JP2021106549A - Information processing method, system and program for sexing chick, and method for sexing chick - Google Patents

Abstract

To provide an information processing method for making it possible to sex a chick accurately even without special skills.SOLUTION: Image data of a chick irradiated with ultraviolet by an illuminator is obtained by capturing the chick by an optical device for visible light (S101). The image data presents an image in which at least flight feather parts are excited by a stroke shift and are colored differently (for example, pink) from ultraviolet, thereby facilitating discrimination between flight feathers and other feathers with different colors. The image data undergoes image processing and a tone change (S102), and an image obtained for sexing is provided for a neural network to sex the chick therein (S103, S104). This processing is performed by a system, thereby eliminating the need for special skills.SELECTED DRAWING: Figure 6

Description

The present invention relates to an information processing method, a system and a program for sexing chicken chicks, and a method for sexing chicken chicks.

Chicken chicks that develop on the incubation farm are confirmed to be male and female by visual inspection of humans, for example, within 1 to 2 days of development, and female chicks are shipped for egg collection. There are three known visual inspection methods: the wing feather discrimination method (also called the "feather discrimination method"), the anal discrimination method, and the color discrimination method (also called the "feather color discrimination method"). A discrimination method suitable for the chicken species is used.

For example, the wing wing discrimination method is a method of discriminating between males and females based on the length of wing wings. Females of primary chicks have long wings, and males of primary chicks have short wings, which are used for sexing of primary chicks of such chicken species. In addition, some males have wings that extend and become longer. It may be difficult to distinguish between males and females simply by the length, but since the lengths of the main wing and aileron are different in females, the tip shape of the wing wings is not uniform and the shape is jagged. On the other hand, in males, the lengths of the main wing and the aileron are the same, so the tip shapes of the wing tend to be the same. Males and females are discriminated by taking into account the length of these wings and the shape of the tip.

The color discrimination method is based on the pattern of the body. For example, in the case of a chicken breed in which two lines appear for females and a single line appears for males, the sex is discriminated from the number of lines. This color discrimination method is applied to a limited variety of Akatama egg-laying chicken breeds.

The anus discrimination method is a method in which the anus of a primary chick is opened, and if a full-scale white protrusion of a poppy is observed from the outside, it is judged to be a male. It is mainly used for sex discrimination for chicken species to which the color discrimination method and the wing feather discrimination method cannot be applied, for example, chickens close to the original species.

The color discrimination method described above makes it easy to distinguish the difference in feather color and can discriminate even without specialized skills. On the other hand, the wing feather discrimination method and the anal discrimination method are performed by a technician who has specialized skills by a first-time chick sexing teacher. In any case, humans perform sex discrimination. Then, at least when performing the wing feather discrimination method or the anal discrimination method, the discriminator picks up the primary chicks one by one and spreads the wings (wing feather discrimination method) or opens the anus (anal discrimination method). Check the condition of the primary chicks and determine the sex.

In recent years, breeds that have been applied to wing feather discrimination are mainly used for egg-collecting chickens that are widely distributed, and it is necessary to discriminate by a discriminator and a technician with specialized skills. As a system for enabling sex discrimination even by a person without specialized skills, for example, there is a system for determining the sex of a chick disclosed in Patent Document 1. The system is based on a mobile platform that transports chicks, a stimulus that is directed at the chicks to open the wings of the chicks that are moving on that mobile platform, a camera that captures the chicks, and male and female based on images taken by the cameras. It is equipped with a computer or the like for determining.

The stimulating means blows air upward from the tilted conveyor forming a part of the mobile platform or from the legs of the chicks on the tilted conveyor toward the face and head of the chicks, as described in, for example, paragraph [0037]. It consists of a fan and a configuration that vibrates the surface of an inclined conveyor. The stimulus given to the chicks by this means of stimulation, that is, the upward blowing air and the vibrating surface, makes the chicks feel unbalanced, and in order to regain that balance, the chicks begin to lift their wings and begin to flap. ..

