Chick gender detection system

By using an inclined conveyor belt and airflow stimulation to encourage chicks to spread their wings in a chick sex detection system, combined with multi-frame image recognition and deep learning algorithms, efficient and accurate automated detection of chick sex is achieved, solving the problems of detection accuracy and misjudgment in existing technologies.

CN120458039BActive Publication Date: 2025-10-24HUAZHONG AGRI UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510971909.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-07-15
Publication Date
2025-10-24
Estimated Expiration
2045-07-15

AI Technical Summary

Technical Problem

Existing methods for sex detection in chicks are not highly accurate when automated, and manual operation can easily cause tissue damage and has a high rate of misjudgment.

Method used

A chick sex detection system based on machine vision and automated control is adopted. By creating a brief weightlessness state through an inclined conveyor belt, the chicks are encouraged to spread their wings. Combined with airflow stimulation and multi-frame image recognition, a deep learning algorithm is used to determine the sex of the chicks, achieving fully automated detection.

Benefits of technology

It improves the accuracy and efficiency of sex detection in chicks, reduces reliance on manual operation and stress response in chicks, and lowers the false positive rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120458039B_ABST
    Figure CN120458039B_ABST
Patent Text Reader

Abstract

The application provides a chick gender detection system, and belongs to the field of agricultural product detection.The system comprises a feeding mechanism, a collecting mechanism, an angle adjusting rod and a detection unit;chick samples are sequentially transmitted through the feeding mechanism, a first conveying belt and a second conveying belt;the collecting mechanism comprises a first camera and a first light source arranged in a first dark box, a second camera, a second light source and a pneumatic nozzle in a second dark box;the first conveying belt passes through the first dark box;the feeding mechanism is horizontally transmitted to the first conveying belt;the first conveying belt is an inclined conveying belt;the second conveying belt is a horizontal conveying belt;the detection unit comprises an acquisition module for acquiring an image sequence of the chick samples when passing through the first dark box and the second dark box through the camera;an angle adjusting rod is used to adjust the inclination angle of the first conveying belt before detection;and a processing module is used to determine the chick gender detection result according to the proportion of detected genders in multiple images.The system improves the chick gender detection precision.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agricultural product detection, and in particular to a system for detecting the gender of chicks. BACKGROUND

[0002] Poultry meat and eggs occupy an increasingly important position in the dietary structure of residents, and chickens are a large part of poultry consumption. In broiler production, male chicks with faster feed conversion rates and growth rates are desired, while in egg production, more female chicks are desired to increase egg production. Generally, female chicks lay eggs early, produce more eggs, and consume less feed, while male chicks can be sold as high-quality chickens when they reach adulthood. Therefore, gender selection of one-day-old chicks and fine management according to gender differences can effectively improve breeding efficiency and economic benefits, and thus gender selection detection methods for one-day-old chicks have urgent practical needs and important significance. Traditional methods for detecting the gender of chicks include the anal palpation method and molecular biology methods. The anal palpation method requires high technical requirements and can cause irritation and damage to the chicks. In actual operation, not only is the labor intensity high, but the identification results are often affected by the subjective inclination of the operator.

[0003] With the continuous progress of poultry breeding technology, methods for gender identification using fast and slow feathering significant sex-linked genetic traits have been widely used. This method identifies different genders of chicks by observing the phenotypic differences in feather growth speed, which can effectively avoid the misjudgment caused by tissue damage and visual fatigue induced by operation compared to the traditional anal palpation method. With the advantages of simple operation, high detection efficiency, etc., the feather speed identification method significantly reduces the stress response of chicks.

[0004] However, the current method for detecting the gender of chicks based on machine vision and automatic control technology has low accuracy and automation level. SUMMARY

[0005] In view of the problems in the prior art, the present application provides a system for detecting the gender of chicks.

