JP2004101204A - Method and apparatus for inspecting fertilized egg - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reliably and nondestructively determine whether a fertilized egg is alive or not and the state of its development in such a way as to be close to inspectors' criteria of judgement. <P>SOLUTION: In this inspection apparatus, the inside of the fertilized egg is irradiated with light to pick up a color image of the inside of the egg. Then blood vessel information such as the presence or absence, thickness, distributed state of blood vessels and internal color information are measured within an inspection region. Or concentration distribution information in the vicinity of the boundary of an air cell is measured by measuring concentration changes from the air cell to an embryo on the basis of concentration distribution information in the vicinity of the boundary of the air cell. On the basis of this, the determination of a normal egg etc. is performed. In addition, in the inspection apparatus, inspection is performed preferably after spot patterns are removed from an egg shell on the basis of color information of the L*a*b* color system. The direction of the image pickup of the color image of the inside of the egg is changed by 90° at image pickup. The fertilized egg is arranged in a light shielding structure, and its image is picked up by a color CCD camera. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
[Technical field to which industry belongs]
The present invention relates to a nondestructive test method and apparatus for determining the life and death of a sperm egg and the state of development. More specifically, the present invention is characterized in that a normal egg, a dead egg, and a poorly-developed egg are selected by examining the state of development based on the internal color information of the fertilized egg. The present invention relates to a destructive inspection method and a nondestructive inspection device for realizing the present method.
[0002]
[Prior art]
In the process of producing an influenza vaccine, fertilized eggs (chicken eggs) are used to propagate the virus. A few days after the influenza virus is injected into a fertilized egg, the virus is collected only from well-developed eggs. This growth state is examined by a skilled inspector irradiating each egg with light and visually observing the state of internal blood vessels and the presence or absence of bleeding. This visual observation is performed in a dark room, and it is necessary to process tens of thousands to hundreds of thousands of eggs a day. It has been pointed out that there is a problem that individual differences may appear.
[0003]
Numerous patent applications have been filed for egg (infertile) inspection methods, but the presence or absence of bleeding or foreign matter depending on the degree of absorption of light of a specific wavelength inside the egg, or the density of the transmitted light image indicates the eggshell. It detects cracks and cannot detect the distribution or development of blood vessels in a fertilized egg.
For example, in a fertilized egg discriminating device (Japanese Patent Application Laid-Open No. 9-127096), irradiation light transmitted and scattered in an egg is detected by a plurality of light receiving elements, and a predetermined frequency component corresponding to a heart pulse is extracted from the detection signal. Then, the presence or absence of a fertilized egg, the active state of the fertilized egg, and the viable / dead state of the fertilized egg are determined based on the presence or absence of the amplitude change and the amount of fluctuation. However, it is difficult to recognize the active state of a fertilized egg in detail based on heart rate information based on the amount of irradiation light emitted from the inside. In fact, a skilled inspector at the work site actually performs the determination work by visually checking not only the state of the internal heart but also the state of the blood vessels.
[0004]
[Problems to be solved by the invention]
An object of the present invention is to determine the life and death of a fertilized egg and its growth state in a non-destructive manner, closer to the judgment standard of the inspector, and reliably.
An object of the present invention is to provide an inspection method and an inspection apparatus capable of judging the life and death of a fertilized egg and the growth state thereof in a detailed and reliable manner as a technical method closer to the work of a skilled inspector.
[0005]
[Means for Solving the Problems]
In the inspection of the growth state, by checking the distribution state of blood vessels and the presence or absence of bleeding, there is a feature that the development state of the fertilized egg can be recognized in more detail, and the present invention determines a normal egg based on the recognized state It is an inspection method. That is, the present invention irradiates light inside the fertilized egg to capture a color image of the inside of the egg, measures blood vessel information in the examination region, and determines a normal egg based on the information. The outline of the method is egg inspection.
[0006]
The present invention, in a color image of the inside of the egg imaged by irradiating light from the top of the fertilized egg, the presence or absence of blood vessels, thickness, distribution state and other information, by using image processing, the results are measured This is a non-destructive inspection method for a fertilized egg that automatically determines a normal egg based on the method. That is, the presence / absence, thickness, distribution state, etc. of blood vessels in a captured color image of the inside of the egg are grasped, and a dead egg or a poorly-developed egg is automatically detected based on the existence of the blood vessel. By irradiating the inside of the egg with light and capturing a color image of the inside of the egg, measuring the blood vessel information in the inspection area such as the presence or absence of blood vessels, thickness, distribution state, etc. The gist of the present invention is a method for testing a fertilized egg, which is characterized in that a normal egg is determined by detecting and distinguishing the egg from a normal egg.
