Method and system for disinfecting, detecting and grading waterfowl hatching eggs

Abstract

The invention relates to the technical field of waterfowl breeding, and discloses a waterfowl hatching egg disinfection detection grading method and system. The method comprises the steps of hatching egg pretreatment, gradient collaborative disinfection, multi-dimensional nondestructive testing and intelligent grading and sorting which are executed in sequence. The gradient collaborative disinfection adopts an alternate collaborative process of primary low-temperature plasma treatment, ozone permeation treatment and secondary low-temperature plasma treatment; machine vision, near infrared spectroscopic analysis and mechanical property detection are synchronously carried out in multi-dimensional nondestructive detection, and time-space registration and fusion of multi-source detection features are completed; a grading evaluation model is constructed, and grading and automatic sorting of hatching egg hatching priorities are achieved. The problems that existing hatching eggs are not thoroughly disinfected and sterilized, embryos are prone to being damaged, the detection dimension is single, and grading is extensive are solved, the sterilization effect and embryo safety are considered, the hatching egg detection precision and grading accuracy are improved, the hatching egg hatching rate and the healthy young rate can be remarkably increased, and the production requirement for large-scale waterfowl breeding is met.

Description

Technical Field

[0001] This invention relates to the field of waterfowl farming technology, and in particular to a pretreatment technology for waterfowl hatching eggs before incubation, specifically a method and system for disinfection, detection and grading of waterfowl hatching eggs. Background Technology

[0002] As a core upstream link in the entire waterfowl farming industry chain, the quality of pre-incubation treatment of hatching eggs directly determines the economic benefits of the entire farming cycle. Among them, the disinfection, sterilization, quality testing, and grading of hatching eggs before incubation are not only the core factors determining the hatching rate, chick survival rate, and chick uniformity, but also key prevention and control points for vertically transmitted diseases such as avian leukosis and salmonella, directly related to the disease prevention and control safety and cost control capabilities of large-scale farming.

[0003] Currently, the pre-incubation processing procedures in large-scale waterfowl hatcheries still generally suffer from two major pain points: First, the disinfection process cannot simultaneously ensure thorough sterilization and embryo safety. Although traditional formaldehyde fumigation disinfection has a wide sterilization range, it is very easy to damage the embryos of hatching eggs and leave harmful residues on the eggshell surface. At the same time, the occupational exposure risks for operators and the environmental compliance pressure of exhaust gas emissions are becoming increasingly prominent. Disinfectant spraying or immersion disinfection can easily cause cross-contamination between hatching eggs during batch processing, and disinfectants can easily penetrate into the interior through the micropores of the eggshell, directly affecting the normal development of the embryo.

[0004] Secondly, in the testing and grading stages, small and medium-sized farms still rely on manual visual observation and spot checks with candling lights, which is inefficient, prone to errors, and unable to detect core indicators such as eggshell thickness, embryo viability, and mechanical strength. Existing automated testing methods are mostly single-dimensional, only capable of detecting single indicators of appearance or internal structure, lacking multi-source data fusion mechanisms and unable to comprehensively evaluate the hatching potential of fertilized eggs. Furthermore, the grading system is crude, making it difficult to achieve refined priority classification for hatching, easily leading to a waste of hatching resources. Existing automated testing technologies include numerous single-dimensional testing schemes, while some patented technologies related to multi-indicator testing exist.

[0005] Publication number CN116636481A discloses a method and equipment for intelligent determination of poultry egg production performance and egg quality. This solution uses machine vision to detect appearance defects in hatching eggs and combines weight detection to complete the grading of hatching eggs. However, this solution has significant technical limitations: 1. It can only obtain the appearance characteristics and weight information of hatching eggs, and cannot detect core internal quality indicators that determine hatching performance, such as eggshell thickness, air cell ratio, yolk albumen status, and embryo viability. The detection dimensions are seriously insufficient, and the matching degree between the grading results and the actual hatching potential of the hatching eggs is extremely low; 2. This solution does not integrate hatching egg pretreatment and disinfection processes, and cannot achieve integrated processing of hatching eggs before incubation. The decentralized process during transportation can easily cause secondary contamination of hatching eggs, and the processing efficiency is difficult to meet the needs of large-scale production.

[0006] Publication number CN111802281B discloses a fiber optic spectral grading detection device and method for fertilized eggs in pre-hatching duck eggs. This scheme achieves non-destructive detection of fertilization information in duck eggs through near-infrared transmission spectroscopy. However, the technical shortcomings of this scheme are quite prominent: it can only make a single determination of whether or not fertilization has occurred, and cannot detect appearance and structural safety indicators such as eggshell cracks, dirt, and mechanical strength. It also cannot quantify and evaluate key hatching performance indicators such as albumen freshness and embryonic development stage, and cannot achieve a comprehensive evaluation of the hatching potential of the eggs. At the same time, the scheme does not involve the disinfection process of the eggs, nor does it form a continuous processing capability from pretreatment and disinfection to grading, making it difficult to directly apply to the pre-treatment production line of eggs in large-scale hatcheries. Summary of the Invention

[0007] Given the existing technical problems such as incomplete sterilization and embryo damage during pre-incubation disinfection of waterfowl hatching eggs, limited and low-precision detection dimensions, crude grading system, and the fact that the dispersed processes can easily lead to secondary pollution and low processing efficiency, a new disinfection, detection and grading method for waterfowl hatching eggs is proposed.

[0008] Its purpose is to achieve fully automated operation of the entire process of pretreatment, disinfection, testing and grading of hatching eggs, improve the disinfection effect and testing accuracy, complete the fine division of hatching egg incubation priority, eliminate secondary pollution, improve processing efficiency, and ultimately improve hatching rate, healthy chick rate and chick uniformity, reduce breeding costs and the risk of disease transmission during the incubation period, and adapt to the production needs of large-scale waterfowl farming.

[0009] The technical solution of this invention is a method for disinfection testing and grading of waterfowl hatching eggs, comprising the following steps:

[0010] S1 pretreatment first removes dust and impurities from the surface of collected waterfowl hatching eggs using a negative pressure dust removal device. Then, neutral electrolyzed water is sprayed onto the surface of the hatching eggs using a spraying device to remove solid impurities adhering to the surface. After spraying, a constant temperature and sterile airflow is delivered to the surface of the hatching eggs through a sterile air curtain to complete the constant temperature and humidity drying treatment of the surface of the hatching eggs, resulting in clean hatching eggs.