[0006] The application provides a chick gender detection system, which comprises a feeding mechanism, a collecting mechanism, an angle adjusting rod and a detection unit; the chick samples to be detected pass through the feeding mechanism, a first conveying belt and a second conveying belt in sequence; the collecting mechanism comprises a first camera and a first light source arranged in a first dark box, and a second camera, a second light source and a pneumatic nozzle arranged in a second dark box; the first conveying belt passes through the first dark box, the feeding mechanism is horizontally transmitted to the first conveying belt, the first conveying belt is an inclined conveying belt, the second conveying belt is a horizontal conveying belt, and the pneumatic nozzle emits airflow to the chick sample when the chick sample passes through the second dark box through the second conveying belt; the detection unit comprises: an acquisition module, which is used for acquiring a first sequence comprising a plurality of first images of the chick sample passing through the first dark box through the first camera, and acquiring a second sequence comprising a plurality of second images of the chick sample passing through the second dark box through the second camera; a correction module, which is used for continuously acquiring gender detection results of the first sequence and gender detection results of the second sequence of a plurality of test chick samples before detection, and adjusting the inclination angle of the first conveying belt through the angle adjusting rod until the gender detection results of the first sequence and the gender detection results of the second sequence of a plurality of test chick samples with a preset proportion are consistent; and a processing module, which is used for determining the gender detection results of the first sequence according to the proportion of the detected gender in the first sequence of the test chick sample, determining the gender detection results of the second sequence according to the proportion of the detected gender in the second sequence, and determining the gender detection results of the chick to be detected according to the overall proportion of the detected gender in the first sequence and the second sequence of the chick to be detected.

[0007] The chick gender detection system provided by the application further comprises a sorting mechanism, wherein the sorting mechanism comprises a blowing device arranged at the end of the second conveying belt, and the blowing device sorts the chick samples according to the detection results of the detection unit.

[0008] The processing module of the chick gender detection system is specifically used for: inputting each first image into a trained gender detection model to output the gender detection results of each first image; inputting each second image into the trained gender detection model to output the gender detection results of each second image; and determining a comprehensive gender detection result according to the proportion of the total number of the detected female or male in the first image and the second image to the total number of the detected gender images; wherein the gender detection results comprise female and male, and the gender detection model is obtained by training sample images of chicks with labeled gender.

[0009] The application provides a chick gender detection system, wherein the feeding mechanism comprises a feeding module and a straightening module; the feeding module comprises a third conveying belt driven by a motor and baffles on both sides of the third conveying belt; the straightening module comprises a plurality of parallel fourth conveying belts, and a V-shaped shunt plate is arranged between two adjacent fourth conveying belts; and the outlet of the fourth conveying belt is connected with the inlet of the first conveying belt.

[0010] The application provides a chick gender detection system, wherein the width of the straightening module is greater than that of the feeding module.

[0011] The application provides a chick gender detection system, wherein the conveying speed of the fourth conveying belt is greater than that of the third conveying belt.

[0012] The application provides a chick gender detection system, further comprising an infrared sensor arranged at the end of the conveying belt, which is used for detecting the passing result of the chick; when no chick passes within a set time, the power supply of the motor corresponding to the conveying belt is cut off.

[0013] The application provides a chick gender detection system, wherein the interaction module comprises a display screen; the interaction module is used for accepting the image number of the first sequence and the second sequence set by a user and determining the proportion of the detected gender image used for determining the gender detection result of the chick.

[0014] The application provides a chick gender detection system, which is different from the detection method of forcibly unfolding the wings by mechanical means, adopts a behavior induction method based on stress reaction, creates a short weightlessness state for the chick by arranging a conveying belt structure with a certain inclination angle, stimulates the instinctive behavior reaction of unfolding the wings to maintain body balance, creates a short weightlessness state for the chick, stimulates the internal behavior reaction of maintaining body balance, and combines with external airflow stimulation to more comprehensively display the fast and slow feather characteristics of the chick, thereby improving the detection accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are only some embodiments of the present application, and all other drawings obtained by those of ordinary skill in the art without creative effort based on these drawings belong to the protection scope of the present application. Among them, the first, second, etc. mentioned below are only for distinguishing the same type of features, and not for limiting the order or position relationship.