The camera then takes an image of the open wing pattern. Then, as described in paragraph [0038], the images taken by the camera are communicated with the computer processor, which processes the images and combines them with the images of the male and female wings in the database of chicks of that breed. compare. By comparing the images, the computer determines the sex of the chick.

Special Table 2018-533367

Since the device disclosed in Patent Document 1 takes a picture with a camera with its wings spread, it takes time and effort to spread it. Further, since the chicks are stimulated by the stimulating means in order to spread the wings, it causes stress on the primary chicks, which is not preferable. In addition, the stimulated chick does not always spread its wings as expected, and may not be able to photograph the wings in the desired posture, orientation, and state. If the wings are not spread correctly, the tip shape of the wings will be shaped. It is not possible to take a clear picture, which causes the discrimination accuracy to drop. Furthermore, since the stimulating means is installed, the size of the device becomes large, and high-speed processing cannot be performed because the processing for expanding the wings is required.

The above-mentioned problems are described as independent of each other, and the present invention does not necessarily have to be able to solve all of the described problems, as long as at least one problem can be solved. In addition, we have the intention to acquire the rights to the structure for solving this problem independently by divisional application, amendment, etc.

(1) In the information processing method according to the present invention, image data obtained by irradiating chicken chicks with ultraviolet rays by an illuminating device and photographing the chicks including a portion excited by Stokes shift from ultraviolet light with an optical device for visible light. The process of acquiring the teacher data image based on the acquired image data, the process of storing the teacher data image based on the acquired image data in the storage means, and the plurality of the teacher data images stored in the storage means of the female chick teacher data image and the male chick teacher. It was classified into data images and trained by a neural network to generate a sex judgment model.

(2) A process of irradiating chicks of unknown sex with ultraviolet rays and acquiring image data obtained by photographing the chicks including a portion excited by Stokes shift from ultraviolet light with an optical device for visible light. It is preferable to give an image for sex determination based on the obtained image data to the neural network, and perform the sex determination using the sex determination model generated by executing claim 1.

(3) The ultraviolet rays may be 365 nm or 395 nm.

(4) The system according to the present invention includes a lighting device that irradiates chicken chicks with ultraviolet rays, an optical device that photographs the chicks irradiated with the light in a visible light region, and measurement data measured by the optical device. It is preferable to include a processing device having a function of acquiring and executing the information processing method according to any one of claims 1 to 3, and a device that executes processing according to the processing result of the processing device. In the embodiment, the device corresponds to a transport sorting device 17, an output device 15 for notifying a determination result, a storage device 16 for storing and holding generated images and models, and the like.

(5) The device may be a sorting device having a function of sorting male chicks and female chicks based on the judgment result when the processing result is a judgment result of male and female.

(6) The program according to the present invention is a program for realizing the function of the processing device according to (4) or (5) in a computer.

(7) The method for distinguishing males and females of chicken chicks according to the present invention is a process of irradiating chicken chicks with ultraviolet rays by a lighting device, and using an optical device for visible light to irradiate the chicken chicks with ultraviolet rays and a portion excited by Stokes shift. When the process of shooting and the image for sex determination based on the captured image data are given to the neural network, and the process of determining the sex is performed using the sex determination model generated by executing (1). good.

When the chicks are exposed to ultraviolet rays, parts such as wings are stroke-shifted from ultraviolet light due to the excitation phenomenon, and different visible light colors appear. In images taken with visible light, for example, the wings and feathers are distinguished. It will be easier to do. Therefore, it is possible to create a teacher data image for creating a sex determination model and a sex determination model suitable for performing sex determination with high accuracy. As a result, even a user who does not have any particular knowledge / skill can appropriately determine the sex. In addition, since the chicken chicks can be treated without spreading their wings, the stress applied to the chicken chicks can be suppressed as much as possible when each treatment is performed.

