Hen's egg state determining device

Abstract

The present invention improves the accuracy of determining the state of a hen's egg using a laser speckle image of the hen's egg, and provides a hen's egg state determining device comprising: an image acquiring unit 3 that acquires a laser speckle image of a hen's egg E; and a determining unit 42 that determines the state of the hen's egg on the basis of image information relating to the yolk part in the laser speckle image.

Description

Egg condition determination device

[0001] The present invention relates to an egg condition determination device.

[0002] Conventionally, a hatching egg inspection system for determining the condition of chicken eggs, such as determining the sex of eggs, has been devised, as shown in Patent Document 1. This hatching egg inspection system focuses on the difference between male and female embryos in the growth rate formed in the early stages of incubation, and captures the degree of blood vessel and / or blood formation that accompanies embryo growth from transmittance.

[0003] However, in the method disclosed in Patent Document 1, the growth of the embryo is evaluated using a transmission image, and the growth of the embryo cannot be evaluated until the embryo has grown to a certain size. As a result, it is not possible to determine the sex of the embryo at an earlier stage, and it is also difficult to improve the accuracy of the determination.

[0004] International Publication No. 2018 / 101139

[0005] Meanwhile, the inventors of the present application are considering using laser speckle images to improve the accuracy of sex determination, etc. By using laser speckle images, it is possible to remove noise from dots on the eggshell. Furthermore, due to their nature, laser speckle images can acquire dynamic information such as the heartbeat of the embryo and the pulsation of blood vessels.

[0006] However, even if dynamic information such as the heartbeat of the embryo and the pulsation of the blood vessels can be obtained using a laser speckle image, it is still difficult to evaluate the growth of the embryo or the blood vessels unless they grow to a certain size. In this situation, the inventors of the present application found that changes that occur as the incubation progresses tend to appear in the yolk portion of the laser speckle image.

[0007] The present invention was made as a result of the above-mentioned investigations, and its main objective is to improve the accuracy of determining the state of a chicken egg using a laser speckle image of the egg.

[0008] That is, the egg condition determination device according to the present invention is characterized by comprising an image acquisition unit that acquires a laser speckle image of an egg, and a determination unit that determines the condition of the egg based on image information of the yolk portion in the laser speckle image.

[0009] With this type of egg condition determination device, the egg condition can be determined with high accuracy by determining the egg condition based on changes in the yolk portion in the laser speckle image. Furthermore, by using a laser speckle image, it is possible to use an image in which noise from dots on the eggshell has been removed, allowing the egg condition to be determined with high accuracy.

[0010] Furthermore, in laser speckle images, changes that occur during incubation are likely to appear, particularly in the yolk portion near the embryo or blood vessels. Therefore, it is desirable that the determination unit determine the state of the egg based on image information of the vicinity of the embryo or blood vessels in the yolk portion. With this configuration, the state of the egg can be determined accurately by determining the state of the egg based on changes in the vicinity of the embryo or blood vessels.

[0011] It is desirable that the determination unit determine the state of the chicken egg based on image information of the embryo or blood vessels in the laser speckle image in addition to image information of the yolk portion. With this configuration, the state of the chicken egg can be determined based not only on changes in the yolk portion but also on changes in the vicinity of the embryo or blood vessels, thereby enabling accurate determination of the state of the chicken egg.

[0012] It is desirable that the determination unit determine the state of the egg using a learning model that has been machine-learned to understand the relationship between image information of the yolk portion in the learning image, which is the laser speckle image, and the state of the egg. With this configuration, by using the machine learning model, the state of the egg can be easily determined from the laser speckle image acquired by the image acquisition unit.

[0013] In order to determine the sex of a chicken egg from a laser speckle image, it is desirable that the determination unit determine the sex of the chicken egg as the state of the chicken egg.

[0014] According to the present invention configured in this manner, it is possible to improve the accuracy of determining the state of a chicken egg using a laser speckle image of the egg.