[0011] S2 gradient synergistic disinfection involves sequentially performing first-level low-temperature plasma treatment, ozone penetration treatment, and second-level low-temperature plasma treatment on clean hatching eggs to complete alternating gradient synergistic disinfection and obtain disinfected hatching eggs.

[0012] Furthermore, the gradient synergistic disinfection in step S2 specifically includes the following steps:

[0013] S21 sends clean hatching eggs into an atmospheric pressure low-temperature plasma reaction chamber, fixes the position of the hatching eggs, sets the plasma discharge parameters and reaction chamber environmental parameters, and performs first-level low-temperature plasma treatment on the hatching eggs.

[0014] S22 transfers the primary treated hatching eggs to a sealed ozone disinfection chamber, sets the ozone concentration, ambient temperature and humidity parameters in the disinfection chamber, and performs ozone permeation treatment on the hatching eggs.

[0015] S23 then sends the ozone-treated hatching eggs back into the low-temperature plasma reaction chamber, sets the plasma discharge parameters, and performs a second-stage low-temperature plasma treatment on the hatching eggs.

[0016] S3 multi-dimensional non-destructive testing simultaneously performs machine vision inspection, near-infrared spectroscopy analysis, and mechanical property testing on disinfected hatching eggs, acquiring multiple sets of detection features for each egg. All detection features are spatiotemporally registered and fused to obtain a multi-dimensional set of detection indicators for a single hatching egg.

[0017] Furthermore, in the machine vision inspection of step S3, panoramic images of the front and back of the hatching egg are acquired by a dual-view area array camera, and the acquired images are transmitted to the image processing unit. The images are then processed by a pre-trained defect detection model to extract the eggshell crack features, dirt area ratio features, and egg shape index features of the hatching egg.

[0018] Furthermore, in the near-infrared spectral analysis and detection in step S3, a probe light is emitted to the hatching egg through the light source of the near-infrared transmission spectral acquisition device, and the transmitted light penetrating the hatching egg is received by the spectrometer to obtain near-infrared transmission spectral data of the hatching egg within a preset wavelength range. After preprocessing the spectral data, it is input into a pre-trained spectral analysis model to extract the characteristics of the hatching eggshell thickness, air cell ratio, yolk index, albumen pH value, and embryonic development stage.

[0019] Furthermore, in the mechanical property testing of step S3, at least three evenly distributed testing points are selected on the equatorial plane of the hatching egg using a high-precision micro-force sensing device. Eggshell breakage strength testing is performed sequentially at each testing point to obtain strength data for each point, thus acquiring the eggshell mechanical strength characteristics of the hatching egg. Using the unique identifier of each hatching egg as a benchmark, all features obtained from machine vision inspection, near-infrared spectroscopy analysis, and mechanical property testing are registered one-to-one, and all registered feature data are normalized.

[0020] The S4 intelligent grading and sorting system is based on a multi-dimensional set of detection indicators. It classifies the indicator types and constructs a grading evaluation model. The grading evaluation model is used to classify the hatching priority level of the hatching eggs. Based on the grading results, it drives the automated actuator to complete the sorting of the hatching eggs.

[0021] Furthermore, step S4 specifically includes the following steps:

[0022] S41 categorizes the various indicators in the multi-dimensional detection indicator set into three types according to preset rules: core veto indicators, key evaluation indicators, and auxiliary optimization indicators.

[0023] S42 constructs a graded evaluation model, assigns weights to key evaluation indicators and quantifies their scores, and combines the deduction rules of auxiliary optimization indicators with the judgment results of core veto indicators to establish graded rules for the priority of hatching eggs.

[0024] S43 uses a grading evaluation model and grading rules to score and classify individual hatching eggs, and transmits the grading results to an automated sorting execution mechanism, which then transfers hatching eggs of different grades to their corresponding storage areas.

[0025] Corresponding to the above methods, the present invention also provides a waterfowl hatching egg disinfection, detection and grading system for performing any of the above-described waterfowl hatching egg disinfection, detection and grading methods, including a hatching egg pretreatment module, a gradient collaborative disinfection module, a multi-dimensional non-destructive testing module, an intelligent grading and sorting module, and a central control module that is electrically and communicatively connected to each of the above modules.

[0026] The pretreatment module for hatching eggs includes a built-in negative pressure dust removal unit, a spraying unit, and a sterile air drying unit, which sequentially perform dust removal, spraying to remove impurities, and air drying on the hatching eggs of waterfowl.

[0027] The gradient synergistic disinfection module has two sets of low-temperature plasma reaction units, a sealed ozone disinfection unit, and a sealed conveying unit built in, which sequentially perform two low-temperature plasma treatments and one ozone penetration treatment on the hatching eggs.

[0028] The multi-dimensional non-destructive testing module has a built-in machine vision inspection unit, near-infrared spectroscopy analysis unit, micro-mechanical detection unit, and multi-source data fusion unit, which simultaneously collects multiple sets of detection features of the hatching eggs and completes feature fusion.

[0029] The intelligent grading and sorting module has a built-in grading determination unit and an automated sorting execution unit to complete the grading and automated sorting of hatching eggs.

[0030] The central control module sends control parameters to each module, collects the operating data and detection data of each module, and coordinates the operating sequence of each module.

[0031] Furthermore, the first low-temperature plasma reaction unit, the sealed ozone disinfection unit, and the second low-temperature plasma reaction unit of the gradient collaborative disinfection module are arranged sequentially along the direction of egg transport. The transport track of the sealed transport unit runs through the cavity of the three units. The disinfection control subunit is communicatively connected to the first low-temperature plasma reaction unit, the sealed ozone disinfection unit, the second low-temperature plasma reaction unit, the sealed transport unit, and the central control module.

[0032] Furthermore, the machine vision inspection unit, near-infrared spectroscopy analysis unit, and micro-mechanical inspection unit of the multi-dimensional non-destructive testing module are arranged sequentially along the egg detection station. The input end of the multi-source data fusion unit is connected to the signal output ends of the machine vision inspection unit, near-infrared spectroscopy analysis unit, and micro-mechanical inspection unit, respectively. The output end of the multi-source data fusion unit is communicatively connected to the grade determination unit of the central control module and the intelligent grading and sorting module.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. This invention employs an alternating gradient synergistic disinfection process consisting of primary low-temperature plasma treatment, ozone infiltration treatment, and secondary low-temperature plasma treatment. This process achieves comprehensive elimination of pathogenic microorganisms from the eggshell surface and the shallow to deep layers of micropores, with the pathogen elimination rate verified to be over 99.9% in the examples. Furthermore, through low-power graded treatment and in-situ decomposition of ozone residues, damage to the eggshell and embryo is significantly reduced, with the embryo damage rate controlled to within 0.2%. Simultaneously, there are no harmful chemical residues throughout the process, avoiding occupational health and environmental risks, and balancing thorough sterilization, embryo safety, and environmental compliance.