FIG. 1 is a diagram showing a preferred embodiment of a system for sexing chicken chicks according to the present invention (for example, generation of a sex determination model). 2A and 2B are examples of image data taken by an optical device, in which FIG. 2A is an image of a female chick and FIG. 2B is an image of a male chick. FIG. 3 is a flowchart showing the function of the processing device. FIG. 4 is a flowchart showing the function of the processing device. FIG. 5 is a diagram showing a preferred embodiment of a system (for example, sex determination) for sexing chicken chicks according to the present invention. FIG. 6 is a flowchart showing the function of the processing device.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. The present invention is not construed as being limited to this, and various changes, modifications, and improvements can be made based on the knowledge of those skilled in the art without departing from the scope of the present invention.

FIG. 1 is a diagram showing a preferred embodiment of a system for creating a sex determination model, which is an example of a system for sex discrimination of primary chicks, which are chicken chicks according to the present invention. This system 10 performs predetermined processing on the lighting device 11 that irradiates the primary chick 1 with a predetermined light, the optical device 12 that captures the primary chick 1, and the image captured by the optical device 12, and determines the sex. A processing device 13 for generating a model, an input device 14, an output device 15, a storage device 16, and the like connected to the processing device 13 are provided.

The lighting device 11 uses a light source that irradiates ultraviolet rays of UV-A (320 to 400 nm) such as an ultraviolet lamp. Various lighting devices can be used as long as the wavelength is within the range of UV-A, but an ultraviolet lamp of 365 nm or 395 nm is preferable, and a lighting device 11 of 395 nm is particularly preferable. Further, when a commercially available ultraviolet lamp is used as the lighting device 11, even if most of the ultraviolet lamps have a wavelength of, for example, 365 nm or 395 nm, ultraviolet rays of only the wavelength are not emitted. , It has a certain spread with the wavelength as its peak, and one of the bases reaches the visible light region castle.

The optical device 12 uses a digital camera having a C-MOS image sensor or a CCD image sensor. This digital camera captures a visible light region (400 to 1000 nm), and captures a color image of the primary chick 1 irradiated with ultraviolet rays. Since the optical device 12 captures the visible light region in this way, it is not possible to capture the ultraviolet light in the ultraviolet region reflected by the primary chick 1.

The present inventor has discovered that when the primary chick 1 is irradiated with ultraviolet rays, an excitation phenomenon occurs, and the light in the ultraviolet region appears as a predetermined color in the visible light region due to the stroke shift. Furthermore, it was found that the portion where the stroke shift occurs is particularly the portion of the wing blade 2 (see FIG. 2). The color due to this excitation becomes a thing corresponding to the wavelength of the ultraviolet rays to be irradiated, for example, it becomes brown at 365 nm and pink at 395 nm. Since these brown and pink colors have wavelengths that are not included in the light source, it can be seen that they are colors based on the stroke shift due to excitation. Ultraviolet rays of other wavelengths also cause stroke shifts in their respective unique colors, but the above-mentioned brown and pink colors are good because the contrast with the feathers 3 is vivid and clear. In particular, pink is more preferable because it can be distinguished from the eyes and the like, and it is easy for people to see at the time of confirmation.

As a result, when an image is taken with an optical device 12 that photographs normal visible light in a state of being irradiated with ultraviolet rays, the portion of the wing feather 2 is reflected in a predetermined excited color, and the color of the portion of the wing feather 2 and the portion of the feather 3 are reflected. The image is such that the difference between the two is clearly distinguishable.

The processing device 13 is a personal computer or other computer, and has a function of executing an implemented application program or the like. The input device 14 is a device for inputting information and commands to a processing device 13 such as a keyboard, a mouse, and a touch panel. The output device 15 is, for example, a monitor device, and displays, for example, image data taken by the optical device 12 and the result (generated image or the like) of the processing device 13 performing predetermined processing on the image data. It is a device. The function or the like mainly executed by the processing device 13 is the function or the like that executes the information processing method.

[Teacher data image creation function]
The processing device 13 has a function of executing, for example, the flowchart shown in FIG. This function creates a teacher data image used for learning to determine male and female by AI. All or part of this function may be realized by, for example, a software program, or by ASIC or dedicated hardware.