[0007]
Further, the present invention is characterized in that a light image is illuminated inside the fertilized egg to capture a color image of the inside of the egg, the internal color information in the test area is measured, and a normal egg is determined based on the information. The gist is a method of testing sperm eggs.
[0008]
The present invention provides a method for measuring a concentration change from an air chamber to a fetus based on concentration distribution information near an air chamber boundary in a color image of the inside of an egg obtained by irradiating light to the inside of a fertilized egg. This is a non-destructive inspection method for fertilized eggs that determines eggs and automatically detects the presence or absence of bleeding near the air chamber boundary. That is, the density distribution information near the air chamber boundary in the captured color image inside the egg is grasped, and based on the information, the presence or absence of a dead egg or the presence of bleeding near the air chamber boundary is automatically detected. Detecting a fertilized egg by irradiating light inside the fertilized egg, capturing a color image of the inside of the egg, measuring concentration distribution information near the air chamber boundary, and determining a normal egg based on the information The law is a gist.
[0009]
The above-described method for inspecting a fertilized egg of the present invention is preferably carried out after removing the spot-like pattern of the eggshell based on the color information of the L * a * b * color system. That is, based on the color information of the L * a * b * color system of the fertilized egg, a spot-like pattern on the eggshell surface that becomes a noise when detecting a blood vessel is removed before the inspection. Based on the color information near the chorioallantoic space from the air chamber to the fetus, the possibility of bleeding in the chorioallantoic space is automatically detected, and if there is a possibility of bleeding, blood vessels and other blood bleeding are detected. This method is a non-destructive inspection method for a fertilized egg in which image correction is performed before a blood vessel test because there is a case where it is difficult to recognize a blood vessel state because of a small color difference from a part.
[0010]
Further, the above-mentioned method for testing a fertilized egg of the present invention, when irradiating light inside the fertilized egg to capture a color image of the inside of the egg, changes the imaging direction to a plurality of color images inside the fertilized egg. Images can be taken and based on them. This is a non-destructive inspection method of a fertilized egg which prevents a leak of a blood vessel by a position and automatically determines a normal egg in a color image of a plurality of fertilized eggs imaged by changing an imaging direction by 90 degrees. .
[0011]
Furthermore, the method for testing a fertilized egg of the present invention, when irradiating light inside the fertilized egg to capture a color image of the inside of the egg, placing the fertilized egg in a structure having a light-shielding property, Images can be taken with a color CCD camera.
According to the present invention, in order to realize the above solution (method), a fertilized egg is arranged in a structure having a light-shielding property, and the above-described inspection method is implemented based on an image captured by a color CCD camera. It relates to an inspection device. That is, the present invention is a non-destructive inspection device for realizing any of the above-mentioned methods for inspecting a fertilized egg, wherein the inspection foil for arranging the fertilized egg, means for irradiating light inside the fertilized egg And a fertilized egg inspection apparatus having a non-destructive inspection unit for determining a normal egg using the captured egg image, and an image capturing unit including a color CCD camera for capturing a color image of the inside of the egg, determination as necessary The gist of the present invention is a fertilized egg inspection device provided with a sorting conveyor for sorting eggs to be inspected according to the results of the above.
[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
《Inspection egg》
Eggs to be tested in the present invention are fertilized eggs such as chickens, and the color of the egg surface may be anything such as white or brown.Especially, in the case of fertilized chicken eggs, the technical significance of the present invention is large. Become. The type of virus such as influenza and the type of injections such as other chemicals, and whether or not they are injected, are not limited. In the case of a destructive inspection, the technical significance of the present invention is great.
[0013]
《Structure of inspection egg》
FIG. 1 is a diagram for explaining the general structure of a normal fertilized egg that is about a dozen days old. For example, the outer structure of a developing chicken egg is covered with a shell called an egg shell. Immediately inside it is the eggshell membrane, and the eggshell and the eggshell membrane are used to exchange oxygen with the interior. There are thick blood vessels just inside the eggshell.