[0035] 2. This invention integrates three major technical approaches: machine vision inspection, near-infrared spectroscopy analysis, and mechanical performance testing. It can acquire core indicators across all dimensions of hatching eggs, including appearance defects, egg shape characteristics, shell thickness, internal yolk protein status, embryo viability, and shell mechanical strength, all at once. Through spatiotemporal registration and fusion of multi-source data, it achieves a comprehensive quantitative characterization of the hatching potential of hatching eggs. Verification by examples shows that the accuracy rate of core indicator detection can reach over 99.5%, and the entire process is non-destructive, without damaging the egg structure or affecting subsequent hatching, providing reliable data support for subsequent accurate grading.

[0036] 3. This invention constructs a three-level indicator system consisting of "core veto indicators, key evaluation indicators, and auxiliary optimization indicators." Combined with a hierarchical evaluation model optimized by weighting using the analytic hierarchy process and machine learning model calibration, it enables precise quantitative scoring and grading of hatching priority for fertilized eggs. It quickly eliminates eggs with no hatching value through core veto items and accurately distinguishes the hatching potential of eggs through multi-indicator weighted scoring, matching different grades of eggs with corresponding hatching resources. This significantly improves the utilization rate of hatching resources and reduces ineffective hatching costs. Simultaneously, an automated sorting execution mechanism ensures rapid implementation of grading results, with sorting efficiency and accuracy far exceeding that of manual and traditional grading methods.

[0037] 4. This invention integrates four core processes—pretreatment, gradient-coordinated disinfection, multi-dimensional non-destructive testing, and intelligent grading and sorting—into a fully closed and time-sequential controlled process, completed entirely in a clean and sealed environment to prevent secondary contamination during egg transfer. Simultaneously, a central control module enables parameter coordination and cycle matching among the modules, achieving continuous and automated processing of individual eggs, significantly improving the processing efficiency of large-scale production. The accompanying full-process traceability mechanism allows for the binding and storage of complete data for each egg, supporting full lifecycle traceability of egg quality. This facilitates optimization of hatching and breeding processes for aquaculture enterprises, promoting the standardization and intelligent upgrading of the waterfowl farming industry. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the overall process of the waterfowl hatching egg disinfection, testing and grading method of the present invention;

[0039] Figure 2 This is a schematic diagram of the overall architecture of the waterfowl hatching egg disinfection, detection and grading system of the present invention;

[0040] Figure 3 This is a schematic diagram of the structural layout of the gradient collaborative disinfection module of the present invention;

[0041] Figure 4 This is a schematic diagram of the multi-dimensional non-destructive testing module structure of the present invention. Detailed Implementation

[0042] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0043] It should be noted that the "waterfowl hatching eggs" mentioned in this invention include, but are not limited to, hatching eggs of waterfowl such as ducks, geese, Muscovy ducks, and mandarin ducks. The following examples use Beijing duck hatching eggs and Anhui white goose hatching eggs as core examples for illustration, and are not intended to limit the scope of protection of this invention.

[0044] Example 1: Disinfection, testing, and grading of Beijing duck hatching eggs

[0045] This embodiment targets Beijing duck hatching eggs laid within 24 hours in large-scale farms. The waterfowl hatching egg disinfection, detection, and grading method and system described in this invention are used for large-scale processing. The specific steps are as follows:

[0046] like Figure 2 As shown, the waterfowl hatching egg disinfection, detection, and grading system used in this embodiment has the following specific structure:

[0047] The system includes a pre-treatment module for hatching eggs, a gradient collaborative disinfection module, a multi-dimensional non-destructive testing module, an intelligent grading and sorting module, a central control module, and a full-process traceability module, all connected in a sealed manner along the direction of hatching egg transport.

[0048] Egg pretreatment module: Built-in negative pressure dust removal unit (negative pressure fan and dust collection device, negative pressure value -5kPa), spraying unit (food-grade neutral electrolyzed water preparation device, high-pressure spray head, wastewater recycling device), aseptic air drying unit (aseptic air curtain machine, constant temperature heating component, temperature and humidity sensor), equipped with an automated feeding device and egg positioning fixture, each egg is fixed in an independent fixture position to avoid collision.

[0049] like Figure 3 As shown, the gradient collaborative disinfection module consists of a first low-temperature plasma reaction unit, a sealed ozone disinfection unit, and a second low-temperature plasma reaction unit arranged sequentially along the conveying direction. The low-temperature plasma reaction unit uses an atmospheric pressure glow discharge flat-plate plasma generator, equipped with a pressure regulating device and a discharge parameter controller. The sealed ozone disinfection unit is equipped with an ozone generator, an ozone concentration detector, a temperature and humidity regulating device, and an exhaust gas decomposition device. The sealed conveying unit uses a closed roller conveyor belt, with the conveying speed matched to the processing time of each station, maintaining a sealed and sterile environment throughout the process.

[0050] like Figure 4 As shown, the multi-dimensional non-destructive testing module is arranged in sequence along the conveying direction, including a machine vision inspection station, a near-infrared spectroscopy analysis station, and a mechanical property testing station, and is equipped with an egg posture adjustment mechanism to ensure that the egg posture is fixed during the testing process.

[0051] Machine vision inspection unit: It adopts two 5-megapixel area array cameras, which are respectively arranged above and below the hatching egg fixture. With the help of a ring shadowless light source and a diffuse backlight, it can realize panoramic image acquisition of hatching eggs without blind spots, and is equipped with an industrial image processing control computer.

[0052] Near-infrared spectroscopy analysis unit: adopts halogen broadband light source and fiber optic spectrometer, with spectral acquisition range covering 800-2500nm, spectral resolution ≤2nm, and equipped with light source collimation component and dark box to shield against ambient light interference;

[0053] Micro-force mechanical detection unit: It adopts a high-precision micro-force sensor with a force measurement range of 0-50N and a detection accuracy of 0.01N. It is equipped with a servo drive mechanism to realize the precise positioning and pressure control of the detection point.

