and remote control to ensure the movement of chassis equipment in non-track environments. The edge computer can send walking instructions to realize the timed inspection function.

[0031] In the present invention, the visual system is equipped with three cameras, which are used to collect images of three layers of chicken cages. Each camera is targeted at a specific chicken cage layer to ensure that all chicken cages can be covered. The three cameras work simultaneously to capture real-time images of the chicken cage layer they are responsible for. These images contain information such as the location and number of chickens and eggs.

[0032] To ensure clear images in all lighting conditions, the device is equipped with an LED light bar as a fill light. This helps improve image quality, especially in low-light environments.

[0033] The images captured by the camera are transmitted to an edge computer, a powerful computing device that can pre-process and analyze the images locally, reducing the amount of data that needs to be transmitted to the cloud.

[0034] The data processed by the edge computer is transmitted to the Alibaba Cloud server through wireless transmission equipment and Wi-Fi network. This allows the powerful computing resources of the cloud to be used for further data analysis and storage.

[0035] The data display screen displays the video and recognition images captured by the camera in real time, as well as the results and related information processed by the edge computer. This provides operators with an intuitive interface for monitoring the working status of the equipment and the real-time situation of the chicken cages.

[0036] The data received by the cloud server is used to analyze the production performance of each chicken cage. By comparing the changes in the number of chickens and eggs over a preset time period, the cloud server can identify low-yielding chicken cages.

[0037] Specifically, the egg production rate is the average egg production percentage of a group in a certain period (day, week, month), which is generally calculated on the basis of feeding days and indicates the egg production intensity of the group. The egg production rate of each chicken cage reflects the egg production performance of a small group within a chicken cage, which can be calculated by continuously counting the egg production and number of chickens in the chicken cage for many days.

[0038] In the present invention, according to the egg production rate standard, an egg production rate of 80% to 90% is considered to be a high egg production rate, so chicken cages with an egg production rate below 80% are judged as low-production chicken cages in this article.

[0039] ; Figure 5 The inspection schematic diagram provided by the present invention is as follows: Figure 5 Figure 2 provides a visual representation of the device's inspection in the chicken cage, showing how the device moves and covers all chicken cage layers, and how data is collected by the image acquisition unit.

[0040] Optionally, the edge calculator is specifically used for: The YOLOv8n-ML-DE model is used to identify the chicken cage QR code, the number of chickens, and the number of eggs in the chicken cage to be tested in the chicken cage image, and the number of the chicken cage to be tested is determined according to the chicken cage QR code; According to the coordinate information of the chicken cage QR code, the counting region of interest corresponding to the chicken cage to be tested is determined, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The serial number of the chicken cage to be tested and the cage count are uploaded to the cloud server together, so that the cloud server can determine whether the chicken cage to be tested is a low-production chicken cage based on the number of chickens and eggs in the chicken cage to be tested within a preset time period.

[0041] In the present invention, the edge computer uses the YOLOv8n-ML-DE model to process the chicken coop images captured by the camera. The model can recognize the QR code in the image and the number of chickens and eggs. Through QR code recognition, the system can determine the unique number of each chicken coop to be tested.

[0042] More specifically, the present invention designs a cage counting method (CDC) for automatically counting chickens and eggs. The method is divided into two stages: the first stage is to set the counting region of interest (CRoI), in which we use the YOLOv8n-ML-DE model to detect targets in motion to ensure accurate identification of the chicken and egg regions. When two QR codes are identified, we calculate and divide the CRoI of chickens and eggs according to their coordinates, thereby achieving accurate region division. The cage counting calculation process of the upper and middle layers In the formula, The chicken coop number stored in the QR code; and They are the horizontal coordinate of the upper vertex and the vertical coordinate of the lower vertex on the left side of the QR code detection box respectively; and are the minimum and maximum functions respectively; is the horizontal coordinate range of the counting area; and The vertical coordinate value ranges for the chicken and egg counting areas are respectively. It should be noted that since the fixed position of the QR code cards in the lower layer is different from that in the first two rows, when calculating the CRoI of each chicken and egg, 100 pixels need to be added to the original basis to achieve accurate counting.