For example, the internal structure of a developing chicken egg has an air chamber, which is a layer of air, at the tip of the egg, and a fetus wrapped in amniotic membrane containing amniotic fluid at the center. Between the fetus and the air chamber is the chorioallantoic space wrapped in the chorioallantoic membrane. In the case of a fertilized egg that is about 10 days old and grows normally, the chorioallantoic space has a certain size, and blood vessels are widely distributed in the chorioallantoic membrane.
[0014]
《Normal egg》
FIG. 2 is a schematic diagram of a captured image for explaining typical characteristics of a normal egg that is about ten days old. In the case of a normal growing sperm egg which is about ten days old, the chorioallantoic space has a certain size, and the fetus exists near the center of the egg. Blood vessels are widely distributed in the chorioallantoic membrane. The shade of the air chamber and chorioallantoic space is distinctly different. No bleeding is observed in the eggshell membrane and in the chorioallantoic membrane and chorioallantoic space. Above the egg, there is an air chamber of a certain size.
[0015]
《Abnormal egg》
(1) Dead egg
FIG. 3 is a schematic view of a picked-up image for explaining typical characteristics of dead eggs aged ten and several days. Normally growing sperm eggs at a dozen days of age have a fixed size of chorioallantoic space, the fetus is located in the center of the egg, and blood vessels are widely distributed on the chorioallantoic membrane However, in the case of dead eggs, there is no distribution of blood vessels, and when blood vessels are present, the distribution is extremely narrow and small. The blood vessels that are distributed are extremely thin and pale. Color differences near the border of the air chamber and chorioallantoic space may be unclear. Bleeding may be seen in the eggshell membranes and on the chorioallantoic membrane and in the chorioallantoic space.
(2) Stunted eggs
FIG. 4 is a schematic view of a picked-up image for explaining typical characteristics of a poorly-developed egg at the age of ten and several days. In the case of a normally growing sperm egg of about a dozen days old, the chorioallantoic space is of a certain size, the fetus is located in the center of the egg, and blood vessels are widely distributed on the chorioallantoic membrane, In the case of poorly developed eggs, their distribution is extremely narrow and small. The blood vessels that are distributed are extremely thin and pale. Poorly developed eggs with abnormal blood vessel conditions may be treated as dead eggs.
(3) Bleeding egg
FIG. 5 is a schematic view of a captured image illustrating typical characteristics of an allantoic hemorrhagic egg at a dozen days of age. Eggs with bleeding inside the chorioallantoic membrane and chorioallantoic space. For example, a test egg in which hemorrhage is observed inside the chorioallantoic space is treated as a normal egg when the fetus is growing normally. In eggs with hemorrhage inside the chorioallantoic space, the color of blood vessels and chorioallantoic fluid tends to be the same color, and it may be difficult to examine the blood vessels distributed in the chorioallantoic membrane. FIG. 6 is a schematic diagram of a captured image illustrating typical characteristics of a bleeding egg near the air chamber boundary at the age of more than ten days. If there is bleeding in the inside of the eggshell, in the chorioallantoic membrane, and in the chorioallantoic space near the boundary between the air chamber and chorioallantoic space, it is treated as a dead egg.
[0016]
《Color information inside the egg》
The color information inside the egg is obtained by irradiating the illumination light from the upper part of the egg (air chamber side), capturing the illumination light transmitted through the eggshell using a color CCD camera, and obtaining a color image (RGB color system).
On a color image obtained by imaging the inside of a normal fertilized egg of about a dozen days old, the air chamber, fetus, chorioallantoic space, and blood vessels have different characteristics such as color, brightness, and shape. The fetal part is dark and has little color. The air chamber, chorioallantoic space, and blood vessels have many colors ranging from yellow to red, and have different lightness.
Using the area inside the egg from the air chamber to the fetus with the chorioallantoic space as the examination area, obtain the position of the fetus (the size of the area), concentration information near the air chamber boundary, and blood vessel information, and determine the normal egg. I do.
Since the inspection area contains a lot of color information from red to yellow, the inspection area is specified using the G component and the R component of the RGB image.
The area of the specified test area is measured and used as a reference for determining a normal egg.
[0017]
<< L * a * b * color system information inside the egg >>
The color image is converted from the RGB color system to the L * a * b * color system, a part of the a * component is extracted, and a monochrome image having its strength as a density value is created. The blood vessel part on the color image contains more reddish components than other parts. Since the a * component in the L * a * b * color system is a component whose hue ranges from red to green, using the a * component is effective in extracting blood vessel information.