[0016] Figure 1 is a structural schematic diagram of the chick gender detection system provided by the present application;

[0017] Figure 2 is a schematic diagram of the chick gender detection result provided by the present application;

[0018] Figure 3 is a structural schematic diagram of the chick gender detection system including the feeding mechanism provided by the present application;

[0019] Figure 4 is a structural schematic diagram of the feeding module provided by the present application;

[0020] Figure 5 is a structural schematic diagram of the whole column module provided by the present application.

[0021] The drawings are as follows: 10, third rack; 11, third motor; 12, third conveying belt; 13, third baffle; 14, third sensor; 15, fourth motor; 16, fourth baffle; 17, fourth conveying belt; 18, fourth sensor; 19, fourth rack; 20, first motor; 21, first conveying belt; 22, second conveying belt; 23, first dark box; 24, first camera; 25, first light source; 26, second dark box; 27, angle adjusting rod; 28, second motor; 29, air blowing device; 30, fifth baffle. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present application more clear, the technical solutions in the present application will be described clearly and completely below by combining the drawings in the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort belong to the protection scope of the present application. Among them, the first, second, etc. mentioned below are only for distinguishing the same type of features, not for limiting the order or position relationship.

[0023] In view of the problems of low accuracy of automatic detection and easy tissue damage caused by manual operation in current one-day-old chick gender identification, the present application provides a one-day-old chick gender detection system based on full automation. The system comprehensively utilizes the physiological response characteristics and fast-slow feather traits of chicks, combines image recognition technology and deep learning algorithm, and realizes rapid and non-destructive detection of early chick gender such as one-day-old chicks. The system aims to improve detection efficiency and accuracy, reduce dependence on manual operation, and reduce chick stress and damage.

[0024] The following will be described in conjunction with Figures 1-5 The present application describes a chick gender detection system. Figure 1 The present application provides a structural schematic diagram of a chick gender detection system, as Figure 1 The present application provides a chick gender detection system, which comprises a feeding mechanism, a collecting mechanism, an angle adjusting rod and a detection unit. The chick samples to be detected are transmitted through the feeding mechanism, a first conveyor belt 21 and a second conveyor belt 22. The collecting mechanism comprises a first camera 24 and a first light source 25 arranged in a first dark box 23, and a second camera, a second light source and a pneumatic nozzle arranged in a second dark box. The first conveyor belt 21 passes through the first dark box 23, and the feeding mechanism is horizontally transmitted to the first conveyor belt 21. The first conveyor belt 21 is an inclined conveyor belt, and the second conveyor belt 22 is a horizontal conveyor belt. The pneumatic nozzle emits airflow towards the chick sample when the chick sample passes through the second dark box through the second conveyor belt. The detection unit comprises an acquisition module for acquiring a first sequence comprising multiple first images of the chick sample passing through the first dark box through the first camera, and a second sequence comprising multiple second images of the chick sample passing through the second dark box through the second camera. A correction module is used to continuously acquire gender detection results of the first sequence and gender detection results of the second sequence of multiple test chick samples before detection, and to adjust the inclination angle of the first conveyor belt through the angle adjusting rod until the gender detection results of the first sequence and the gender detection results of the second sequence of a preset proportion of the test chick samples are consistent. A processing module is used to determine the gender detection results of the first sequence according to the proportion of detected gender of the test chick samples, to determine the gender detection results of the second sequence according to the proportion of detected gender of the test chick samples, and to determine the gender detection results of the chick to be detected according to the overall proportion of detected gender in the first sequence and the second sequence of the chick to be detected.