[0043] The second stage is to realize automatic cage counting. First, when using the pyzbar library to identify the QR code information, the image taken in each chicken cage area is grayed out, and then the pyzbar.decode function is used to parse the QR code to extract the type of QR code, the chicken cage number and location information. Next, the smaller value is selected from the identified QR code information as the number of the chicken cage to be tested. Then, in the current frame Detect the center coordinates of all chickens and eggs , calculate the number of target observations whose center point coordinates are within the CRoI range The calculation process is as shown in formula (2) to determine the number of chickens and eggs in each CRoI. When the QR code information changes, the algorithm will re-count the observations of the video frame corresponding to the new chicken cage number. The target number is predicted by calculating the mean of these observations. , as shown in formula (3). Ensure that the target quantity changes between different video frames can be accurately tracked and recorded. Through these steps, the algorithm can realize the automatic cage counting of different chicken cages to identify low-yield chicken cages.

[0044] In the formula, is the frame number, For detection Total number of frames in the chicken cage; is the mean of all frame prediction values, Round up the decimal point.

[0045] In the process of automatic cage counting, there is a large error between the counting region of interest (CRoI) obtained by the inspection robot when it moves and the actual chicken cage area, resulting in errors in cage counting. In order to reduce this error, this study uses a monocular camera calibration model to calibrate the CRoI and realize the mutual conversion between the pixel coordinates and the world coordinates of the CRoI. The coordinate transformation process is (4) and (5): In the formula, is the coordinate of the center point of the image; is the boundary coordinate of the chicken coop; and are the horizontal distances from the camera to the QR code and the chicken coop respectively; for The transformed world coordinates. for Corrected coordinates; Divided into horizontal and vertical pixel focal length.

[0046] Optionally, the YOLOv8n-ML-DE model specifically refers to: Replace the Ciou loss function of the YOLOv8n model with Wiou(v2), add the MLCA attention mechanism to the lower layer of the spatial pooling pyramid SPPF of the backbone part, and replace the Detect of the Head with Efficient_Detect.

[0047] Based on the YOLOv8n model, the present invention proposes a YOLOv8n-ML-DE model with an improved structure. Figure 6 The schematic diagram of the YOLOv8n-ML-DE network framework structure provided by the present invention is as follows: Figure 6 As shown in the figure, first, the Ciou loss function is replaced by Wiou (v2) to improve the accuracy of bounding box regression, and the Wiou loss function (Weighted Intersection over Union) is used to optimize the positioning accuracy according to the importance, size or other relevant features of the bounding box. A dynamic weight distribution based on object visibility, size and difficulty is introduced, and the loss function is mathematically optimized, especially for small-sized targets that are difficult to identify (such as chicken heads) and eggs that are easily mixed with the background, which significantly improves the detection accuracy of the model and reduces missed detections.

[0048] Secondly, the MLCA attention mechanism is added to the lower layer of the spatial pooling pyramid SPPF in the backbone part to strengthen the extraction of key features and improve the model's detection performance for small targets such as chicken heads and eggs. Finally, the Head Detect is replaced with the lightweight Efficient_Detect to improve computational efficiency while maintaining target detection accuracy.

[0049] Figure 7 The schematic diagram of the process flow of the low-yield chicken cage identification method provided by the present invention is as follows: Figure 7 As shown, including: Step 710, the chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instruction or the preset inspection path; In the present invention, the chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instruction or the preset inspection path. In this step, the chassis equipment first receives the instruction through the controller, which may be the instruction sent by the operator through the remote control module, or the instruction automatically generated by the navigation module according to the preset path.

