X-ray imaging-based goose body size and bone phenotype precise measurement method, system and device

Abstract

The application discloses a goose body size and bone phenotype precision measurement method, system and device based on X-ray imaging, and belongs to the technical field of image recognition. The method comprises the following steps: acquiring goose basic identification information and generating a detection ID, collecting weight and determining an effective weight, and then associating; based on the associated effective weight, estimating the estimated body length and thickness distribution of the measured goose by using a body shape prior model, and combining the single imaging effective field length to adaptively calculate the X-ray imaging parameters; conveying the measured goose into a lead room for preventing radiation, performing adaptive segmented imaging according to the calculated X-ray imaging parameters, and evaluating the image coverage rate in real time; based on the segmented images after evaluation, pre-processing, feature matching and splicing are performed on each segment image to generate a complete goose body X-ray panoramic image, and the goose body size is calculated based on the complete goose body X-ray panoramic image. The application significantly improves the efficiency, safety, data integrity and measurement accuracy of goose body size measurement.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, and in particular to a method, system and device for accurate measurement of goose body size and skeletal phenotype based on X-ray imaging. Background Technology

[0002] Currently, livestock and poultry breeding and selection have entered a new stage of deep integration of phenomics and genomics. High-throughput, precise phenotypic data have become the core support for analyzing the genetic mechanisms of complex economic traits and breeding superior breeds. In goose breeding, body size traits and skeletal development are core indicators for evaluating growth performance, slaughter performance, and animal health, directly determining the accuracy of breeding models and the efficiency of genetic improvement. At present, research on goose body size and skeletal phenotypes still faces two technical bottlenecks: First, the thick feathers on the goose's body make it difficult for traditional visible light vision to penetrate the feather layer, making it impossible to accurately identify key skeletal sites such as the keel and pelvis; manual caliper measurements have drawbacks such as low efficiency, high stress, and single indicator dimensions, seriously affecting the authenticity, repeatability, and high-throughput acquisition of body size data. Second, the evaluation of skeletal quality, such as bone density, still relies on destructive in vitro testing after slaughter, which cannot achieve live, non-invasive, and dynamic tracking, making it difficult to reveal the temporal patterns of skeletal development and its association with genetic variation.

[0003] To overcome these bottlenecks, researchers have recently attempted to introduce machine vision technology based on visible light or depth cameras into livestock and poultry body size measurement. However, due to issues such as feather occlusion, posture changes, and lighting interference in geese, problems such as incomplete contour extraction, unstable key point identification, and insufficient measurement accuracy easily arise. Furthermore, most existing systems use fixed fields of view and fixed exposure parameters, making it difficult to adapt to geese of different sizes. When the goose is large, a single image cannot completely cover the target area, and incomplete image stitching further affects the accuracy and stability of body size parameter extraction. In contrast, X-ray imaging technology, due to its strong penetrating power, can effectively penetrate feathers and soft tissue structures to directly obtain information on bones and key structures. It has significant advantages in imaging clarity, structural integrity, and robustness against occlusion, and has been maturely applied in fields such as medical diagnosis, industrial non-destructive testing, and foreign object detection in food. This provides a new technological possibility for the automated, high-throughput acquisition of live goose body size and non-destructive bone density testing, and is expected to fill the gap in live skeletal phenotypic data for goose breeding. However, the application of X-ray technology in poultry body size measurement is still in the initial exploratory stage, and a mature dedicated body size measurement system has not yet been formed. Existing X-ray inspection equipment is mostly portable or open-structured, lacking dedicated radiation shielding and safety protection design—it is not equipped with a complete lead room or multi-layer composite shielding structure, resulting in radiation leakage risks in actual use and making it difficult to meet the engineering safety requirements for long-term, continuous inspection scenarios.

[0004] In summary, the current field of automated goose body measurement still faces core challenges such as limited imaging methods, insufficient acquisition of structural information, lack of parameter control and fixation, and inadequate radiation safety protection. This makes it difficult to collect goose body size and phenotypic data in a high-throughput, accurate, and intelligent manner. Therefore, the development of a dedicated X-ray hardware system for live goose restraint and imaging is crucial. This system can effectively penetrate feather obstruction and accurately locate key skeletal points. By integrating a deep learning-based automatic skeletal key point recognition algorithm, it enables intelligent extraction of core body size parameters such as back length, sternum width, keel length, pelvic width, and tibia length. This fundamentally solves the pain points of traditional manual measurement, such as low efficiency, high stress, and large errors. This technology provides robust technological support for accelerating the breeding of high-quality meat goose varieties and large-scale propagation of superior breeds. Summary of the Invention

[0005] To address the aforementioned shortcomings in existing technologies, this application provides a method, system, and device for accurate measurement of goose body size and skeletal phenotype based on X-ray imaging. These solutions resolve issues such as difficulty in identifying key points due to feather obstruction in existing goose body size measurements, low efficiency and high stress associated with manual measurement, fixed X-ray imaging parameters that cannot be adapted to different body types, incomplete coverage of multi-segment imaging, and lack of dedicated radiation protection.

[0006] To achieve the aforementioned objectives, the technical solution adopted in this application is as follows: First aspect: This application provides a method for precise measurement of goose body size and skeletal phenotype based on X-ray imaging, including: S1: Obtain the basic identification information of the goose being tested, generate a unique detection task ID, collect the weight information of the goose being tested, obtain the effective weight after weight stability determination, and associate the effective weight with the unique detection task ID. S2: Based on the effective weight after association, the estimated body length and thickness distribution of the tested goose are estimated using a pre-trained offline body shape prior model. The body shape prior model includes a weight-body length regression model and a thickness distribution model of the tested goose in an inverted posture. S3: Based on the estimated body length and thickness distribution, combined with the effective field of view of a single imaging, adaptively calculate X-ray imaging parameters including the number of imaging segments, tube voltage and tube current; S4: The goose to be tested is transported to the detection area inside the radiation-proof lead room. Adaptive segmented imaging is performed based on the calculated X-ray imaging parameters, and the image coverage is evaluated in real time. S5: Based on the segmented images after evaluation, preprocess, feature match and stitch each segment to generate a complete panoramic X-ray image of the goose body, and calculate the goose body size based on the complete panoramic X-ray image of the goose body.