FIG. 7 is a schematic view of a picked-up image for explaining typical characteristics of a spotted egg at a dozen days of age. The spots on the egg surface have a similar shade to the other areas, but the brightness values may be significantly higher than the other areas. When a monochrome image is created from an RGB image, the image is easily affected by the luminance value, and spots become noise when extracting blood vessel information. In the L * a * b * color system, since the color is expressed by combining the hue information a * and b * with the luminance information L *, when the a * component is converted to a monochrome image, the luminance value can be ignored and the luminance can be increased. There is an advantage of suppressing a speckle. Since it is difficult to completely remove the speckles due to the difference in the hue between the speckles and the other parts only by the conversion of the color system, a process for removing the remaining speckles such as smoothing from the image is added. This spot removal method has a great effect when handling original color information inside an egg.
FIG. 5 is a schematic view of a captured image illustrating typical characteristics of an allantoic hemorrhagic egg at a dozen days of age.
When bleeding is observed in the allantois cavity, since the color of the blood in the allantois membrane and other parts are similar due to the bleeding, the color in the examination area is constant, and the a * component image has low density and low density. Contrast is likely to occur, and blood vessel extraction becomes difficult. The possibility of bleeding inside the chorioallantoic space is determined based on the density information of the examination region. If it is determined that there is bleeding, the brightness of the image is changed before the blood vessel extraction processing to enhance the contrast.
[0018]
《Concentration information around the air chamber inside the egg》
Comparing a fertile egg that has grown normally over a dozen days of age with a poorly developed egg or a dead egg, the concentration of the normal egg rapidly decreases from the air chamber to the fetus. In eggs and dead eggs, the attenuation of the concentration value from the air chamber often changes gradually. From the RGB color image, a monochrome image with a brightness value as a density value is created, and the amount of change in the density value from the vicinity of the boundary between the air chamber and the inside of the eggshell membrane to the fetus is measured, and used as a judgment material for determining a normal egg.
FIG. 6 is a schematic diagram of a captured image illustrating typical characteristics of a bleeding egg near the air chamber boundary at the age of more than ten days.
When there is bleeding near the air chamber boundary on a monochrome image created from an RGB color image, the density value near the air chamber boundary is often lower than the density value of the chorioallantoic space. A test area for confirming the presence or absence of bleeding is provided near the air chamber boundary, and the density information in that area is compared with the density information in other parts where the test area is shifted at a fixed interval from the fetus near the air chamber boundary. The ratio is determined, and the presence or absence of bleeding is determined based on the magnitude. (FIG. 14)
[0019]
《Vascular information inside the egg》
Since there is a difference in the state and distribution of blood vessels in the test area between normal eggs, poorly developed eggs, and dead eggs, it is effective to grasp the state and distribution of blood vessels in determining a normal egg. Using the above-described speckle removal processing image, only a reddish portion is extracted as a blood vessel portion. After the extraction of the blood vessels, the length, thickness, and distribution range are measured, and used as judgment data for determining a normal egg. This method has an effect of extracting a blood vessel portion with high accuracy.
[0020]
《Inspection equipment》
In the apparatus for enabling nondestructive inspection of fertilized eggs according to the present invention, the supplied eggs are subjected to nondestructive inspection by image processing to determine normal eggs and to discriminate between normal eggs and dead eggs (including poorly developed eggs). I do. In the non-destructive inspection part, a fertilized egg is arranged in a structure having a light-shielding property, and an inspection is performed based on an image taken by a color CCD camera. In the inspection device, a color image of the inside of the egg is taken from four directions by rotating the foil support on the egg placed on the inspection foil by 90 degrees. Normal egg determination is performed using the obtained egg images from four directions. Sorted on a sorting conveyor according to the result of normal egg judgment.
[0021]
[Action]
Based on the internal color information of the fertilized egg, a poorly developed egg, a dead egg, an egg having bleeding near the air chamber boundary, and a normal egg can be reliably and non-destructively determined by color image processing. In the case of a fertilized egg, it is possible to perform a nondestructive test for determining a normal egg regardless of the color of the egg surface or the presence or absence of virus injection. Since the blood vessel information inside the fertilized egg is measured in a non-destructive manner, it can be determined according to the state of development of the egg.