[0050] The controller then generates corresponding drive control instructions, and the drive module executes these instructions to control the chassis equipment to move along the planned path. During the movement, the visual system on the chassis equipment, including the camera, fill light device and edge calculator, is transported together to the location of the chicken cage to be inspected. After arriving, the visual system immediately starts working, the camera collects images of the chicken cage, the fill light device provides the necessary lighting, and the edge calculator is ready to process and analyze the collected images for subsequent counting of the number of chickens and eggs and identification of low-yield chicken cages.

[0051] Step 720: The camera in the visual system is used to collect the image of the chicken cage to be tested. The edge calculator in the visual system identifies the chickens, eggs and the QR code of the chicken cage in the image of the chicken cage through the YOLOv8n-ML-DE model, and determines the number of the chicken cage to be tested according to the QR code of the chicken cage. In the present invention, the visual system aims its camera at the chicken coops to be inspected and captures images of the chicken coops. Subsequently, the edge calculator in the visual system uses the YOLOv8n-ML-DE model to perform in-depth analysis on these images and identify the chickens, eggs, and chicken coop QR codes in the images. Based on the information in the QR code, the edge calculator can determine the specific number of each chicken coop to be inspected.

[0052] The invention not only realizes the automatic counting of the number of chickens and eggs, but also accurately identifies each chicken cage through QR code technology, providing accurate basic information for subsequent data management and analysis. This information is then uploaded to the cloud server for more in-depth production performance analysis to identify low-yield chicken cages.

[0053] Step 730, the edge calculator determines the counting region of interest corresponding to the chicken cage to be tested according to the coordinate information of the chicken cage QR code, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; In the present invention, the edge calculator first identifies the QR code from the image captured by the camera and extracts the coordinate information of the QR code. These coordinate information indicate the specific location of the QR code in the image, which in turn helps the edge calculator determine the CRoI of the chicken coop associated with the QR code. CRoI refers to a specific area in the image used to count chickens and eggs, and the determination of this area is crucial to ensure the accuracy of the counting.

[0054] After determining the CRoI, the edge calculator will analyze the image content in the area and calculate the number of chickens and eggs. This step usually involves image segmentation and object recognition algorithms that can distinguish and count the chickens and eggs in the image. By counting the chickens and eggs in the CRoI, the edge calculator can obtain the cage count results for each cage to be tested.

[0055] Finally, the edge computer organizes these cage counting results and generates a counting report containing the number of chickens and eggs for each cage to be tested. This report can be used to monitor the production status of the cage in real time and can also be used as basic data for subsequent analysis of the production performance of the cage.

[0056] Step 740: The wireless transmission device in the visual system uploads the serial number of the chicken cage to be tested and the cage count to the cloud server, so that the cloud server determines whether the chicken cage to be tested is a low-yield chicken cage according to the cage count of the chicken cage to be tested.

[0057] In the present invention, the wireless transmission device in the visual system is responsible for uploading the cage number and cage count data obtained by the edge calculator to the cloud server. This process is continuous and involves transmitting the cage number and count results to the cloud through the network. After receiving this data, the cloud server will determine whether each cage to be tested is a low-yield cage based on the cage count.

[0058] This is achieved by comparing the number of chickens and eggs in the chicken cage with the preset production efficiency standards. If the production data of a chicken cage is lower than these standards, the cloud server will mark it as a low-production chicken cage. Such automated data processing and analysis processes make the management of chicken farms more efficient and accurate, and help to take timely measures to improve production efficiency.

[0059] In this invention, the improved low-yield chicken cage identification algorithm is deeply integrated with the robot system, realizing the efficient linkage of automatic inspection, cage counting and low-yield chicken identification functions. Robot integration enables the entire system to have autonomous movement, accurate detection and remote management capabilities, greatly reducing the need for manual operation and improving the management automation level of the farm.