[0007] Furthermore, the basic identification information includes batch information, individual number, and auxiliary identification parameters for unique identification; The weight stability determination includes: acquiring the weight time series signal by continuous sampling, performing sliding window filtering, and taking the weight mean at the corresponding time as the effective weight when the weight change rate per unit time is lower than a set threshold or the signal variance converges to a preset range.

[0008] Furthermore, the expression for the weight-body length regression model is:

[0009] In the formula, This is an estimated body length. For effective weight, The weight-body length regression coefficient is... For the intercept term; in, and The following data was obtained by performing a least-squares fit on a historical sample dataset, with the fitting objective being to minimize the sum of squared residuals:

[0010] In the formula, For the first The measured length of each sample. For the first The actual measured weight corresponding to each sample The total number of samples; The expression for the thickness distribution model of the goose under the inverted hanging posture is:

[0011] In the formula, Let be the thickness distribution function of the goose being tested in an inverted position. The coordinates of the goose being measured are along its length when it is hanging upside down. , and The measurements are the leg thickness, chest thickness, and neck thickness of the goose in an inverted position. This represents the coordinate distance from the leg to the chest area. This represents the coordinate distance from the chest to the neck region.

[0012] Further, S3 includes: The number of imaging segments and step distance are determined based on the estimated body length and the effective field of view of a single X-ray imaging system. The calculation formula is as follows:

[0013] In the formula, For the number of shooting segments, It is a rounding function. The effective coverage length for a single imaging session. The effective field of view length for a single imaging session. The preset overlap ratio between adjacent fields of view. The step distance between adjacent imaging positions; The tube voltage and tube current corresponding to the imaging segment are determined based on the estimated body length and thickness distribution.

[0014] Furthermore, the step of determining the tube voltage and tube current corresponding to the imaging segment based on the estimated body length and thickness distribution includes: For imaging segments covering the chest region, increase tube voltage and tube current; For imaging segments covering the leg and neck regions, reduce tube voltage and tube current; The expression for determining the tube voltage is:

[0015] In the formula, For the first The tube voltage corresponding to each imaging segment For the first The maximum thickness corresponding to each imaging segment For tube voltage mapping function, These are the position coordinates along the length of the body. For the first The starting position coordinates of each imaging segment and This corresponds to the tube voltage boundary value. and These are the minimum and maximum values ​​of the goose body thickness in the training samples.

[0016] Further, S4 includes: S401: The X-ray source beam is controlled by tube voltage and tube current. The flat panel detector simultaneously acquires the first segment of the projected image, packages and stores the image data, the corresponding imaging segment position coordinates and the actual exposure parameters, and performs image coverage evaluation. S402: Control the stepper motor drive mechanism to move the goose body moving module by one step distance, so that the next imaging segment enters the field of view between the X-ray source and the flat panel detector; S403: Read the tube voltage and tube current corresponding to the imaging segment after the movement is in place, perform imaging acquisition and storage again, and evaluate the image coverage until the acquisition of all imaging segments is completed.

[0017] Further, the image coverage assessment includes: A1: Extract the starting and ending boundaries of the goose's outline in the image, and combine them with the position coordinates of the current imaging segment to calculate the cumulative length covered along the length direction of the goose's body, thus obtaining the image coverage rate. The calculation formula is as follows:

[0018] In the formula, For image coverage, This represents the cumulative coverage length. A2: When the image coverage is found to be below the preset threshold, the location and extent of the uncovered area are analyzed, and corresponding compensation strategies are adopted based on the analysis results. The corresponding compensation strategy based on the analysis results includes: if the uncovered area is located at both ends of the goose, it is determined that the end coverage is insufficient, and one or more compensation scanning segments are added in the corresponding direction, with the moving step distance set at half of the normal step distance; if the uncovered area is located in a certain section in the middle, it is determined that the middle is missing, the missing section is re-divided into one or more supplementary imaging segments, the stepper motor drive mechanism is adjusted to send the missing area into the field of view, and supplementary imaging is performed.

[0019] Further, S5 includes: S501: Perform preprocessing on each image segment, including grayscale normalization, noise filtering, and contrast enhancement. S502: Based on the preprocessed images, feature points between adjacent image segments are extracted using a hybrid algorithm including SIFT, SURF and ORB algorithms. The Euclidean distance or Hamming distance between the feature points is calculated for matching. The set is then stitched together based on the candidate matching points between the two adjacent image segments to obtain a complete panoramic X-ray image of the goose. S503: Based on the complete X-ray panoramic image of the goose, perform contour extraction, use an edge detection algorithm to identify the boundary between the goose and the background, and generate a closed goose contour curve; S504: Based on the closed outline curve of the goose body, the system automatically locates each key measurement point according to the anatomical features of the goose in the upside-down posture, and calculates auxiliary parameters of the goose body, including body length, chest depth, body width, leg length and neck length, based on the key points. S505: Summarize the goose's body outline curve, auxiliary parameters of key points, weight data, and panoramic X-ray images of the goose to obtain a complete record of the goose's body dimensions.

[0020] The second aspect: This application provides a system for precise measurement of goose body size and skeletal phenotype based on X-ray imaging, including: A radiation protection module is used to form a closed radiation-proof lead room and shield X-ray leakage. The radiation protection module includes a sheet metal outer shell, a steel-lead-steel composite protective layer, an electrical cabinet layer, a wiring layer, and a lead curtain. The RFID reader / writer module is used for non-contact reading of the electronic tags attached to the goose being tested to obtain basic identification information; An image acquisition module is used to emit X-rays and receive transmission projection images. The image acquisition module includes an X-ray source, a flat panel detector, and a synchronous Z-axis lifting module. The X-ray source and the flat panel detector are arranged opposite each other on both sides of the synchronous Z-axis lifting module, and are driven by the synchronous Z-axis lifting module to move up and down synchronously in the vertical direction. The motion execution module is used to transport the goose being tested and collect its weight. The motion execution module includes a goose movement module and a goose support. The goose movement module includes a conveyor line and a stepper motor, which is used to automatically send the goose being tested into the detection area inside the radiation-proof lead room and send it out after imaging is completed. The goose support includes a foot inversion device and a fixing device. The foot inversion device is equipped with a weight sensor, which is used to automatically read the weight data after the goose is in place and upload it to the user control module. The user control module, including the host PC computer software platform and the slave PLC electrical control system, is used to adaptively calculate imaging parameters based on the collected weight data and basic identification information, control the action execution module to complete the transportation and segmented imaging, perform image coverage assessment, and complete image stitching and body size parameter extraction.