Specifically, the processing device 13 acquires image data captured by the optical device 12 (S101). Next, the processing device 13 performs image processing on the acquired image data, adjusts the white balance, and changes the color tone (S102). Specifically, a process for suppressing the blue component is performed. As described above, many commercially available ultraviolet lamps constituting the lighting device 11 also have a wavelength component in the visible light region, and therefore, when photographed with a normal camera, the light component in the visible light region close to the ultraviolet region is captured. As a result, the image becomes bluish as a whole. In the present embodiment, since the blue component is suppressed by image processing, the image after image processing becomes an image that is easy for humans to see.

By executing the image processing described above, the teacher data image C is generated. Then, the processing device 13 stores the generated teacher data image C in the storage device 16 (S103).

The above-mentioned processing is executed for different primary chicks, and a predetermined number or more of male teacher data images and female teacher data images are generated and stored in the storage device 16. Then, these teacher data images may be stored in the storage device 16 in association with the chicken species information.

In the present embodiment, since the image processing for color tone conversion is performed to suppress the blue component, the teacher data image after the image processing becomes an image that is easy for humans to see. Etc. can be easily performed. Further, the acquired image data may be used as it is as a teacher data image without performing this image processing (S102). However, it is more preferable to perform image processing as in the present embodiment because the accuracy of sex determination is higher when the image is color-converted and used.

[Male / female judgment model generation function]
The processing device 13 has a function of executing, for example, the flowchart shown in FIG. This function generates a sex determination model for determining sex by AI based on the teacher data image created by the above-mentioned function.

Specifically, the processing device 13 accesses the storage device 16 and acquires, for example, a teacher data image created for the same chicken breed (S201). Then, the acquired teacher data image is artificially classified into male and female (S202). In this classification operation, for example, a predetermined number of teacher data images acquired by the processing device 13 are displayed on the display device which is the output device 15. Then, the model creation user looks at the image displayed on the monitor, determines whether it is a male image or a female image, operates an input device, for example, and labels the male / female on each teacher data image. This labeling can be performed by, for example, adding male / female information to each teacher data image, dividing the teacher data image into a male holder and a female holder and storing them in the corresponding holders, and various other methods.

Next, the processing device 13 gives the female teacher data image (C-1) and the male teacher data image (C-2) to the neural network, and learns by deep running (S203). The neural network generates a sex determination model for determining sex based on the given female teacher data image and the male teacher data image (S204). The processing device 13 stores the generated male / female determination model in the storage device 16 in association with the chicken breed. This model creation work can also be a method of creating on an external computer having a neural network.

FIG. 5 is a diagram showing an embodiment of a system for discriminating between male and female primary chicks according to the present invention. The system 10'for discriminating between male and female primary chicks is a predetermined lighting device 11 for irradiating the primary chick 1 with a predetermined light, an optical device 12 for photographing the primary chick 1, and an image captured by the optical device 12. A processing device 13 that performs the above-mentioned processing and determines the sex, an input device 14, an output device 15, a storage device 16, a transport sorting device 17, and the like connected to the processing device 13 are provided. As the lighting device 11 and the optical device 12, for example, those having the same functions as those in the system 10 described above may be used. The storage device 16 stores and retains the sex determination model for each chicken species created by the system 10 described above. Further, the storage device 16 may store, for example, the image data acquired by the optical device 12, the result of the processing device 13 performing predetermined processing on the image data (generated image, etc.), and the like. Further, it is particularly preferable that the lighting device 11 used in the system 10'is the same as the lighting device 11 used in the system 10 for generating the teacher data image / sex determination model.

The processing device 13 is a computer or the like having an arithmetic processing ability having a neural network and having a function of executing an implemented application program or the like. Specific functions will be described later. The output device 15 is, for example, a display device, a lamp, a speaker, or the like, and notifies the user (product user) of the determination result of male and female performed by the processing device 13 by display display or light color and sound.