[0060] Figure 8 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 8 As shown, the electronic device may include: a processor 810, a communications interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communications interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the low-yield chicken cage identification method, the method comprising: the chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instruction or the preset inspection path; The camera in the visual system is used to collect images of the chicken cages to be inspected. The edge calculator in the visual system uses the YOLOv8n-ML-DE model to identify the chickens, eggs, and the QR code of the chicken cages to be inspected in the images of the chicken cages, and determines the number of the chicken cages to be inspected according to the QR code of the chicken cages. The edge calculator determines the counting region of interest corresponding to the chicken cage to be tested according to the coordinate information of the chicken cage QR code, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The wireless transmission device in the visual system uploads the serial number of the chicken cage to be tested and the cage count to the cloud server, so that the cloud server determines whether the chicken cage to be tested is a low-yield chicken cage according to the cage count of the chicken cage to be tested.

[0061] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0062] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the low-yield chicken cage identification method provided by the above methods, the method comprising: the chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instruction or the preset inspection path; The camera in the visual system is used to collect images of the chicken cages to be inspected. The edge calculator in the visual system uses the YOLOv8n-ML-DE model to identify the chickens, eggs, and the QR code of the chicken cages to be inspected in the images of the chicken cages, and determines the number of the chicken cages to be inspected according to the QR code of the chicken cages. The edge calculator determines the counting region of interest corresponding to the chicken cage to be tested according to the coordinate information of the chicken cage QR code, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The wireless transmission device in the visual system uploads the serial number of the chicken cage to be tested and the cage count to the cloud server, so that the cloud server determines whether the chicken cage to be tested is a low-yield chicken cage according to the cage count of the chicken cage to be tested.

[0063] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the low-yield chicken cage identification method provided by the above methods, the method comprising: the chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instruction or the preset inspection path; The camera in the visual system is used to collect images of the chicken cages to be inspected. The edge calculator in the visual system uses the YOLOv8n-ML-DE model to identify the chickens, eggs, and the QR code of the chicken cages to be inspected in the images of the chicken cages, and determines the number of the chicken cages to be inspected according to the QR code of the chicken cages. The edge calculator determines the counting region of interest corresponding to the chicken cage to be tested according to the coordinate information of the chicken cage QR code, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The wireless transmission device in the visual system uploads the serial number of the chicken cage to be tested and the cage count to the cloud server, so that the cloud server determines whether the chicken cage to be tested is a low-yield chicken cage according to the cage count of the chicken cage to be tested.

[0064] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0065] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A low-yield chicken cage identification system, characterized in that: include: Vision systems, chassis equipment and power equipment; The power supply device is electrically connected to the chassis device and the visual device respectively, and the visual system is arranged on the chassis device; Wherein, the chassis equipment includes: a controller, a navigation module, a remote control module and a driving module, and the controller is respectively connected to the navigation module, the remote control module, the driving module and the visual system for communication; The controller is used to generate a driving control instruction according to the information transmitted by the navigation module and the remote control module, and the driving module is used to control the movement of the chassis device according to the driving control instruction; The visual system includes: a fixed bracket, at least one camera, a fill light device and an edge calculator, wherein the fixed bracket is fixed above the chassis device, and the camera, the fill light device and the edge calculator are all fixed on the fixed bracket; the camera is used to collect the image of the chicken cage to be tested, and the edge calculator is used to automatically count the image of the chicken cage to obtain the number of chickens and eggs in the chicken cage to be tested, and transmit the number of chickens and eggs in the chicken cage to be tested to the cloud server, so that the cloud server can determine whether the chicken cage to be tested is a low-yield chicken cage based on the number of chickens and eggs in the chicken cage to be tested within a preset time period.