[0021] Third aspect: This application provides a device for precise measurement of goose body size and skeletal phenotype based on X-ray imaging, comprising: External sheet metal, radiation-proof lead room, operating area, lead curtain, goose body support and weight sensor; The radiation-proof lead room contains a testing area, which includes an electrical cabinet, a flat panel detector, an X-ray source, a goose-moving module, a first synchronous Z-axis lifting module, and a second synchronous Z-axis lifting module. The electrical cabinet integrates electrical control components. The X-ray source and the flat panel detector are positioned opposite each other on either side of the first and second synchronous Z-axis lifting modules, and are driven synchronously in the vertical direction by these modules. The goose-moving module is used for transporting and positioning the goose within the testing area. An overflow port is located at the bottom of the testing area. The radiation-proof lead room adopts a steel-lead-steel composite structure; The operating area integrates a display screen, mouse, keyboard, emergency stop button, and start button; the lead curtain is installed at the material inlet and outlet of the radiation-proof lead room to dynamically shield radiation when materials enter and exit; the weight sensor is installed above the goose support to collect the weight data of the goose being measured in real time.

[0022] The beneficial effects of this application are: This application features full-process automation, significantly improving measurement efficiency and data repeatability. It also employs multiple radiation protection designs to ensure safety for engineering applications. Furthermore, real-time acquired weight data, combined with an offline-trained body shape prior model, automatically calculates imaging parameters, ensuring consistent and clear images and optimizing radiation dose. In addition, real-time image coverage assessment and dynamic compensation ensure the integrity of data acquisition. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0024] Figure 1 This is a flowchart illustrating a method for accurately measuring the body size and skeletal phenotype of a goose based on X-ray imaging, as provided in an embodiment of this application.

[0025] Figure 2 This is a schematic diagram of a system for accurately measuring the size and skeletal phenotype of a goose based on X-ray imaging, provided in an embodiment of this application.

[0026] Figure 3 This is a schematic diagram of the external structure of a goose body size and skeletal phenotype precision measurement device based on X-ray imaging, provided in an embodiment of this application.

[0027] Figure 4 This is a schematic diagram of the internal structure of a goose body size and skeletal phenotype precision measurement device based on X-ray imaging, provided in an embodiment of this application.

[0028] Figure 5 A schematic diagram of a radiation-protected lead room for a precise measurement device for goose body size and skeletal phenotype based on X-ray imaging, provided as an embodiment of this application.

[0029] Figure 6 A schematic diagram of the overflow port of a goose body size and skeletal phenotype precision measurement device based on X-ray imaging, provided in an embodiment of this application.

[0030] Among them: 1-external sheet metal; 2-radiation-proof lead room; 3-operating area; 4-lead curtain; 5-goose body support; 6-weight sensor; 7-electrical cabinet; 8-flat panel detector; 9-X-ray source; 10-goose body moving module; 11-first synchronous Z-axis lifting module; 12-second synchronous Z-axis lifting module; 13-overflow port. Detailed Implementation

[0031] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0032] Example 1: This application provides a method for precise measurement of goose body size and skeletal phenotype based on X-ray imaging. This method can be found in [reference needed]. Figure 1 , Figure 1 The diagram shown is a flowchart illustrating a method for accurately measuring the body size and skeletal phenotype of a goose based on X-ray imaging, according to an embodiment of this application. The method includes: S1: Obtain the basic identification information of the goose being tested, generate a unique detection task ID, collect the weight information of the goose being tested, obtain the effective weight after weight stability determination, and associate the effective weight with the unique detection task ID.

[0033] In one embodiment of this application, the system first acquires the basic identification information of the goose being tested, including batch information, individual number, and auxiliary identification parameters for unique identification. Based on the basic identification information, a unique detection task ID is generated for data association and traceability management throughout the entire process of weighing, transportation, imaging, and measurement. The basic identification information can be acquired in various ways, including manual input by operators through a user control interface; or non-contact reading of the electronic tag attached to the goose being tested using, for example, RFID technology read / write module. When a combination of manual input and automatic identification is used, the system prioritizes the automatic identification result and prompts the operator to confirm the identification result.

[0034] During system operation, the operator hangs the goose to be measured onto the goose support frame, and the weight information of the goose is collected by a weighing unit (weight sensor) set on the frame. The weighing unit acquires the weight time series signal in a continuous sampling manner and performs sliding window filtering and stability determination processing on the signal. When preset stability conditions are met, such as the weight change rate per unit time being lower than a set threshold and the signal variance converging to a preset range, the system uses the weight mean at the corresponding time as the valid weight input.

[0035] Optionally, the stability criterion includes: continuity N The difference between the maximum and minimum weight values ​​within a sampling period does not exceed the allowable deviation value, or the coefficient of variation within the sliding window is lower than the preset percentage threshold. If the stability condition cannot be met within the preset time window, the system will issue a prompt message, suggesting that the operator check the goose's posture or the working status of the weighing unit, and restart weight collection after confirming that the abnormality has been eliminated.

[0036] Furthermore, the effective weight input is associated with the generated unique detection task ID and data-bound with the subsequent imaging parameter calculation module, serving as the basic input for body shape prior modeling and adaptive parameter planning.

[0037] S2: Based on the effective weight after association, the estimated body length and thickness distribution of the tested goose are estimated using a pre-trained offline body shape prior model. The body shape prior model includes a weight-body length regression model and a thickness distribution model of the tested goose in an inverted posture.

[0038] In one embodiment of this application, the body shape prior model is pre-trained offline and stored in the storage unit of the user control system before the system goes online. During the actual measurement process, only forward inference calculation is performed, and the model parameters are not updated online to ensure the real-time performance and consistency of each measurement.

[0039] Body length estimation is performed using a weight-body length mapping function. Let the effective weight of the goose being measured be... W (kg), estimated body length L (mm) is represented by a linear regression model:

[0040] In the formula, This is an estimated body length. For effective weight, The weight-body length regression coefficient is... This is the intercept term.

[0041] in, and The data was obtained by least-squares fitting of a historical sample dataset containing several goose samples with known weights and measured body lengths. The fitting objective was to minimize the sum of squared residuals.