[0054] Multi-source data fusion unit: Employs an embedded industrial controller to receive, register, preprocess, and fuse multi-source detection data.

[0055] Intelligent grading and sorting module: It has a built-in grade determination unit (embedded industrial control computer) and a multi-channel automated sorting actuator (pneumatic push rod, multi-channel conveyor belt), which can realize the synchronous sorting of hatching eggs of 4 grades.

[0056] Central control module: It adopts PLC + industrial touch screen, and is equipped with configuration control system to realize parameter setting, timing coordination, operation status monitoring and data interaction of each module.

[0057] End-to-end traceability module: Utilizes RFID reading and writing devices to equip each hatching egg with a unique RFID tag, enabling the binding, recording, and storage of data for each hatching egg throughout the entire process.

[0058] like Figure 1 As shown, the specific steps of the waterfowl hatching egg disinfection detection and grading method used in this embodiment are as follows:

[0059] S1 Egg Pretreatment: Beijing duck hatching eggs laid within 24 hours are placed in a feeding device and conveyed to the pretreatment module via a conveyor belt. First, a negative pressure dust removal unit removes dust and loose impurities from the surface of the eggs. Then, a spray unit sprays neutral electrolyzed water, controlling the spray pressure at 0.15 MPa, the effective chlorine concentration of the electrolyzed water at 60 mg / L, the pH value of the electrolyzed water at 6.8-7.2, and the water temperature at a constant 23℃, to remove feces, dried egg liquid, and other impurities from the surface of the eggs. After spraying, the eggs are sent to the drying station, where a sterile airflow at 24℃ is delivered through a sterile air curtain. The drying time is 45 seconds. After testing, the surface moisture content of the dried eggs is 2.2%, resulting in clean hatching eggs.

[0060] S2 Gradient Synergistic Disinfection: Clean hatching eggs are fed into the gradient synergistic disinfection module via a sealed conveyor belt, circulating entirely within a Class 10,000 cleanroom environment with no external exposure, and undergoing the following three-step gradient synergistic sterilization process:

[0061] S21 Level 1 Low-Temperature Plasma Pretreatment: Hatching eggs are conveyed into the first low-temperature plasma reaction chamber via a conveyor belt. After the chamber's feed door automatically closes, plasma discharge treatment is initiated. This process is tailored to the characteristics of Beijing duck hatching eggs, which have thin shells and small micropores. The plasma discharge power is controlled at 65W, the reaction chamber pressure is stabilized at 0.1MPa, and the treatment time is 15 seconds. The active particles generated by the low-power glow discharge achieve the initial elimination of pathogenic microorganisms on the eggshell surface and in the shallow micropores, while avoiding thermal damage to the eggshell caused by high-power discharge. After treatment, the chamber's discharge door automatically opens, and the hatching eggs are conveyed to the next station via a sealed conveyor belt.

[0062] This process utilizes an atmospheric pressure glow discharge flat-plate plasma generator. The electrodes are made of food-grade 304 stainless steel, with a 1mm thick alumina insulating layer sprayed onto the electrode surface to prevent arc burns to the hatching eggs during discharge. The distance between the upper and lower electrodes is set at 10cm, perfectly matching the height of the hatching egg fixture, ensuring that the entire surface of the hatching eggs is within a uniform discharge area. Dry, clean air is used as the working gas, eliminating the need for additional special working gases and making it suitable for the on-site conditions of large-scale farms. The gas flow rate is controlled at 5L / min, and the discharge frequency is 40kHz. The cavity sealing door uses an interlocking timing control: plasma discharge starts 0.5s after the feed door is completely closed, and the discharge door opens 0.5s after the discharge ends, ensuring the cavity remains airtight throughout the process and preventing the leakage of active particles from affecting surrounding workstations.

[0063] S22 Ozone Permeation Treatment: Hatching eggs are fed into a sealed ozone disinfection chamber via a closed conveyor belt. The chamber's sealed door is closed, and the ozone concentration inside the chamber is controlled at 4 mg / m³, with an ambient temperature of 22℃ and a relative humidity of 50%. The treatment time is 25 seconds, which completes the killing of pathogenic microorganisms deep within the eggshell micropores. After the treatment is completed, the discharge sealed door is opened, and the hatching eggs are discharged. The residual ozone in the chamber is treated by the exhaust gas decomposition device before being released.

[0064] The ozone disinfection chamber adopts a top-feed, bottom-return circulating airflow layout, with an internal airflow velocity of 1.5 m / s. The spacing between individual hatching eggs in the egg-laying fixture is ≥2 cm to ensure that the ozone flow evenly covers the entire surface of the eggs. After ozone treatment, the chamber first performs a 1-second residual gas extraction before opening the discharge door. The matching exhaust gas decomposition device uses a manganese-based ozone decomposition catalyst with a space velocity of 10,000 h⁻¹. -1 The ozone decomposition efficiency is ≥99.9%, and the exhaust emission concentration meets the GB3095-2012 ambient air quality standard.

[0065] S23 Secondary Low-Temperature Plasma Passivation Treatment: Hatching eggs are fed into the second low-temperature plasma reaction chamber via a sealed conveyor belt. The chamber's sealing door is closed, and the plasma discharge power is controlled at 40W for 8 seconds. This process decomposes residual ozone on the surface of the hatching eggs and simultaneously kills any remaining pathogenic microorganisms. After the treatment is completed, the hatching eggs are sent out, resulting in sterilized hatching eggs.

[0066] The electrode structure, working gas, and discharge frequency of the plasma generator are consistent with those of the primary processing unit. The timing control of the cavity sealing door is matched with that of the primary processing unit. After the secondary processing is completed and before the hatching eggs are discharged, the cavity is purged with clean air for 0.5 seconds to avoid residual active particles affecting subsequent testing.

[0067] The sealed conveyor unit between the three disinfection processes uses a closed roller conveyor belt. The conveying environment is Class 10,000 cleanroom, and the cavity maintains a positive pressure of 10Pa. The conveying time of a single hatching egg between adjacent workstations is ≤2s. There are no links in the process that are exposed to the external environment, thus avoiding secondary contamination.

[0068] Full-process time-sequential collaborative control: The central control module synchronously controls the cycle time of the three disinfection processes. One second before the previous station finishes processing, the feeding door of the next station is in a ready-to-open state, ensuring that the entire disinfection process of a single hatching egg is uninterrupted and guaranteeing precise control of the processing time.