[0025] As shown in the figure, the first conveyor belt 21 can be driven by the first motor 20, and the second conveyor belt 22 can be driven by the second motor 28. The system is provided with a section of downwardly inclined slope conveyor belt, i.e. the first conveyor belt 21. On the first conveyor belt 21, the chicks will lose balance temporarily due to the change of slope when passing through the slope from the horizontal feeding mechanism, and instinctively spread their wings due to stress reaction. The first camera 24 located above the conveyor belt and in the dark box will capture the corresponding image in real time. The main control unit obtains multiple images from the video stream captured by the first camera 24 as first images for subsequent gender detection. Subsequently, the chick sample is transported to the second conveyor belt along with the first conveyor belt.

[0026] To avoid external environmental light interference, the first camera 24 and the subsequent second camera are entirely enclosed in the dark box with the corresponding detection area, and industrial light sources (corresponding to the first light source 25 and the second light source) are arranged inside to provide light for the chicks, ensuring stable and reliable imaging quality. Considering that some chicks may not be sensitive to the slope of the first conveyor belt 21 with sudden slope change, resulting in failure to spread or fully spread their wings to reveal the fast and slow feather characteristics and cause misjudgment. The present application particularly provides the second dark box 26 after the inclined first conveyor belt 21 is conveyed to the horizontal second conveyor belt 22, and a gas flow is emitted to the chick sample through the pneumatic nozzle inside the second dark box 26 to stimulate the stress reaction of the chick, so as to spread the wings and reveal the fast and slow feather characteristics. At this moment, a second sequence of multiple second images is captured by the second camera, and the detection of gender characteristics is carried out in combination with the first sequence of first images. Finally, the detection results of the chicks are determined comprehensively according to the detection results containing multiple first images and multiple second images.

[0027] For the chick sample to be detected, the detection unit can select multiple frames from the video stream captured by the first camera, and select the same multiple frames from the video stream captured by the second camera, such as 5 clear images, to form an image sequence containing 10 images (first images and second images). Subsequently, the image sequence will be input into the deployed target detection model to perform gender recognition analysis on the fast and slow feather characteristics revealed by the spread wings of the chick, and the gender is determined according to the recognition ratio, and the final detection result is output. For example, if the sequence is 10 images, and the ratio of male chicks is strictly greater than 50%, the gender detection result is male chick. The detection result corresponding to the image sequence is a detection sequence, as shown in Table 1, and the detection result of the image can be referred to Figure 2 (female represents female chick, and male represents male chick).

[0028] Table 1

[0029]

[0030] In addition, the first conveyor belt 21 is an angle-adjustable conveyor belt connected to the base plate by an angle adjustment rod 27, which controls the slope angle. Before detection, the accuracy of the adjustment angle is determined by a plurality of test chick samples. For example, the gender detection results of the first sequence and the second sequence of 10 test chick samples are continuously obtained, and if the gender detection results of the first sequence and the second sequence of 8 test chick samples are consistent, the angle adjustment is ended, and the gender detection of the chick samples to be detected is performed.

[0031] The chick gender detection system of the present application is different from the existing detection method of forcibly unfolding the wings by mechanical means. It adopts a behavior induction method based on stress reaction. A conveyor belt structure with a certain inclination angle is provided to create a short weightlessness state for the chicks, stimulate their instinctive wing unfolding behavior to maintain body balance. By creating a short weightlessness state for the chicks, stimulating their internal behavior reaction to maintain body balance, and combining with external airflow stimulation, the fast and slow feather characteristics of the chicks can be more fully displayed, thereby improving the detection accuracy. At the same time, by acquiring multiple images to construct a gender detection sequence, and combining with the significant features of the chick wing unfolding behavior in the image for discrimination. Finally, the gender judgment is output according to the proportion of the recognition result, reducing the gender misjudgment caused by motion blur, thereby improving the detection accuracy. For different breeds of chicks, their sensitivity to inclination is different. The present application adjusts the inclination according to the detection result of the second image triggered by the pneumatic nozzle reaction. The inclination angle of the first conveyor belt can be adjusted to a reasonable range to avoid a small part of the chicks losing balance due to insufficient inclination angle stimulation or too large inclination angle, resulting in inaccurate detection results. And using a preset proportion consistent in a plurality of test chick samples as a reference can also filter out the comparative influence of a small part of chicks that are not sensitive to air blowing stimulation, improving the reliability of angle adjustment.