[0042] In the formula, For the first The measured length of each sample. For the first The actual measured weight corresponding to each sample The total number of samples.

[0043] To ensure the model has sufficient generalization ability and statistical significance, the training sample set should meet the following quantity requirements: at least [number of samples collected]. ≥200 valid samples, and the samples should cover the typical weight distribution range of the tested goose breed (e.g., from minimum mature weight to maximum mature weight). Training samples are derived from historical measurement data of previous batches or from specialized collection experiments. Each sample must simultaneously record the accurate weight value and the true value of body length measured manually or by instrument. Abnormal samples are removed to ensure the quality of training data.

[0044] Since this application scenario only requires establishing a rough mapping between weight and body shape, and does not require high-precision continuous distribution fitting, all the interpolation and smoothing parameters mentioned above are calibrated during the offline training phase and can be directly called upon after online operation without the need for online adjustment. Specifically, during the offline training process, the system determines parameters such as interpolation weights, spline basis function coefficients, and boundary conditions between typical parts through historical sample data, and the calibrated parameters are then fixed in the configuration file of the user control system.

[0045] Because the goose being measured was fixed in an upside-down position, its thickness distribution along its length (from head to bottom) exhibited a regular variation: the thickness was smaller in the leg and neck areas, and the thickness was greatest in the chest area due to its greatest chest depth. Based on this, let the coordinates along the body length be... x Its value range is 0≤ x ≤ L ;in x =0 corresponds to the end of the leg. x = L Corresponding to the top of the head. Then the thickness distribution function. The following can be estimated using a piecewise function:

[0046] In the formula, Let be the thickness distribution function of the goose being tested in an inverted position. The coordinates of the goose being measured are along its length when it is hanging upside down. The obtained estimated body length represents the top of the head. This represents the coordinate distance from the leg to the chest area. The coordinate distance from the chest to the neck region is used during modeling. and and Proportional, specifically the value is: =0.3 and =0.7 , , and The thicknesses of the legs, chest, and neck of the goose in an inverted position were measured, and the characteristic thicknesses of each region were compared with the body length. It satisfies the pre-defined regression relationship. In a typical implementation, chest thickness is taken as body length. 0.2 times the body length, with the thickness of the legs and neck respectively taken as the body length. The values ​​are 0.05 and 0.10 times higher. These parameters are adjustable to users through the software platform to support customized parameter adjustments for different goose populations.

[0047] Among them, the characteristic thickness and body length of each region L It satisfies the pre-defined regression relationship. Typically, the thickness of the chest is about 0.15 to 0.25 times the body length, and the thickness of the legs and neck is about 0.05 to 0.10 times the body length.

[0048] S3: Based on the estimated body length and thickness distribution, combined with the effective field of view of a single imaging, adaptively calculate X-ray imaging parameters including the number of imaging segments, tube voltage, and tube current.

[0049] In one embodiment of this application, the number of shooting segments is based on the estimated body length. L The ratio between the effective field of view of a single X-ray imaging system and the effective field of view of the system is adaptively determined. One implementation method is as follows: Let the effective field of view of a single imaging be... The preset overlap ratio between adjacent fields of view is Its value ranges from 0.15 to 0.35, preferably 0.25, then the effective coverage length of a single image is... The initial number of shooting segments is calculated using the following formula:

[0050] In the formula, This is the initial number of shooting segments. This is an up-rounding function that ensures complete coverage along the entire length of the goose.

[0051] Furthermore, the step distance between adjacent imaging positions Same as the effective coverage length, determined as:

[0052] This ensures that adjacent fields of view form continuous spatial coverage with redundant overlapping areas, which are used for feature matching and error correction in subsequent image stitching. Optionally, the system also supports a manual intervention mode. When the operator manually specifies the number of shooting segments, the system prioritizes this manual value, but it needs to verify whether the corresponding total coverage length meets the coverage requirements.

[0053] The tube voltage and tube current corresponding to each imaging segment are determined based on the estimated volume length.L And a priori thickness distribution model for the inverted posture. For the imaging segment covering the chest region, since this region has the greatest thickness and the most severe X-ray attenuation, a higher tube voltage and tube current are required to ensure sufficient penetration and image signal-to-noise ratio; for the imaging segment covering the leg and neck regions, since the thickness is smaller, the tube voltage and tube current can be reduced accordingly to reduce unnecessary radiation dose and avoid image overexposure.

[0054] Let the first The maximum thickness corresponding to the imaging segment is Then the first segment voltage Determined according to the following mapping function:

[0055] In the formula, For the first The tube voltage corresponding to each imaging segment For the first The maximum thickness corresponding to each imaging segment This is a tube voltage mapping function used to map the maximum thickness within a certain imaging segment to the corresponding tube voltage value. The coordinates are along the length of the body, ranging from 0 to... L , For the first The starting position coordinates of each imaging segment and These are the minimum and maximum values ​​of the goose body thickness in the training samples. and These are the corresponding tube voltage boundary values, and both were calibrated through offline experiments.

[0056] The tube current can be determined using a similar mapping relationship, or by keeping the tube current constant while adjusting only the tube voltage. The specific method depends on the actual performance of the imaging system.

[0057] After calculating the number of imaging segments, tube voltage, and tube current, the system encapsulates the imaging parameter set and sends it to the image acquisition module. If an adaptive correction mechanism is triggered during subsequent imaging, the initial parameter set will be dynamically updated.

[0058] In existing technologies, X-ray imaging parameters are mostly set using fixed values ​​or manual experience, which is difficult to adapt to the significant differences in thickness distribution of geese of different body sizes. This easily leads to underexposure in thick areas such as the chest (underexposed images, loss of detail) or overexposure in thin areas such as the legs and neck (image saturation, blurred boundaries). This application uses real-time acquired goose weight data, combined with an offline-trained body size prior model (weight-body length linear regression model and thickness distribution model under inverted posture), to adaptively calculate the number of imaging segments, tube voltage, and tube current. In particular, for the thickness distribution pattern of thick chest and thin legs and neck under inverted posture, a segmented differentiated exposure strategy is adopted to match the exposure parameters of each imaging segment with the actual thickness of the corresponding area, thereby ensuring consistent image clarity across the entire goose body and significantly improving the accuracy of subsequent body size measurements.