[0069] S3 Multi-Dimensional Non-Destructive Testing: After disinfection, the hatching eggs are directly sent to the multi-dimensional non-destructive testing module, and testing is performed synchronously at each station, specifically including:

[0070] S31 Machine Vision Inspection: Hatching eggs are sent to the vision inspection station, where a dual-view area array camera simultaneously captures panoramic images of the front and back of the hatching eggs. The images are transmitted to an image processing industrial control computer, where they are processed by a pre-trained defect detection model to extract features such as eggshell cracks (distinguishing between continuous and non-continuous cracks and recording crack length), dirt area ratio, and egg shape index.

[0071] The pre-trained defect detection model is a YOLOv8 defect detection model. Its training process involves: collecting 120,000 images of the surface of Beijing duck eggs; having professionals label the eggs with tags for through-cracks, non-through microcracks, dirt, and egg shape contours; and dividing the images into training, validation, and test sets in an 8:1:1 ratio. The training set images undergo data augmentation processing including normalization, random flipping, scaling, and brightness adjustment. Transfer learning is then performed based on the YOLOv8n pre-trained weights, with 300 training epochs, an initial learning rate of 0.001, and a batch size of 16. After training, the model is validated on the test set, achieving a crack detection accuracy of ≥99.6% and a dirt recognition accuracy of ≥99.8%, meeting industrial inspection requirements.

[0072] S32 Near-Infrared Spectroscopy Analysis: Hatching eggs are sent to the spectral detection station. A halogen light source emits broadband detection light into the eggs, and the spectrometer receives the transmitted light that penetrates the eggs, acquiring near-infrared transmission spectral data of the eggs in the 800-2500nm band. The raw spectral data is then subjected to Savitzky-Golay smoothing (window size 7, polynomial order 2), multivariate scattering correction, and baseline correction preprocessing. The preprocessed spectral data is then input into a pre-trained spectral analysis model to extract the characteristics of eggshell thickness, air cell ratio, yolk index, albumen pH value, and embryonic development stage.

[0073] The pre-trained spectral analysis model was the PLS-DA model. The training process was as follows: Near-infrared transmission spectral data of 15,000 Beijing duck hatching eggs were collected. Simultaneously, the eggshell thickness, air cell ratio, yolk index, albumen pH value, and true labeled values ​​of embryonic development stage of the corresponding hatching eggs were obtained through destructive detection methods. After preprocessing the spectral data, the training set and validation set were divided in a 7:3 ratio to construct the PLS-DA model. The number of latent variables in the model was optimized by leave-one-out cross-validation. Finally, the prediction determination coefficient R² of the model for each indicator was ≥0.93, and the prediction relative error was ≤5%, which met the detection requirements.

[0074] S33 Mechanical Performance Testing: The hatching egg is sent into the mechanical testing station. The servo mechanism drives the micro-force sensor to select four evenly distributed testing points on the equatorial plane of the hatching egg. Pressure testing is performed on each testing point in sequence, and the critical pressure value when the eggshell deforms is recorded. The average value of the four testing points is taken to obtain the mechanical strength characteristics of the eggshell.

[0075] The micro-force sensor used is paired with a pressure head made of medical-grade silicone. The pressure head has a diameter of 10mm, and its spherical curvature matches the equatorial curvature of the egg. The control parameters for the pressure application process are: pressure rate 0.5N / s. When the force change rate is detected to be ≥50% / 0.1s, it is determined to be the critical point of eggshell deformation. The pressure application is stopped immediately and the critical pressure value is recorded to avoid eggshell breakage. After the detection of a single detection point is completed, the servo mechanism drives the sensor to reset. The reset time is ≤0.5s. The egg is rotated 90° by the egg posture adjustment mechanism to switch to the next detection point, ensuring that the four detection points are evenly distributed along the equatorial plane.

[0076] S34 Multi-Source Feature Fusion: Based on the unique RFID identifier of the hatching egg tool, all features obtained by machine vision inspection, near-infrared spectral analysis, and mechanical performance testing are spatiotemporally registered one-to-one; min-max normalization is performed on all registered feature data to map all index values ​​to the 0-1 range, eliminating dimensional differences, and obtaining a standardized multi-dimensional detection index set for the hatching egg, which is synchronously transmitted to the central control module and the intelligent grading and sorting module.

[0077] S4 Intelligent Grading and Sorting:

[0078] S41 Indicator Classification: According to preset rules, the multi-dimensional detection indicator set is divided into three categories: core rejection indicators (penetrating cracks in the eggshell, unqualified embryo viability), key evaluation indicators (eggshell thickness, eggshell mechanical strength, air cell ratio, egg shape index), and auxiliary optimization indicators (proportion of soiled area, non-penetrating microcracks). Among them, the criteria for judging unqualified embryo viability are: mismatch between embryonic development stage and post-laying time, yolk index exceeding the normal threshold, and albumen pH value exceeding the suitable range for incubation.

[0079] Through-cracks in eggshells: Identified by the defect detection model, cracks that penetrate the inner and outer walls of the eggshell, are visible as continuous light-transmitting channels under backlight, and have a length ≥0.5mm are considered through-cracks, triggering a core rejection and directly classifying them as D-level discarded hatching eggs.

[0080] Embryo viability is unacceptable; meeting any of the following conditions will result in disqualification:

[0081] The yolk index exceeds the normal threshold range of 0.40-0.55.

[0082] The protein pH value is outside the suitable incubation range of 7.6-8.2.

[0083] If the embryonic development stage does not match the postpartum period, and the embryonic development of the hatching egg enters the blastocyst stage or later within 24 hours after laying, it is considered a developmental abnormality.

[0084] S42 Grading Evaluation Model Construction and Rule Setting: Based on the analytic hierarchy process (AHP), combined with industry data on Beijing duck egg hatching and expert experience, a judgment matrix was constructed to determine the weights of key evaluation indicators.

[0085] Eggshell thickness weighted at 0.25, eggshell mechanical strength weighted at 0.25, air cell ratio weighted at 0.3, and egg shape index weighted at 0.2; the comprehensive score is out of 100 points, and the scores are quantified based on the suitable incubation range for each indicator.