[0032] In some embodiments, a sorting mechanism is further included, which comprises an air blowing device 29 arranged at the end of the second conveyor belt 22. The air blowing device 29 sorts the chick samples according to the detection result of the detection unit, and the sorting mechanism 29 is provided with a fifth baffle 30 at both ends.

[0033] For example, at the end of the second conveyor belt 22, two groups of air blowing devices 29 are installed in each channel, and the airflow is controlled by electromagnetic valves. According to the gender detection result signal transmitted in real time by the detection unit, the corresponding nozzle is triggered to spray airflow, and the chicks are sorted into male and female channels by non-contact pneumatic guidance and transported to the corresponding storage box for sorting. Through this system, the automation of the detection and sorting process during the chick gender detection process is realized, which significantly improves the gender sorting efficiency and reduces the dependence on manual labor and labor intensity.

[0034] In some embodiments, the processing module is specifically configured to: input each first image into the trained gender detection model to output a gender detection result of each first image; input each second image into the trained gender detection model to output a gender detection result of each second image; and determine a comprehensive gender detection result according to a total number of the first images and the second images detected as female or male, and a proportion of the total number of the detected gender images.

[0035] The gender detection result includes female and male, and the gender detection model is obtained by training sample images of chicks spreading out primary and secondary feathers according to labeled genders. For example, 5 first images and 5 second images are selected, and when more than half of the images are detected as male, the gender detection result is male.

[0036] In some embodiments, the feeding mechanism includes a feeding module and a straightening module; the feeding module includes a third conveyor belt 12 driven by a motor and third baffles 13 on both sides of the third conveyor belt 12; and the straightening module includes a plurality of parallel fourth conveyor belts 17, a V-shaped splitter plate between adjacent two fourth conveyor belts 17, and a connection between the outlet of the fourth conveyor belt 17 and the inlet of the first conveyor belt 21.

[0037] As shown in Figures 3-5 The feeding module can adopt a wide conveyor belt structure to achieve efficient batch feeding. It is provided with a third rack 10, and the conveyor belt is driven by a third motor 11 to ensure the stability and controllability of the transmission. Third baffles 13 are installed on both sides to limit the lateral movement of chicks during transportation, prevent chicks from falling due to jumping or sliding, and improve the safety and stability of system operation. The straightening module is provided with a fourth rack 19, and fourth baffles 16 are provided on both sides. A plurality of conveyor belts are designed in parallel, and the width of each conveyor belt is slightly larger than the body width of the chick to ensure that the chicks are arranged in single row and connected and divided by the inverted V-shaped groove guide plate between the conveyor belts, effectively solving the problems of stacking and congestion. The fourth conveyor belt 17 is driven by a fourth motor 15.

[0038] Firstly, the chicks are placed in the feeding module and transported to the fourth conveyor belt 17 of the straightening module through the wide conveyor belt (third conveyor belt 12). The V-shaped splitter plate arranged at the connection between two adjacent fourth conveyor belts 17 divides and straightens the multiple chicks gathered closely into single row, and then enters the narrow conveyor belt.

[0039] In some embodiments, the width of the straightening module is greater than the width of the feeding module. The straightening module is slightly larger than the feeding module, which effectively realizes the transition of chicks from disordered stacking state to single row and orderly arrangement, and facilitates subsequent single gender detection.

[0040] In some embodiments, the fourth conveyor belt 17 has a higher conveying speed than the third conveyor belt 12. In order to achieve effective alignment, the conveying speed of the conveyor belt in the alignment module is higher than that of the feeding conveyor belt, so as to expand the individual spacing after the chicks enter the alignment module, ensure single passage, and facilitate subsequent chicks to enter the collection mechanism for detection.