2. The low-yield chicken cage identification system according to claim 1, characterized in that: The navigation module includes: a geomagnetic sensor, a site sensor and an obstacle avoidance sensor; The geomagnetic sensor is arranged at the bottom of the chassis device, and is used to detect the preset magnetic track signal in real time when in working state, and the controller outputs the speed of the driving wheel after calculation and processing to keep it in the track state; The site sensor locates the equipment through the detected site position, so as to realize the turning and U-turn of the chassis equipment; Wherein, the obstacle avoidance sensor is used to implement emergency braking of the chassis equipment when an obstacle is detected.

3. The low-yield chicken cage identification system according to claim 1, characterized in that: The remote control module includes: a receiver and a controller; The receiver is used to receive control instructions and implement manual control of the chassis equipment according to the controller.

4. The low-yield chicken cage identification system according to claim 1, characterized in that: The driving module specifically includes: a servo motor, a driver, a reducer, a driving wheel and a driven wheel; the driving wheels are respectively arranged on the left and right sides of the chassis of the chassis equipment, and the driven wheels are arranged on the front and rear sides of the chassis of the chassis equipment. The driver is used to drive the communication connection with the controller, adjust the speed of the servo motor by proportional-integral-differential control, and increase the torque through the reducer to drive the driving wheel to rotate; The driven wheel utilizes a universal structure to assist the device in turning and adjusting its posture.

5. The low-yield chicken cage identification system according to claim 1, characterized in that: The visual system specifically includes: a first camera, a second camera, a third camera, a fill light device, an edge calculator and a data display screen; The setting position of the first camera corresponds to the height of the first layer of chicken cages; the setting position of the second camera corresponds to the height of the second layer of chicken cages; the setting position of the third camera corresponds to the height of the third layer of chicken cages; Wherein, the fill light is used to fill light for the shooting area of ​​each camera; Among them, the data display screen is used to display the chicken cage images collected by each camera.

6. The low-yield chicken cage identification system according to claim 1, characterized in that: The edge calculator is specifically used for: The YOLOv8n-ML-DE model is used to identify the chickens, eggs and the QR code of the chicken cage to be tested in the chicken cage image, and the number of the chicken cage to be tested is determined according to the QR code of the chicken cage; According to the coordinate information of the two-dimensional code of the chicken cage, the counting region of interest corresponding to the chicken cage to be tested is determined, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The serial number of the chicken cage to be tested and the cage count are uploaded to the cloud server together, so that the cloud server can identify whether the chicken cage to be tested is a low-production chicken cage based on the number of chickens and eggs in the chicken cage to be tested within a preset time period.

7. The low-yield chicken cage identification system according to claim 6, characterized in that: The YOLOv8n-ML-DE model specifically refers to: Replace the Ciou loss function of the YOLOv8n model with Wiou(v2), add the MLCA attention mechanism to the lower layer of the spatial pooling pyramid SPPF of the backbone part, and replace the Detect of the Head with Efficient_Detect.

8. A method for identifying low-yield chicken cages based on the low-yield chicken cage identification system according to any one of claims 1 to 7, characterized in that: include: The chassis equipment transports the loaded visual system to the chicken cage to be inspected according to the control instructions or the preset inspection path; The camera in the visual system is used to collect images of the chicken cages to be inspected. The edge calculator in the visual system uses the YOLOv8n-ML-DE model to identify the chickens, eggs, and the QR code of the chicken cages to be inspected in the images of the chicken cages, and determines the number of the chicken cages to be inspected according to the QR code of the chicken cages. The edge calculator determines the counting region of interest corresponding to the chicken cage to be tested according to the coordinate information of the chicken cage QR code, so as to obtain the cage count of the chicken cage to be tested according to the number of chickens and eggs in the counting region of interest; The wireless transmission device in the visual system uploads the serial number of the chicken cage to be tested and the cage count to the cloud server, so that the cloud server determines whether the chicken cage to be tested is a low-yield chicken cage according to the cage count of the chicken cage to be tested.

9. 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 computer program, the method for identifying low-yield chicken cages as claimed in claim 8 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying low-yield chicken cages as claimed in claim 8 is implemented.

Images