[0086] The quantitative scoring rules for key evaluation indicators (out of 100 points) are shown in the table below:

[0087] Key Indicators Weight Full marks Suitable range for Beijing duck hatching eggs Scoring Calculation Rules Eggshell thickness 0.25 25 points 0.33-0.38mm (reference value 0.35mm) For measured values ​​between 0.33 and 0.38 mm, a value of 0.35 mm receives a perfect score of 25 points, with 1 point deducted for every 0.01 mm deviation. Measured values ​​< 0.30 mm or > 0.40 mm receive 0 points for this item. Eggshell mechanical strength 0.25 25 points 35-45N (reference value 40N) If the measured value is within the range of 35-45N, a score of 25 points is awarded for 40N, with 1 point deducted for each deviation of 1N; if the measured value is <30N or >50N, 0 points are awarded for this item. air chamber ratio 0.3 30 points 1.5%–3.0% (benchmark 2.0%) If the measured value is within the range of 1.5% to 3.0%, a score of 2.0% is awarded, with 2 points deducted for each deviation of 0.2%. If the measured value is greater than 5.0%, 0 points are awarded for this item. Egg-shaped index 0.2 20 points 0.72-0.76 (Base value 0.74) The measured value is between 0.72 and 0.76. A value of 0.74 receives the full score of 20 points, and 2 points are deducted for each deviation of 0.01. If the measured value is <0.65 or >0.85, 0 points are awarded for this item.

[0088] Set the scoring rules for auxiliary optimization indicators:

[0089] Dirty area percentage: Based on the panoramic projection area of ​​the hatching egg, a dirty area percentage of ≥5% will deduct 5 points, a dirty area percentage of ≥10% will deduct 10 points, and the maximum deduction for a single hatching egg is 10 points.

[0090] Non-penetrating microcracks: Microcracks that only break the surface / inner layer of the eggshell and have no continuous light transmission channel. For each additional 1mm of length ≥2mm, 2 points will be deducted, with a maximum deduction of 10 points for a single hatching egg; for a total crack length ≥10mm, all 10 points will be deducted.

[0091] The hierarchical evaluation model of this invention adopts a fusion architecture of "analytic hierarchy process (AHP) weight assignment + XGBoost machine learning model calibration and optimization", and the complete implementation details are as follows:

[0092] Model architecture: Based on the basic score obtained by the analytic hierarchy process, the score is calibrated again by the XGBoost machine learning model to improve the matching degree between the grading results and the actual incubation effect.

[0093] Training dataset: Collected data on all dimensions of detection indicators and subsequent full-cycle hatching results of 50,000 Beijing duck hatching eggs, labeled with successful / failed hatching and healthy / weak chicks, and divided into training set, validation set and test set in a ratio of 8:1:1.

[0094] The core parameters of the model are: tree depth 6, learning rate 0.05, number of iterations 200, loss function is binary cross-entropy, and optimizer is gradient boosting tree algorithm.

[0095] Inference rules: The model input consists of normalized key evaluation indicators and full features of auxiliary optimization indicators. The output is the predicted value of the hatching success rate of the hatching eggs. Based on the predicted value, the basic score of the analytic hierarchy process is calibrated within a range of ±5 points to obtain the comprehensive score of the hatching eggs, thus ensuring the accuracy of the grading results.

[0096] Establish hierarchical rules:

[0097] Grade A priority hatching eggs: No core veto items, comprehensive score ≥ 90 points.

[0098] Grade B conventional hatching eggs: No core veto items, overall score 70-89 points.

[0099] Grade C downgraded hatching eggs: No core veto items, overall score 60-69 points.

[0100] Grade D Elimination Eggs: Existing core veto items, or overall score < 60 points.

[0101] S43 Grading and Sorting: The grading unit calls the grading evaluation model and grading rules to comprehensively score and grade individual hatching eggs, and sends the grading results to the sorting execution mechanism; the sorting execution mechanism uses pneumatic pushers to push hatching eggs of different grades to the corresponding grade conveyor belts and storage areas to complete the entire sorting process.

[0102] Meanwhile, the end-to-end traceability module binds and stores the pre-processing parameters, disinfection parameters, multi-dimensional detection indicators, and grading results of each hatching egg with a unique RFID identifier, supporting end-to-end traceability queries.

[0103] Example 2: Disinfection, testing, and grading of Anhui White Goose hatching eggs. This example targets Anhui White Goose hatching eggs within 36 hours post-laying. The method and system described in this invention are used for processing. The system hardware architecture is consistent with Example 1. The tooling dimensions, equipment parameters, and model weights are adjusted based on the size and shell structure characteristics of the goose hatching eggs. For hatching eggs from large waterfowl such as Anhui White Goose, grading parameter adaptation rules are simultaneously added.

[0104] The appropriate ranges for key evaluation indicators are as follows: suitable range for eggshell thickness is 0.45-0.50 mm, suitable range for eggshell mechanical strength is 45-55 N, suitable range for egg shape index is 0.70-0.74, and suitable range for air cell ratio is 2.0%-3.5%.

[0105] Weight re-optimization: The judgment matrix was reconstructed based on the analytic hierarchy process. The optimized weights are: eggshell mechanical strength 0.3, air cell ratio 0.25, eggshell thickness 0.25, and egg shape index 0.2. Consistency test (CR < 0.1) was performed simultaneously to ensure the rationality of the weights.

[0106] The threshold values ​​of the core rejection indicators are adapted to the breed characteristics of goose eggs. The suitable range for the yolk index (the ratio of yolk height to yolk diameter, which is a core indicator of albumen freshness) is adjusted to 0.42-0.58, and the suitable range for albumen pH value is adjusted to 7.5-8.3.

[0107] The specific implementation steps are as follows:

[0108] S1 Egg Pretreatment: The eggs are fed into the pretreatment module. First, surface impurities are removed by negative pressure dust removal. Then, neutral electrolyzed water is sprayed through the spray unit. The spray pressure is controlled at 0.2 MPa and the effective chlorine concentration of the electrolyzed water is 75 mg / L to remove surface impurities. After spraying, the eggs are dried in a sterile air curtain at 25°C for 60 seconds. After drying, the surface moisture content of the eggs is 2.5%, resulting in clean eggs.

[0109] S2 gradient synergistic disinfection: S21 first-level low-temperature plasma treatment: control plasma discharge power 75W, reaction chamber gas pressure 0.11MPa, treatment time 18s.

[0110] S22 Ozone Infiltration Treatment: Control the ozone concentration in the disinfection chamber to 4.5 mg / m³, ambient temperature to 24℃, relative humidity to 52%, and treatment time to 28 seconds.