[0041] The V-shaped diverter plate combined with the differential conveying belt structure guides the disordered chicks to be aligned into a single passage mode, and expands the individual spacing of the chicks through the speed difference between the conveying belts, thereby providing stable single recognition conditions for subsequent gender detection.

[0042] In some embodiments, an infrared sensor is further arranged at the end of the conveying belt to detect the passage result of the chicks. When no chick is detected to pass within a set time, the power supply of the corresponding motor of the conveying belt is cut off.

[0043] For example, the third sensor 14 is arranged at the end of the third conveying belt 12, and the fourth sensor 18 is arranged at the end of the fourth conveying belt 17.

[0044] In some embodiments, an interaction module is further included, which includes a display screen. The interaction module is used to accept the number of images of the first sequence and the second sequence set by the user, and to determine the proportion of the detected gender images of the gender detection result of the chicks. The display screen is used to display the gender detection result.

[0045] The system further includes an interaction module with a human-computer interaction interface, which can monitor the running state of the equipment in real time, support online adjustment of running parameters and fault alarm, and ensure the stability of system operation, the continuity of sorting process and the accuracy of detection result. The running parameters include the number of images of the first image, or the number of images of the first image and the second image, and the proportion of the detected gender in multiple images of the first image, or the proportion of the detected gender in multiple images of the first image and multiple images of the second image.

[0046] In combination with the above embodiments, four groups of comparative experiments are conducted to evaluate the influence of different stimulation methods and image acquisition strategies on the accuracy of gender judgment of one-day-old chicks. The four groups of experiments respectively use ramp stimulation, air blowing stimulation, ramp and air blowing stimulation at the same time, and ramp and air blowing stimulation in sequence but respectively acquiring image sequences for gender judgment. Among them, the first group, the second group and the third group respectively acquire one image for judgment, and the fourth group acquires video streams through ramp and air blowing stimulation in sequence, acquires 5 frames of images for each chick, a total of 10 frames of images, and uses a multi-frame voting mechanism to determine the gender: if the number of frames judged as a certain gender is strictly greater than 5 frames, it is determined as that gender, otherwise it is determined as gender doubtful.

[0047] The image acquisition device selects a U3-3890CP-C-HQ industrial camera produced by Germany IDS company, which is equipped with a CMOS area image sensor, has a resolution of 12 million pixels (4096x3000), and can realize 34 frames / second video acquisition under the maximum resolution, which not only meets the demand of high-quality image acquisition, but also supports multi-frame video acquisition scene. Compared with CCD camera, CMOS camera has the advantages of fast response, low power consumption and high sensitivity, which is suitable for precise capture of dynamic behavior of chicks in the experiment. The camera is equipped with FOCViS UH0820 10MP C lens with a focal length of 8 mm, which can provide high-quality imaging effect under close-range conditions. The image acquisition system is operated through IDS peak Cockpit software to realize flexible adjustment of parameters such as exposure time, frame rate and white balance, and ensure the stability and consistency of image quality under different light conditions.

[0048] In terms of image processing and gender determination, a lightweight deep learning model based on SAG-YOLO is used. The model replaces the original YOLO feature extraction module with StarNet backbone network, which significantly reduces the model parameter quantity and computational complexity. The Additive CGLU module is introduced to optimize the original C2f module, which combines additive similarity function and gate mechanism to effectively enhance the multi-scale feature interaction and detail information capture ability. In addition, a detection head Head based on group normalization (Group Normalization) is constructed, and the shared convolution structure is used to improve the coordination of feature extraction and target detection.

[0049]

[0050] In summary, the system induces the instinctive wing opening behavior of chicks during the short imbalance process by controlling the speed difference of the conveyor belt and the angle of the slope conveying device. Compared with relying on mechanical manual intervention or other forced methods to force the chicks to open their wings passively, it can effectively reduce the physical damage caused by the equipment to individual chicks and improve the accuracy of detection and system stability.