[0059] Furthermore, using uniform exposure parameters often sets tube voltage and current based on areas of large thickness, resulting in thinner areas receiving unnecessary excessive radiation. This application employs a segmented differentiated exposure strategy, using higher exposure parameters for thicker areas such as the chest to ensure penetration, while correspondingly lowering exposure parameters for thinner areas such as the legs and neck. This effectively reduces the cumulative radiation dose received by the goose while ensuring image quality, and also extends the lifespan of the X-ray source.

[0060] S4: The goose to be tested is transported to the detection area inside the radiation-proof lead room. Adaptive segmented imaging is performed based on the calculated X-ray imaging parameters, and the image coverage is evaluated in real time.

[0061] In one embodiment of this application, after the imaging parameters are calculated, the action execution module initiates the transport process. Specifically, driven by a stepper motor, the goose-moving module smoothly transports the support carrying the goose to be tested through a lead curtain into the detection area inside the radiation-proof lead room. During transport, the system monitors the goose's position in real time using photoelectric sensors or limit switches installed along the transport path. When the goose reaches the preset imaging start position, the system issues a pause signal, the stepper motor stops operating and maintains the locked position, and simultaneously sends confirmation information to the user control system. If an abnormality is detected during transport, such as excessive transport resistance or a position sensor failing to respond within a timeout period, the system automatically stops transport and issues an alarm. The process resumes after the operator troubleshoots the problem. The lead curtain is passively opened when the goose enters and automatically returns to its original position after the goose has completely passed through, minimizing radiation leakage.

[0062] Once the goose is in position and locked, the system performs adaptive segmented imaging based on the calculated set of imaging parameters. The set of imaging parameters includes the number of imaging segments, the tube voltage and tube current corresponding to each segment, and the step distance between adjacent imaging positions.

[0063] Before imaging begins, the system first adjusts the flat panel detector and X-ray source to a height aligned with the target area of ​​the goose body using a synchronous Z-axis lifting module, based on the preset initial imaging position coordinates, to ensure that the imaging center falls at a suitable position along the thickness direction of the goose body. Subsequently, the system follows the set initial exposure parameters. U 1 (tube voltage corresponding to the first imaging segment) I 1 (the tube current corresponding to the first imaging segment) controls the X-ray source to emit a beam, and the flat panel detector synchronously acquires the first segment of projected image, and packages and stores the image data, the position coordinates of the segment and the actual exposure parameters.

[0064] After completing the first imaging segment, the system controls a stepper motor drive mechanism to move the goose-shaped moving module one step distance, bringing the next imaging segment into the field of view between the X-ray source and the flat panel detector. Once in position, the system reads the corresponding tube voltage and current for that segment and performs imaging acquisition again. This "movement-imaging-storage" cycle is repeated until all image segments are acquired.

[0065] During and after segmented imaging, the system simultaneously performs X-ray image coverage assessment of the goose body to ensure that each acquired image segment can completely cover the entire length of the goose being measured, thus avoiding subsequent measurement errors due to missing coverage.

[0066] Specifically, after each imaging segment is completed, the system analyzes the effective imaging area of ​​that segment in real time, extracts the start and end boundaries of the goose's outline in the image, and calculates the cumulative length covered along the length of the goose's body by combining the position coordinates of the current imaging segment. The system then calculates the cumulative coverage length. Compared with estimated body length L Compare and calculate image coverage :

[0067] Furthermore, when the coverage rate is detected to be below a preset threshold (e.g., 40%), the system further analyzes the location and extent of the uncovered area. If the uncovered area is located at either end of the goose's body (e.g., the end of the leg or the top of the head), it is determined to be insufficient end coverage; if the uncovered area is located in the middle section (e.g., due to the goose's bent posture causing a local deviation from the field of view), it is determined to be a mid-section missing area. For different types of insufficient coverage, the system adopts corresponding compensation strategies: for insufficient end coverage, the system adds one or more compensation scanning segments in the corresponding direction, setting the movement step distance to half the normal step distance to ensure sufficient overlap between the compensation segment and the existing image; for mid-section missing areas, the system re-divides the missing section into one or more re-imaging segments, adjusts the stepper motor drive mechanism to bring the missing area into the field of view, and performs re-imaging. After the compensation imaging is completed, the system recalculates the coverage rate until the preset threshold requirement is met.

[0068] In existing technologies, segmented imaging typically uses a fixed number of segments and a fixed step distance. When the goose's posture shifts or its size exceeds the normal range, insufficient coverage at the ends or missing areas in the middle region can easily occur, leading to subsequent image stitching failures or incomplete body size measurement data. This application performs a simultaneous X-ray image coverage assessment of the goose during and after segmented imaging, calculating the ratio of cumulative coverage length to estimated body length in real time. When insufficient coverage is detected, compensation scan segments are automatically added until a preset coverage threshold is met. This dynamic compensation mechanism effectively avoids data loss due to posture shifts or abnormal body size, ensuring that a complete whole-goose X-ray image is obtained for each measurement.

[0069] The system also evaluates whether the actual overlap ratio between adjacent image segments meets the preset requirements. Let the first... j Section and the j The theoretical overlap ratio of +1 segment is Actual overlap ratio The degree of overlap in the goose's body contour features between the two images is determined by calculating this. If... (in If the minimum allowable overlap ratio is 0.20, it indicates insufficient overlap, which may lead to gaps or registration failures during subsequent stitching; if (in If the maximum allowed overlap ratio is 0.60, it indicates that the overlap redundancy is too high. The system records this information to optimize the step distance setting for subsequent measurements.

[0070] S5: Based on the segmented images after evaluation, preprocess, feature match and stitch each segment to generate a complete panoramic X-ray image of the goose body, and calculate the goose body size based on the complete panoramic X-ray image of the goose body.

[0071] In one embodiment of this application, after all imaging segments are acquired and coverage is evaluated, the system performs segmented image feature matching and stitching processing to fuse multiple X-ray projection images into a complete panoramic image of the goose.

[0072] First, the image processing module preprocesses each image segment, including grayscale normalization, noise filtering, and contrast enhancement. Grayscale normalization eliminates the overall brightness inconsistency caused by differences in exposure parameters between different imaging segments by mapping the grayscale histograms of each segment to a uniform reference distribution. Noise filtering uses Gaussian filtering or median filtering to remove random noise. Contrast enhancement uses histogram equalization or adaptive Gamma correction to highlight the goose's outline and internal structural features.