[0111] S23 Level 2 Low Temperature Plasma Passivation Treatment: Control the plasma discharge power to 45W, and the treatment time to 9s to complete the disinfection.

[0112] S3 Multidimensional Nondestructive Testing: Targeting the characteristics of Anhui White Goose hatching eggs, the training datasets of the defect detection model and the spectral analysis model were optimized to adapt to the eggshell thickness and shape characteristics of goose hatching eggs. Simultaneously, surface defect features, eggshell structure features, internal embryo viability features, and eggshell mechanical strength features of the hatching eggs were collected to complete the fusion of multi-source features and obtain a standardized multidimensional detection index set.

[0113] S4 Intelligent Grading and Sorting: Based on the hatching characteristics of Anhui White Goose hatching eggs, the weights of the indicators in the grading evaluation model have been adjusted: eggshell mechanical strength weight 0.3, air cell ratio weight 0.25, eggshell thickness weight 0.25, and egg shape index (the ratio of the short axis to the long axis of the hatching egg, a core parameter adapted to incubation equipment) weight 0.2. The quantitative scoring thresholds and grading rules of each indicator have been optimized to complete the grading and automated sorting of hatching eggs.

[0114] Example 3: Verification of the stability of process parameter boundary values

[0115] The core verification objective of this embodiment is to confirm that the process parameters of the present invention can still maintain stable sterilization effect, embryo safety and detection accuracy under the lower limit of the protection range, and adapt to the actual production scenario of fluctuating hatching egg quality and unstable on-site conditions in large-scale breeding.

[0116] This embodiment processes Cherry Valley duck hatching eggs laid within 24 hours in large-scale farms. The core process parameters all adopt the lower limit of the scope of protection of this invention. The specific implementation process is as follows:

[0117] S1 Egg Pretreatment: Considering the thin shells and susceptibility to corrosion of Cherry Valley duck eggs, the spray pressure was controlled at 0.1 MPa, and neutral electrolyzed water with an effective chlorine concentration of 50 mg / L was used for spraying to remove impurities. After spraying, the eggs were air-dried at 22℃ under a sterile air curtain for 60 seconds. After treatment, the surface moisture content of the eggs was measured to be 2.8%, which meets the cleanliness requirements of subsequent disinfection processes.

[0118] S2 gradient synergistic disinfection: Gradient sterilization is completed using lower limit parameters. Specific process parameters are as follows:

[0119] S21 Level 1 Low Temperature Plasma Pretreatment: Control the discharge power to 50W, the gas pressure in the reaction chamber to 0.08MPa, and the treatment time to 20s. The low power and long treatment time method is adopted to ensure the shallow sterilization effect while avoiding heat damage to the eggshell.

[0120] S22 Ozone Infiltration Treatment: Control the ozone concentration in the disinfection chamber to 3mg / m³, ambient temperature to 20℃, relative humidity to 45%, and treatment time to 30s. The treatment time is extended to compensate for the infiltration sterilization capacity of low-concentration ozone.

[0121] S23 secondary low-temperature plasma passivation treatment: control the discharge power to 30W, the treatment time to 10s, and complete the decomposition of residual ozone and the elimination of residual pathogens.

[0122] The S3 multi-dimensional non-destructive testing and the S4 intelligent grading and sorting process are consistent with the equipment, model and process of Example 1, completing the full-process testing and grading of hatching eggs.

[0123] After the batch processing in this embodiment was completed, third-party testing verified that the pathogen eradication rate of the hatching eggs reached 99.91%, the embryo damage rate was only 0.15%, and the accuracy rate of core indicator detection was 99.58%. Subsequent full-cycle incubation tests showed that the hatching rate and chick survival rate of this batch of hatching eggs were not significantly different from those in Example 1, which fully demonstrates that the parameter protection scope of the present invention has excellent feasibility and working condition stability.

[0124] Example 4: Parameter Boundary Value Verification Example. This example verifies another set of boundary values ​​for the process parameter range of the present invention. The processing object is Muscovy duck hatching eggs within 24 hours post-laying. The core process parameters are as follows:

[0125] S1 hatching egg pretreatment: spray pressure 0.2MPa, effective chlorine concentration of electrolyzed water 80mg / L, air drying temperature 26℃, air drying time 30s, surface moisture content of hatching eggs 2.6%;

[0126] S2 gradient synergistic disinfection:

[0127] S21 Level 1 Low Temperature Plasma Treatment: Discharge power 80W, reaction chamber pressure 0.12MPa, treatment time 10s;

[0128] S22 ozone infiltration treatment: ozone concentration 5mg / m³, ambient temperature 25℃, relative humidity 55%, treatment time 20s;

[0129] S23 Secondary low-temperature plasma treatment: discharge power 50W, treatment time 5s; S3-S4 steps are the same as in Example 1, to complete the detection and grading of hatching eggs.

[0130] The pathogen eradication rate of this embodiment was 99.92%, the embryo damage rate was 0.18%, and the detection accuracy was 99.55%. The subsequent hatching effect was not significantly different from that of Example 1, which further verified the reliability of the parameter range of the present invention.

[0131] For hatching eggs from different breeds of waterfowl such as ducks and geese, the suitable range of core pretreatment parameters is as follows: effective chlorine concentration of neutral electrolyzed water 50-80 mg / L, pH value 6.5-7.5, spray pressure 0.1-0.2 MPa, air drying temperature 22-26℃, and air drying time 30-60 seconds. These parameters can be adjusted within the range according to the eggshell thickness and surface contamination level to ensure effective impurity removal while avoiding eggshell corrosion.

[0132] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for disinfection testing and grading of waterfowl hatching eggs, characterized in that, Includes the following steps: S1 pretreatment involves removing surface impurities from the collected waterfowl hatching eggs, followed by constant temperature and humidity air drying to obtain clean hatching eggs. S2 gradient synergistic disinfection involves sequentially performing first-level low-temperature plasma treatment, ozone penetration treatment, and second-level low-temperature plasma treatment on clean hatching eggs to complete alternating gradient synergistic disinfection and obtain disinfected hatching eggs. S3 multi-dimensional non-destructive testing simultaneously performs machine vision inspection, near-infrared spectroscopy analysis, and mechanical property testing on disinfected hatching eggs, acquiring multiple sets of detection features of the hatching eggs, and performing spatiotemporal registration and fusion of all detection features to obtain a multi-dimensional detection index set corresponding to a single hatching egg. The S4 intelligent grading and sorting system is based on a multi-dimensional set of detection indicators. It classifies the indicator types and constructs a grading evaluation model. The grading evaluation model is used to classify the hatching priority level of the hatching eggs. Based on the grading results, it drives the automated actuator to complete the sorting of the hatching eggs.