The transport sorting device 17 includes a transport device such as a belt conveyor that transports the primary chicks 1, a device that picks up the primary chicks 1 after determination by a robot arm or the like and sorts them into males and females. Further, the sorting work may not be automatically performed by a machine like a robot arm, but may be manually performed while checking the notification result of the output device 15.

FIG. 6 is a flowchart illustrating the function of the processing apparatus 13 for determining the sex of the primary chicks in the incubation farm or the like. First, in the present embodiment, sex determination is sequentially performed on the primary chicks of the same chicken species. Therefore, prior to the start of operation, the input device 14 is used to specify the chicken species to be determined, and the chicken species to be determined are transmitted to the processing device 13. If the input device 14 for transmitting the chicken species is, for example, a selection switch, a button, or the like provided attached to the processing device 13, it may be possible to give an instruction with a simple operation. In addition, various methods can be adopted in addition to instructing the chicken species while looking at the selection screen displayed on the monitor of the output device 15 by using pointing of a keyboard, a mouse, or the like.

In the sex determination work, the primary chicks 1 to be determined are sequentially supplied onto the belt conveyor of the transport sorting device 17 by hand or by using a predetermined device. The primary chick 1 is conveyed by a belt conveyor without being able to spread its wings in the placed state, and reaches the photographing area of the optical device 12 which is irradiated with a predetermined light by the lighting device 11. Preferably, the belt conveyor may be temporarily stopped when it reaches this photographing area, but it may be continuously conveyed depending on the capacity of the optical device 12 or the like.

The optical device 12 sends the image data captured and acquired to the processing device 13 in real time. The processing device 13 acquires the image data sent from the optical device 12 (S101). Next, the processing device 13 performs image processing for adjusting the white balance on the acquired image data (S102), generates an image (C') to be discriminated between males and females, and stores it in the storage device 16 (S103). The series of processes from S101 to S103 may be the same as the corresponding processes in the above-mentioned "teacher data image creation function".

Further, when the teacher data image used for generating the sex determination model is not performed, if the image processing for adjusting the white balance is not performed, the input image data is input without performing the image processing in the processing step S102 in this apparatus as well. The image (C') to be discriminated between males and females is determined as it is, and is stored in the storage device 16 or given to the neural network.

Further, the processing device 13 gives the generated image (C') to the neural network (D) in which the sex determination model (E) corresponding to the chicken species is registered (S104). This neural network (D) uses the sex determination model in the processing step S204 of the “male / female determination model creation function”.

The neural network determines whether the given image (C') is a male primary chick or a female primary chick (S104). The processing device 13 controls the operations of the output device 15 and the transport sorting device 17 based on the determination result. That is, the determination result is notified to the user (product user) by the display display of the output device 15 or the color and sound of the light, and the female and the male are sorted by the transport sorting device 17. That is, the primary chicks judged to be male and female are picked up and transferred to the male area or the female area, respectively.

As described above, in the present embodiment, the primary chicks 1 are simply placed on the belt conveyor of the transport sorting device 17, and the posture and appearance of the primary chicks 1 are not particularly adjusted, for example, by spreading the wings. You can make a judgment. Therefore, it is good that the primary chicks can be discriminated without stress.

Further, although only one optical device 12 is drawn in the drawing, it is preferable to arrange a plurality of cameras at different positions and photograph the primary chicks from different directions. In this way, if the AI / neural network is configured to shoot from various angles with multiple cameras and the AI / neural network determines based on the multiple images taken, the placement, posture, and orientation of the primary chicks can be arbitrarily improved. , It is good because the sex can be determined more quickly and with high accuracy. In the determination based on a plurality of cameras, for example, if the measurement data taken by any of the cameras can be determined to be a female, the primary chick to be determined may be a female. Further, for a plurality of cameras, for example, it is advisable to place the cameras at six surrounding locations.

On the other hand, when one camera is installed, it is preferable that the primary chicks are placed sideways with respect to the camera, that is, the wings and wings are bag-made by a person or various guide mechanisms.