[0073] Furthermore, after preprocessing, the system extracts and matches feature points between adjacent image segments. Assume that adjacent image segments have spatially overlapping regions. The system extracts feature points within these overlapping regions using a hybrid feature vector derived from SIFT (Scale Invariant Feature Transform), SURF (Accelerated Robust Feature Transform), and ORB (Binary Descriptor for Fast Rotation) algorithms. After feature point extraction, the system performs matching by calculating the Euclidean or Hamming distance between feature point descriptors, stitching together the candidate matching point pairs between adjacent image segments. Upon completion of the stitching, the system generates a complete X-ray panoramic image covering the entire goose, saves it to a storage unit, and associates it with a unique detection task ID.

[0074] After completing the segmented image stitching and obtaining an X-ray panoramic image covering the entire goose, the system enters the body size measurement stage, automatically extracting various body size parameters of the goose being measured based on the panoramic image.

[0075] Specifically, the system first performs contour extraction on the panoramic image, using edge detection algorithms (including the Canny and Sobel operators) to identify the boundary between the goose and the background, generating a closed contour curve of the goose. Based on this, the system automatically locates key measurement points according to the anatomical features of the goose in its inverted posture, including the apex of the head, the thinnest point of the neck, the widest point of the chest, and the end points of the legs. The localization of these key points can be achieved based on a preset geometric feature model, and all localization parameters are trained and calibrated offline.

[0076] After key point localization is completed, the system calculates the following body size parameters: body length, which is the vertical projection distance from the top of the head to the end of the leg; chest depth, which is the maximum vertical distance from the back to the sternum along the horizontal direction; body width, which is the maximum lateral width of the chest area; and auxiliary parameters such as leg length and neck length. The actual physical dimensions of each parameter are converted using calibration coefficients, which are obtained in advance through imaging calibration of standard phantoms of known dimensions and are expressed as pixel equivalents (mm / pixel).

[0077] Finally, the system summarizes all body size parameters, weight data, and panoramic images to generate a complete measurement record, and then enters the information entry and data management stage, automatically completing the entire closed loop from weight acquisition, parameter calculation, segmented imaging to body size measurement.

[0078] Example 2: This application provides a system for precise measurement of goose body size and skeletal phenotype based on X-ray imaging. This system can be found in [reference needed]. Figure 2 It includes: RFID reading and writing module, radiation protection module, image acquisition module, action execution module and user control module. The modules work together to form a complete detection process.

[0079] The RFID reader / writer module is used for non-contact reading of the electronic tags attached to the goose being tested to obtain basic identification information.

[0080] The radiation protection module is used to form a sealed radiation-proof lead room and shield against X-ray leakage. This module includes a sheet metal outer shell, a steel-lead-steel composite protective layer, an electrical cabinet layer, a wiring layer, and a lead curtain. Specifically, the radiation-proof lead room adopts a steel-lead-steel composite structure. The steel plates on both sides support and fix the intermediate lead plate. The lead layer serves as the primary radiation shielding material, and its thickness is determined comprehensively based on the tube voltage and current of the X-ray source and the available installation space.

[0081] In this embodiment, the thicknesses of the steel-lead-steel composite structure are 2mm, 3mm, and 2mm, respectively.

[0082] The image acquisition module, used to emit X-rays and receive transmission projection images, includes an X-ray source, a flat panel detector, and a synchronous Z-axis lifting module. The X-ray source and the flat panel detector are positioned opposite each other on both sides of the synchronous Z-axis lifting module. The synchronous Z-axis lifting module drives the two to move up and down synchronously in the vertical direction to ensure that the imaging center is always aligned, thereby obtaining a clear projection image.

[0083] The motion execution module is used to transport the goose to be tested. The motion execution module includes a goose movement module and a goose support. The goose movement module includes a conveyor line and a stepper motor, which is used to automatically send the goose to be tested into the detection area inside the radiation-proof lead room and send it out after imaging is completed. The goose support is used to keep the goose's posture stable at the imaging position. It includes a foot inversion device and a fixing device. The foot inversion device is equipped with a weight sensor, which is used to automatically read the weight data after the goose is in place and upload it to the user control module. This weight data is used as the input condition for subsequent adaptive setting of shooting parameters.

[0084] The user control module includes a host PC computer software platform and a slave PLC electrical control system. It enables human-machine interaction through a monitor, mouse, keyboard, emergency stop button, and start button located in the operation area. It is used to adaptively calculate imaging parameters based on the collected weight data and basic identification information, control the action execution module to complete the transportation and segmented imaging, perform image coverage assessment, and complete image stitching and body size parameter extraction.

[0085] In existing technologies, goose body size measurement often employs manual measurement or semi-automatic imaging methods, which suffer from cumbersome operation procedures, large human errors, and low measurement efficiency. This application's embodiment addresses these issues by centrally managing the entire process—from weight acquisition after goose mounting, automatic calculation of imaging parameters, transport and positioning, segmented imaging, image stitching to body size measurement, and data entry—all under the unified control of a user control module (upper-level PC software platform and lower-level PLC electrical control system). All subsystems operate automatically and collaboratively. Operators only need to complete two steps: individual information entry and goose mounting; the remaining steps require no manual intervention, significantly simplifying the operation, lowering the operational threshold and reducing human error, and improving measurement efficiency.

[0086] Furthermore, existing radiation shielding lead rooms often employ a single-thickness lead plate structure, resulting in issues such as heavy weight, high cost, and separation of structural support and radiation shielding functions. This application adopts a steel-lead-steel composite multi-layer shielding structure. The steel plates on both sides simultaneously provide support and auxiliary shielding, while the middle lead layer serves as the primary radiation shielding material, with its thickness determined based on pipe voltage, pipe current, and installation space. Simultaneously, a lead curtain is placed at the device opening to achieve dynamic shielding, and the outer sheet metal layer, electrical cabinet layer, and wiring layer provide electrical safety isolation. This design effectively reduces the overall weight and manufacturing costs while ensuring effective radiation protection.

[0087] Example 3: This application provides a device for precise measurement of goose body size and skeletal phenotype based on X-ray imaging. This device can be found in [reference needed]. Figures 3-4 The device's exterior includes an external sheet metal frame 1, a radiation-proof lead room 2, an operating area 3, a lead curtain 4, and a goose-shaped support frame 5.