2. The method for disinfection detection and grading of waterfowl hatching eggs according to claim 1, characterized in that: In step S1, dust and impurities on the surface of the hatching eggs are first removed by a negative pressure dust removal device, and then neutral electrolyzed water is sprayed onto the surface of the hatching eggs by a spraying device to remove solid impurities attached to the surface of the hatching eggs. After spraying, a constant temperature sterile airflow is delivered to the surface of the hatching eggs through a sterile air curtain to complete the air drying treatment of the surface of the hatching eggs.

3. The method for disinfection detection and grading of waterfowl hatching eggs according to claim 1, characterized in that: The gradient-coordinated disinfection in step S2 specifically includes the following steps: S21 sends the clean hatching eggs into the atmospheric pressure low-temperature plasma reaction chamber, fixes the position of the hatching eggs, sets the plasma discharge parameters and reaction chamber environmental parameters, and performs first-level low-temperature plasma treatment on the hatching eggs. S22 transfers the primary treated hatching eggs to a sealed ozone disinfection chamber, sets the ozone concentration and ambient temperature and humidity parameters in the disinfection chamber, and performs ozone permeation treatment on the hatching eggs. S23 sends the ozone-treated hatching eggs back into the low-temperature plasma reaction chamber, sets the plasma discharge parameters, and performs a second-stage low-temperature plasma treatment on the hatching eggs.

4. The method for disinfection detection and grading of waterfowl hatching eggs according to claim 1, characterized in that: In the machine vision inspection of step S3, panoramic images of the front and back of the hatching egg are acquired by a dual-view area array camera. The acquired images are transmitted to the image processing unit, and the images are processed by a pre-trained defect detection model to extract the eggshell crack features, dirt area ratio features, and egg shape index features of the hatching egg.

5. The method for disinfection detection and grading of waterfowl hatching eggs according to claim 1, characterized in that: In the near-infrared spectral analysis and detection in step S3, the light source of the near-infrared transmission spectral acquisition device emits detection light to the hatching egg, and the transmitted light penetrating the hatching egg is received by the spectrometer to obtain the near-infrared transmission spectral data of the hatching egg within a preset wavelength range. After preprocessing the spectral data, it is input into a pre-trained spectral analysis model to extract the characteristics of the hatching egg's shell thickness, air cell ratio, yolk index, albumen pH value, and embryonic development stage.

6. The method for disinfection detection and grading of waterfowl hatching eggs according to claim 1, characterized in that: In the mechanical performance testing of step S3, at least three evenly distributed testing points are selected on the equatorial plane of the hatching egg using a high-precision micro-force sensing device. The eggshell fracture resistance strength test is performed on each testing point in sequence to obtain the strength data of each testing point and obtain the mechanical strength characteristics of the eggshell. Based on the unique identifier of a single hatching egg, all features obtained by machine vision detection, near-infrared spectroscopy analysis and mechanical performance testing are registered one-to-one, and all registered feature data are normalized.

7. The method for disinfection detection and grading of waterfowl hatching eggs according to claim 1, characterized in that: Step S4 specifically includes the following steps: S41 categorizes the various indicators in the multi-dimensional detection indicator set into three types according to preset rules: core veto indicators, key evaluation indicators, and auxiliary optimization indicators. S42 constructs a graded evaluation model, assigns weights to key evaluation indicators and quantifies their scores, and combines the deduction rules of auxiliary optimization indicators with the judgment results of core veto indicators to establish graded rules for the priority of hatching eggs. S43 uses a grading evaluation model and grading rules to score and classify individual hatching eggs, and transmits the grading results to an automated sorting execution mechanism, which then transfers hatching eggs of different grades to their corresponding storage areas.

8. A disinfection, detection, and grading system for waterfowl hatching eggs, characterized in that, The method for disinfection, detection and grading of waterfowl hatching eggs according to any one of claims 1-7 includes a hatching egg pretreatment module, a gradient collaborative disinfection module, a multi-dimensional non-destructive testing module, an intelligent grading and sorting module, and a central control module that is electrically and communicatively connected to each of the above modules respectively. The pre-treatment module for hatching eggs includes a built-in negative pressure dust removal unit, a spraying unit, and a sterile air drying unit, which sequentially perform dust removal, spraying to remove impurities, and air drying on waterfowl hatching eggs. The gradient synergistic disinfection module has two sets of low-temperature plasma reaction units, a closed ozone disinfection unit, and a closed conveying unit built in, and performs two low-temperature plasma treatments and one ozone penetration treatment on the hatching eggs in sequence. The multi-dimensional non-destructive testing module has a built-in machine vision testing unit, near-infrared spectroscopy analysis unit, micro-mechanical testing unit, and multi-source data fusion unit, which simultaneously collects multiple sets of detection features of the hatching eggs and completes feature fusion. The intelligent grading and sorting module has a built-in grading determination unit and an automated sorting execution unit to complete the grading and automated sorting of hatching eggs; The central control module sends control parameters to each module, collects the operating data and detection data of each module, and coordinates the operating sequence of each module.

9. A disinfection, detection, and grading system for waterfowl hatching eggs according to claim 8, characterized in that: The first low-temperature plasma reaction unit, the sealed ozone disinfection unit, and the second low-temperature plasma reaction unit of the gradient collaborative disinfection module are arranged sequentially along the direction of egg transport. The transport track of the sealed transport unit runs through the cavity of the three units. The disinfection control subunit is communicatively connected to the first low-temperature plasma reaction unit, the sealed ozone disinfection unit, the second low-temperature plasma reaction unit, the sealed transport unit, and the central control module.

10. A disinfection, detection, and grading system for waterfowl hatching eggs according to claim 8, characterized in that: The machine vision inspection unit, near-infrared spectroscopy analysis unit, and micro-mechanical inspection unit of the multi-dimensional non-destructive testing module are arranged sequentially along the egg detection station. The input end of the multi-source data fusion unit is connected to the signal output ends of the machine vision inspection unit, near-infrared spectroscopy analysis unit, and micro-mechanical inspection unit, respectively. The output end of the multi-source data fusion unit is communicatively connected to the grade determination unit of the central control module and the intelligent grading and sorting module.

Images