[0015] 1 is a diagram schematically illustrating the configuration of an egg condition determination device according to one embodiment of the present invention; 2 is a graph illustrating the relationship between flow velocity and contrast value in a laser speckle image of the same embodiment; 3 is a laser speckle image of a chicken egg on the third day after the start of incubation, and 4 is a laser speckle image of a chicken egg on the fourth day after the start of incubation, according to the same embodiment; 5 is a graph illustrating the contrast value (average value) of the yolk portion on the "third day after the start of incubation" and the contrast value (average value) of the yolk portion on the "fourth day after the start of incubation" in the laser speckle image of the same embodiment;

[0016] Hereinafter, an embodiment of an egg condition determining device according to the present invention will be described with reference to the drawings.

[0017] <Device Configuration> The egg condition determination device 100 of this embodiment observes the internal condition of a hatching egg, that is, an egg E, during the incubation stage (during incubation) of the egg E. By observing the internal condition of the egg E in this manner, it becomes possible to non-destructively determine the sex of a chick that will hatch from the egg E based on the difference between male and female in the internal condition of the egg E.

[0018] Specifically, as shown in FIG. 1, the egg condition determination device 100 has a light irradiation unit 2 that irradiates light L1 onto an egg E, an imaging unit 3 that captures scattered light L2 generated inside the egg E irradiated with light L1, and an information processing device 4 that determines the internal condition of the egg E from the captured image obtained by the imaging unit 3.

[0019] In this embodiment, the egg E to be inspected is placed on the measurement table 10 with its major axis facing horizontally and then inspected. The egg E to be inspected is, for example, an egg on the third or fourth day after the start of incubation. Note that while Fig. 1 shows an example in which a single egg E is placed on the measurement table 10 and inspected, it is also possible to place multiple eggs E on a setter tray (not shown) having multiple egg seats so that they can be inspected all at once, or it is of course possible to irradiate the eggs E with light during transport and inspect them one after another.

[0020] The light irradiating unit 2 irradiates the egg E with light L1 from the blunt end (air cell side) of the egg, facilitating internal scattering of the light L1 within the egg. The light irradiating unit 2 also irradiates light L1 with a wavelength that is highly permeable through the egg shell and easily absorbed by the embryo, blood vessels, or blood. Wavelengths that are easily absorbed by the embryo, blood vessels, or blood are specifically wavelengths that are easily absorbed by hemoglobin and myoglobin.

[0021] Specifically, the light irradiation unit 2 can use a laser light source 21 that irradiates laser light in the near-infrared wavelength range (for example, 750 nm or more and less than 1000 nm, specifically 785 nm). The light irradiation unit 2 may have an irradiation optical fiber that guides the laser light from the laser light source 21 and irradiates the desired position on the egg. Alternatively, the light irradiation unit 2 may use a laser light source 21 that irradiates laser light in the green wavelength range (for example, 532 nm). The light irradiation unit 2 may also use a light source such as a xenon lamp, which is a type of spectral lamp.

[0022] The imaging unit 3 has a camera 31 that captures scattered light L2 emitted from the egg E from above the egg E. The imaging unit 3 receives the scattered light L2 and generates a laser speckle image. The imaging unit 3 may be provided in front of the camera 31 and may have a bandpass filter for clarifying the shape of the embryo. In this embodiment, the imaging unit 3 functions as an image acquisition unit that acquires a laser speckle image.

[0023] The information processing device 4 judges the state of the egg E based on image information of the yolk portion in the laser speckle image obtained by the imaging unit 3. The information processing device 4 is configured by a dedicated or general-purpose computer having a CPU, internal memory, an input / output interface, an AD conversion unit, etc. The information processing device 4 performs the functions of an image processing unit 41, a judgment unit 42, etc., through cooperation between the CPU and peripheral devices based on the egg measurement program stored in the internal memory.

[0024] The image processing unit 41 calculates image information of the yolk portion in the laser speckle image obtained by the imaging unit 3 from the image. Specifically, the image processing unit 41 calculates, for example, a contrast value as image information of the yolk portion. The image processing unit 41 of this embodiment can calculate image information (for example, contrast values) of multiple locations in the yolk portion and their average value. Here, the image information of multiple locations desirably includes image information of the vicinity of the embryo or blood vessels in the yolk portion.