[0088] The operation area 3 integrates a display screen, mouse, keyboard, emergency stop button, and start button, facilitating parameter setting and process control for operators. A lead curtain 4 is installed at the device opening to dynamically shield radiation during material entry and exit. A weight sensor 6 is installed above the goose support 5 to collect real-time weight data of the goose being measured. Furthermore, the overall support of the device utilizes an aluminum alloy frame, assembled from aluminum blocks of the same specifications but different lengths connected by threads. Compared to traditional welded steel frames, this structure offers higher overall strength under the same cross-sectional conditions and boasts advantages such as light weight, ease of processing, low cost, and excellent appearance, maintaining structural stability during long-term use.

[0089] like Figure 4As shown, the interior of the device is a detection area consisting of a radiation-proof lead room 2, which includes an electrical cabinet 7, a flat panel detector 8, an X-ray source 9, a goose-body moving module 10, and a first synchronous Z-axis lifting module 11 and a second synchronous Z-axis lifting module 12 located on the flat panel detector and the X-ray source. The electrical cabinet 7 is used to integrate electrical control components; the flat panel detector 8 and the X-ray source 9 are arranged opposite each other, and are driven by the first synchronous Z-axis lifting module 11 and the second synchronous Z-axis lifting module 12 to move them synchronously up and down in the vertical direction to accommodate geese of different sizes; the goose-body moving module is used to realize the transportation and positioning of the goose within the detection area.

[0090] like Figure 5 As shown, the radiation-shielding lead room adopts a steel-lead-steel composite structure. Steel plates on both sides support and fix the middle lead plate. The lead layer serves as the main radiation shielding material, and its thickness is determined comprehensively based on the tube voltage, tube current, and equipment installation space of the X-ray source 9. In this embodiment, the thicknesses of the steel-lead-steel composite structure are 2 mm, 3 mm, and 2 mm, respectively. Furthermore, considering that geese may produce excrement due to stress during the testing process, such as... Figure 6 As shown, an overflow port 13 is provided at the bottom of the detection area to drain urine and other liquids in a timely manner, which facilitates cleaning and maintenance and avoids the impact of liquid accumulation on the equipment and imaging quality.

[0091] This application features full-process automation, significantly improving measurement efficiency and data repeatability. It also employs multiple radiation protection designs to ensure safety for engineering applications. Furthermore, real-time weight data, combined with an offline-trained body shape prior model, automatically calculates imaging parameters, ensuring consistent and clear images and optimizing radiation dose. In addition, real-time image coverage assessment and dynamic compensation ensure the integrity of data acquisition.

[0092] It should be noted that those skilled in the art will recognize that the embodiments described herein are for the purpose of helping readers understand the principles of this application, and should be understood as not limiting the scope of protection of this application to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this application without departing from the essence of this application, and these modifications and combinations are still within the scope of protection of this application.

Claims

1. A method for precise determination of goose body size and skeletal phenotype based on X-ray imaging, characterized in that, include: S1: Obtain the basic identification information of the goose being tested, generate a unique detection task ID, collect the weight information of the goose being tested, obtain the effective weight after weight stability determination, and associate the effective weight with the unique detection task ID. S2: Based on the effective weight after association, the estimated body length and thickness distribution of the tested goose are estimated using a pre-trained offline body shape prior model. The body shape prior model includes a weight-body length regression model and a thickness distribution model of the tested goose in an inverted posture. S3: Based on the estimated body length and thickness distribution, combined with the effective field of view of a single imaging, adaptively calculate X-ray imaging parameters including the number of imaging segments, tube voltage and tube current; S4: The goose to be tested is transported to the detection area inside the radiation-proof lead room. Adaptive segmented imaging is performed based on the calculated X-ray imaging parameters, and the image coverage is evaluated in real time. S5: Based on the segmented images after evaluation, preprocess, feature match and stitch each segment to generate a complete panoramic X-ray image of the goose body, and calculate the goose body size based on the complete panoramic X-ray image of the goose body.

2. The method for precise determination of goose body size and skeletal phenotype based on X-ray imaging according to claim 1, characterized in that, The basic identification information includes batch information, individual number, and auxiliary identification parameters used for unique identification; The weight stability determination includes: acquiring the weight time series signal by continuous sampling, performing sliding window filtering, and taking the weight mean at the corresponding time as the effective weight when the weight change rate per unit time is lower than a set threshold or the signal variance converges to a preset range.

3. The method for precise measurement of goose body size and skeletal phenotype based on X-ray imaging according to claim 1, characterized in that, The expression for the weight-body length regression model is: In the formula, This is an estimated body length. For effective weight, The weight-body length regression coefficient is... For the intercept term; in, and The following data was obtained by performing a least-squares fit on a historical sample dataset, with the fitting objective being to minimize the sum of squared residuals: In the formula, For the first The measured length of each sample. For the first The actual measured weight corresponding to each sample The total number of samples; The expression for the thickness distribution model of the goose under the inverted hanging posture is: In the formula, Let be the thickness distribution function of the goose being tested in an inverted position. The coordinates of the goose being measured are along its length when it is hanging upside down. , and The measurements are the leg thickness, chest thickness, and neck thickness of the goose in an inverted position. This represents the coordinate distance from the leg to the chest area. This represents the coordinate distance from the chest to the neck region.

4. The method for precise determination of goose body size and skeletal phenotype based on X-ray imaging according to claim 3, characterized in that, The S3 includes: The number of imaging segments and step distance are determined based on the estimated body length and the effective field of view of a single X-ray imaging system. The calculation formula is as follows: In the formula, For the number of shooting segments, It is a rounding function. The effective coverage length for a single imaging session. The effective field of view length for a single imaging session. The preset overlap ratio between adjacent fields of view. The step distance between adjacent imaging positions; The tube voltage and tube current corresponding to the imaging segment are determined based on the estimated body length and thickness distribution.