[0051] Finally, the device embodiments described above are only illustrative, wherein the units illustrated as separate components can or can not be physically separated, and the components illustrated as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0052] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0053] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A system for detecting the sex of a chick, characterized in that, The utility model relates to a chick gender detection device and method, including: The feeding mechanism, the collection mechanism, the angle adjusting rod and the detection unit; The chick sample to be detected passes through the feeding mechanism, the first conveying belt and the second conveying belt in turn; The collection mechanism includes a first camera and a first light source arranged in a first dark box, and a second camera, a second light source and a pneumatic nozzle arranged in a second dark box, the first conveying belt passes through the first dark box, the feeding mechanism is horizontally transmitted to the first conveying belt, the first conveying belt is an inclined conveying belt, the second conveying belt is a horizontal conveying belt, the pneumatic nozzle emits airflow to the chick sample when the chick sample passes through the second dark box through the second conveying belt; The detection unit includes: The acquisition module is used for acquiring a first sequence containing multiple first images through the first camera when the chick sample passes through the first dark box, and acquiring a second sequence containing multiple second images through the second camera when the chick sample passes through the second dark box; The correction module is used for detecting the gender detection results of the first sequence and the gender detection results of the second sequence of multiple test chick samples, and adjusting the inclination angle of the first conveying belt through the angle adjusting rod until the gender detection results of the first sequence and the gender detection results of the second sequence of a preset proportion of the multiple test chick samples are consistent; The processing module is used for determining the gender detection results of the first sequence according to the proportion of the detected gender of the test chick sample, determining the gender detection results of the second sequence according to the proportion of the detected gender of the second sequence, and determining the gender detection results of the chick to be detected according to the overall proportion of the detected gender in the first sequence and the second sequence of the chick to be detected; The processing module is specifically used for: Inputting each first image into the trained gender detection model to output the gender detection result of each first image; Inputting each second image into the trained gender detection model to output the gender detection result of each second image; According to the total number of the detected female or male in the first image and the second image as a whole, the proportion of the total number of the detected gender image, the comprehensive gender detection result of the chick to be detected is determined; Wherein, the gender detection result includes female, male, and the gender detection model is obtained after training according to the sample image of the chick with labeled gender.

2. The chick gender detection system according to claim 1, characterized in that, It also includes a sorting mechanism, the sorting mechanism includes a gas blowing device arranged at the end of the second conveying belt, and the gas blowing device blows and sorts the chick sample according to the detection result of the detection unit.

3. The chick gender detection system according to claim 1, wherein The feeding mechanism includes a feeding module and a straightening module; The feeding module includes a third conveying belt driven by a motor and baffles on both sides of the third conveying belt; The straightening module includes multiple parallel fourth conveying belts, and a V-shaped shunt plate between adjacent two fourth conveying belts, and the outlet of the fourth conveying belt is connected with the inlet of the first conveying belt.

4. The chick gender detection system according to claim 3, wherein The width of the straightening module is greater than the width of the feeding module.

5. The chick gender detection system according to claim 3, wherein The conveying speed of the fourth conveying belt is greater than the conveying speed of the third conveying belt.

6. The chick gender detection system according to claim 1, wherein It also includes an infrared sensor arranged at the end of the conveying belt, which is used for detecting the passing result of the chick, and cutting off the power supply of the corresponding motor of the conveying belt when no chick is detected within a set time.

7. The chick gender detection system according to claim 1, wherein Also included is an interaction module, which includes a display screen; The interaction module is used to accept the image number of the first sequence and the second sequence set by the user, and to determine the proportion of the detected gender image of the chick gender detection result; The display screen is used to display the gender detection result.

Citation Information

Patent Citations

  • System and method of determining health & gender of chick

    CN108601336A

  • Facility for automatically determining the sex of a chick

    WO2023281198A1