Further, in the present embodiment, for example, in order to continuously and sequentially determine and sort the primary chicks sent in large quantities, a transport sorting device 17 having a transport function such as a conveyor belt is used. However, the present invention is not limited to this, and for example, the primary chicks may be manually placed at a predetermined position and sorted manually based on the sex determination result.

Further, in the above-described embodiment, the male / female determination model or the like is stored in the storage device 16 and the male / female determination model is used to determine the sex at the site such as in the incubation farm, but the present invention is limited to this. In this case, for example, the captured image may be sent to a server and the determination may be made there. However, if an image is sent each time, it may not be possible to smoothly determine a large number of primary chicks, so it is preferable to enable on-site determination as in the present embodiment. In this case, for example, it is advisable to set a limit on the number of males and females that can be used, and when the limit is reached, a business model in which additional males and females can be determined by purchasing a new key.

Further, since the optical device captures ordinary visible light, for example, a mobile terminal such as a smartphone or a tablet or other various devices having a camera function can be used. In this way, the system can be configured inexpensively and easily. Then, the algorithm for determining the sex using the neural network may be installed as an application, for example, or the captured image may be sent to the server and determined on the server side. However, in order to perform high-speed processing as described above, it is advisable to discriminate by an application installed on a smartphone or the like.

In addition, when irradiated with ultraviolet rays, patterns appear sparsely in places other than the wings. Then, as for the pattern that appears, a pattern that does not appear in a normal color image taken by irradiating visible light appears, so by using it in combination with a color discrimination method based on that pattern or wing feather discrimination, the anus It can be applied to chicken breeds that could only be identified.

Although various aspects of the present invention have been described above with reference to embodiments and modifications, these embodiments and descriptions have not been made for the purpose of limiting the scope of the present invention, and are intended for understanding the present invention. It should be added that it was provided to contribute. The scope of the present invention is not limited to the configurations and manufacturing methods explicitly described herein, but also includes combinations of various aspects of the invention disclosed herein. Of the present invention, the configuration for which a patent is sought is specified in the appended claims, but even if the configuration is not currently specified in the claims, it is disclosed in the present specification. I would like to mention that there is a possibility of claiming a patent for this configuration in the future.

1: Primary chick 10: System 10': System 11: Lighting device 12: Optical device 13: Processing device 14: Input device 15: Output device 16: Storage device 17: Transport sorting device

Claims (7)

A process of irradiating chicken chicks with ultraviolet rays by an illuminating device and acquiring image data obtained by photographing the chicks including a portion excited by Stokes shift from ultraviolet light with an optical device for visible light.
A process of storing a teacher data image based on the acquired image data in a storage means,
Information processing that classifies the plurality of teacher data images stored in the storage means into teacher data images of female chicks and teacher data images of male chicks and trains them with a neural network to generate a sex determination model. Method. A process of irradiating chicken chicks of unknown sex with ultraviolet rays and acquiring image data obtained by photographing the chicks including a portion excited by Stokes shift from ultraviolet light with an optical device for visible light.
An information processing method in which an image for sex determination based on the acquired image data is given to a neural network, and sex determination is performed using the sex determination model generated by executing claim 1. The information processing method according to claim 1 or 2, wherein the ultraviolet rays are 365 nm or 395 nm. A lighting device that irradiates chicken chicks with ultraviolet rays,
An optical device that photographs the chicks irradiated with the light in the visible light region,
A processing device having a function of acquiring measurement data measured by the optical device and executing the information processing method according to any one of claims 1 to 3.
A system including a device that executes processing according to the processing result of the processing device. The system according to claim 4, wherein the device is a sorting device having a function of sorting male chicks and female chicks based on the judgment result when the processing result is a judgment result of male and female. A program for realizing the function of the processing apparatus according to claim 4 or 5 in a computer. Processing to irradiate chicken chicks with ultraviolet rays by a lighting device,
A process of photographing the chicks including a portion excited by a Stokes shift from ultraviolet light with an optical device for visible light.
An image for sex determination based on captured image data is given to a neural network, and a process of determining sex is performed using the sex determination model generated by executing claim 1. How to distinguish male and female chicks.

Images