5. The method for precise determination of goose body size and skeletal phenotype based on X-ray imaging according to claim 4, characterized in that, The determination of tube voltage and tube current corresponding to the imaging segment based on the estimated body length and thickness distribution includes: For imaging segments covering the chest region, increase tube voltage and tube current; For imaging segments covering the leg and neck regions, reduce tube voltage and tube current; The expression for determining the tube voltage is: In the formula, For the first The tube voltage corresponding to each imaging segment For the first The maximum thickness corresponding to each imaging segment For tube voltage mapping function, These are the position coordinates along the length of the body. For the first The starting position coordinates of each imaging segment and This corresponds to the tube voltage boundary value. and These are the minimum and maximum values ​​of the goose body thickness in the training samples.

6. The method for precise measurement of goose body size and skeletal phenotype based on X-ray imaging according to claim 5, characterized in that, S4 includes: S401: The X-ray source beam is controlled by tube voltage and tube current. The flat panel detector simultaneously acquires the first segment of the projected image, packages and stores the image data, the corresponding imaging segment position coordinates and the actual exposure parameters, and performs image coverage evaluation. S402: Control the stepper motor drive mechanism to move the goose body moving module by one step distance, so that the next imaging segment enters the field of view between the X-ray source and the flat panel detector; S403: Read the tube voltage and tube current corresponding to the imaging segment after the movement is in place, perform imaging acquisition and storage again, and evaluate the image coverage until the acquisition of all imaging segments is completed.

7. The method for precise determination of goose body size and skeletal phenotype based on X-ray imaging according to claim 6, characterized in that, The image coverage assessment includes: A1: Extract the starting and ending boundaries of the goose's outline in the image, and combine them with the position coordinates of the current imaging segment to calculate the cumulative length covered along the length direction of the goose's body, thus obtaining the image coverage rate. The calculation formula is as follows: In the formula, For image coverage, This represents the cumulative coverage length. A2: When the image coverage is found to be below the preset threshold, the location and extent of the uncovered area are analyzed, and corresponding compensation strategies are adopted based on the analysis results. The corresponding compensation strategy based on the analysis results includes: if the uncovered area is located at both ends of the goose, it is determined that the end coverage is insufficient, and one or more compensation scanning segments are added in the corresponding direction, with the moving step distance set at half of the normal step distance; if the uncovered area is located in a certain section in the middle, it is determined that the middle is missing, the missing section is re-divided into one or more supplementary imaging segments, the stepper motor drive mechanism is adjusted to send the missing area into the field of view, and supplementary imaging is performed.

8. The method for precise determination of goose body size and skeletal phenotype based on X-ray imaging according to claim 1, characterized in that, The S5 includes: S501: Perform preprocessing on each image segment, including grayscale normalization, noise filtering, and contrast enhancement. S502: Based on the preprocessed images, feature points between adjacent image segments are extracted using a hybrid algorithm including SIFT, SURF and ORB algorithms. The Euclidean distance or Hamming distance between the feature points is calculated for matching. The set is then stitched together based on the candidate matching points between the two adjacent image segments to obtain a complete panoramic X-ray image of the goose. S503: Based on the complete X-ray panoramic image of the goose, perform contour extraction, use an edge detection algorithm to identify the boundary between the goose and the background, and generate a closed goose contour curve; S504: Based on the closed outline curve of the goose body, the system automatically locates each key measurement point according to the anatomical features of the goose in the upside-down posture, and calculates auxiliary parameters of the goose body, including body length, chest depth, body width, leg length and neck length, based on the key points. S505: Summarize the goose's body outline curve, auxiliary parameters of key points, weight data, and panoramic X-ray images of the goose to obtain a complete record of the goose's body dimensions.

9. A system for accurately measuring the body size and skeletal phenotype of a goose based on X-ray imaging as described in any one of claims 1-8, characterized in that, include: A radiation protection module is used to form a closed radiation-proof lead room and shield X-ray leakage. The radiation protection module includes a sheet metal outer shell, a steel-lead-steel composite protective layer, an electrical cabinet layer, a wiring layer, and a lead curtain. The RFID reader / writer module is used for non-contact reading of the electronic tags attached to the goose being tested to obtain basic identification information; An image acquisition module is used to emit X-rays and receive transmission projection images. The image acquisition module includes an X-ray source, a flat panel detector, and a synchronous Z-axis lifting module. The X-ray source and the flat panel detector are arranged opposite each other on both sides of the synchronous Z-axis lifting module, and are driven by the synchronous Z-axis lifting module to move up and down synchronously in the vertical direction. The motion execution module is used to transport the goose being tested and collect its weight. The motion execution module includes a goose movement module and a goose support. The goose movement module includes a conveyor line and a stepper motor, which is used to automatically send the goose being tested into the detection area inside the radiation-proof lead room and send it out after imaging is completed. The goose support includes a foot inversion device and a fixing device. The foot inversion device is equipped with a weight sensor, which is used to automatically read the weight data after the goose is in place and upload it to the user control module. The user control module, including the host PC computer software platform and the slave PLC electrical control system, is used to adaptively calculate imaging parameters based on the collected weight data and basic identification information, control the action execution module to complete the transportation and segmented imaging, perform image coverage assessment, and complete image stitching and body size parameter extraction.

10. An apparatus for accurately measuring the size and skeletal phenotype of a goose based on X-ray imaging as described in any one of claims 1-8, characterized in that, include: External sheet metal (1), radiation-proof lead room (2), operating area (3), lead curtain (4), goose body support (5), and weight sensor (6); The radiation-proof lead room (2) contains a detection area, which includes an electrical cabinet (7), a flat panel detector (8), an X-ray source (9), a goose movement module (10), a first synchronous Z-axis lifting module (11), and a second synchronous Z-axis lifting module (12). The electrical cabinet (7) is used to integrate electrical control components. The X-ray source (9) and the flat panel detector (8) are positioned opposite each other on both sides of the first synchronous Z-axis lifting module (11) and the second synchronous Z-axis lifting module (12), and are driven synchronously in the vertical direction by the first synchronous Z-axis lifting module (11) and the second synchronous Z-axis lifting module (12). The goose movement module (10) is used to realize the transportation and positioning of the goose in the detection area. An overflow port (13) is provided at the bottom of the detection area. The radiation-proof lead room (2) adopts a steel-lead-steel composite structure; The operation area (3) integrates a display screen, mouse, keyboard, emergency stop button and start button; the lead curtain (4) is set at the material inlet and outlet of the radiation-proof lead room (2) to dynamically shield radiation when materials enter and exit; the weight sensor (6) is set above the goose support (5) to collect the weight data of the goose being measured in real time.

Images