[0025] In a laser speckle image, as shown in Figure 2, the faster the flow velocity or movement of the imaged object, the smaller the contrast value that appears. Figure 3 shows laser speckle images of a chicken egg E on the third day after the start of incubation and a laser speckle image of a chicken egg E on the fourth day after the start of incubation. Comparing the contrast values ​​at multiple locations in the yolk portion of each of these images, as shown in Figure 4, the contrast value in the yolk portion on the fourth day is smaller than that on the third day. This is thought to reflect the fact that the movement of the molecules that make up the yolk is faster on the fourth day. The laser speckle image shown in Figure 4 was taken of a chicken egg E on the third day after it was cracked, and shows multiple locations (four locations) where contrast values ​​were calculated.

[0026] The determination unit 42 determines the internal state of the egg E based on the image information (e.g., contrast values) obtained by the image processing unit 41. Specifically, the determination unit 42 determines the internal state of the egg E based on image information (e.g., contrast values) from multiple locations obtained by the image processing unit 41 or the average value thereof. Here, the determination unit 42 can determine the sex of the egg E. In other words, the determination unit 42 can determine the state of the egg using a laser speckle image of the egg E on the third or fourth day after the start of incubation.

[0027] More specifically, the determination unit 42 determines the state of the egg using a learning model that has been machine-learned to determine the relationship between the image information of the yolk portion in the learning image, which is a laser speckle image, and the state of the egg. Here, the learning model may be generated using a machine learning function provided in the information processing device 4, or may be generated by a machine learning device separate from the information processing device 4. The learning model of this embodiment is generated by machine learning to determine the relationship between the image information (e.g., contrast values) of multiple locations obtained by the image processing unit 41 or their average value and the sex of the egg.

[0028] <Effects of this embodiment> With the egg condition determination device 100 of this embodiment configured as described above, the egg condition can be determined with high accuracy by determining the egg condition based on changes in the yolk portion in the laser speckle image. Furthermore, by using a laser speckle image, it is possible to use an image in which noise from dots on the eggshell has been removed, allowing the egg condition to be determined with high accuracy.

[0029] Other Embodiments The present invention is not limited to the above-described embodiments.

[0030] For example, the determination unit may determine the state of the chicken egg based on image information of the embryo or blood vessels in the laser speckle image in addition to image information of the yolk portion in the laser speckle image.

[0031] Furthermore, in the above embodiment, the imaging unit 3 functions as an image acquisition unit that acquires a laser speckle image, but the image acquisition unit may be configured to accept laser speckle image data acquired externally.

[0032] Furthermore, if the learning model is generated by a machine learning device other than the information processing device 4, it is possible to configure the learning model to be obtained and used from another machine learning device or server via a communication line such as the Internet.

[0033] Furthermore, the present invention is not limited to the above-described embodiment, and it goes without saying that various modifications are possible without departing from the spirit of the present invention.

[0034] According to the present invention, it is possible to improve the accuracy of determining the state of a chicken egg using a laser speckle image of the egg.

[0035] 100: Egg condition determination device E: Egg 2: Light irradiation unit L1: Light L2: Scattered light 3: Image capture unit (image acquisition unit) 4: Information processing device 41: Image processing unit 42: Determination unit

Claims

1. A chicken egg condition determination device comprising: an image acquisition unit that acquires a laser speckle image of a chicken egg; and a determination unit that determines the condition of the chicken egg based on image information of the yolk portion in the laser speckle image.

2. The chicken egg condition determination device according to claim 1, wherein the determination unit determines the condition of the chicken egg based on image information of the vicinity of the embryo or blood vessels in the yolk portion.

3. The chicken egg condition determination device according to claim 1 or 2, wherein the determination unit determines the condition of the chicken egg based on image information of the embryo or blood vessels in the laser speckle image in addition to image information of the yolk portion.

4. The egg condition determination device according to claim 1 or 2, wherein the determination unit determines the condition of the egg using a learning model that has been machine-learned to determine the relationship between image information of the yolk portion in the learning image, which is the laser speckle image, and the condition of the egg.

5. The egg condition determination device according to claim 1 or 2, wherein the determination unit determines the sex of the egg as the condition of the egg.

Images