Skilled inspectors visually check the condition of internal blood vessels and the presence or absence of bleeding to judge a fertilized egg with a good growth state. Have been watched. On the other hand, in the above-mentioned conventional technology, whether or not a fertilized egg is used, or whether the fertilized egg is active or alive or dead, is determined based on the heartbeat information of the egg. I can't say. The present invention uses an image processing method to automatically recognize the blood vessel distribution and the state of bleeding in order to make a judgment closer to the judgment criteria of the inspector, so that the life and death of the fertilized egg and its development state are not destroyed. Can be reliably determined.
In addition, the present invention adopts an image processing method as a technical method closer to the work of a skilled inspector, and automatically recognizes a blood vessel distribution state in an egg centering on the chorioallantoic space, which is a place for virus propagation, and as a result, It is possible to provide an inspection device capable of determining the life and death of a fertilized egg and the state of development of the fertilized egg in a detailed and reliable manner.
[0022]
【Example】
Examples of the present invention will be described in detail. The present invention is not limited by these examples.
[0023]
Example 1
8 to 11 are schematic diagrams showing a schematic configuration of an embodiment of an inspection apparatus for realizing the inspection method of the present invention.
In the inspection device, a large number of eggs conveyed by the dedicated tray are transferred one by one to an inspection foil, a plurality of eggs are arranged on an alignment conveyor, and a plurality of eggs are conveyed to the image pickup unit at once by picking and pressing.
The transported eggs are placed in a light-shielding shooting cylinder in order to raise the foil support and prevent disturbance during image processing.
At the top of the photographing tube, there is a tube with a built-in lighting device containing a lighting device. The foil support is raised until the tip of the tube contacts the egg, and the egg is illuminated from the air chamber side.
The distal end portion of the illumination built-in tube is made of a soft material so that the illumination light does not leak outside from the distal end even if the egg shape or size is different.
Therefore, reflection of the illumination light on the egg surface can be prevented, and the illumination light passing only inside the egg is captured by the color CCD camera to obtain color information inside the egg.
The color CCD camera is installed on the opposite side of the imaging tube, and captures the first egg image at the rising end of the support. After the photographing, the support base is rotated by 90 degrees in response to the photographing completion signal, and the second frame is photographed after stopping. Similarly, the third and fourth sheets are photographed, images of one inspection egg are obtained from four directions, and a normal egg determination process is performed using these images.
When the support table is lowered by the fourth image capturing completion signal, and the eggs are then transported from the alignment conveyor, the captured eggs are pushed out from the image capturing unit and received by the sorting conveyor.
After the image is captured and before it is transferred to the sorting conveyor, the eggs are subjected to normal egg determination processing, and a determination result signal is output to the sorting conveyor.
[0024]
FIG. 12 is a process flowchart of the inspection method (normal egg determination) of the present invention.
In the inspection method of the present invention, the presence or absence of bleeding near the air chamber boundary is confirmed using color images (RGB) of the inspection eggs captured from four directions by the inspection device.
FIG. 13 is a processing flowchart of the technique for testing for bleeding near the air chamber boundary according to the present invention.
The obtained color image (RGB) is converted into a monochrome image based on the luminance information.
FIG. 20 is a diagram for briefly explaining a method of setting a test area for the presence or absence of bleeding near the air chamber boundary.
The monochrome image emphasizes low density and suppresses noise having a high luminance value, selects a pixel that satisfies the condition shown in FIG. 20, and specifies the pixel as a test area for checking the presence or absence of bleeding near the air chamber boundary.
The specified examination region is further shifted to the fetus at a constant interval, and three comparative examination regions are set from the vicinity of the air chamber boundary to the fetus.
FIG. 14 shows an example of setting a bleeding inspection area.
For each of the four specific inspection areas, an average density value is calculated from the image after noise suppression.
Calculate each gray-scale ratio between the gray-scale average value closest to the vicinity of the air chamber boundary and the gray-scale average values of the three regions from the fetus, and determine the maximum value as the air-chamber boundary and blood vessel distribution region (chorioallantoic space) of the test egg Specified as the shading ratio of
[0025]
FIG. 15 shows a simple example of the average gray value in each bleeding confirmation test area of a normal egg and an egg having bleeding near the air chamber boundary.
When there is bleeding near the air chamber boundary, there is a large difference in shading between the vicinity of the air chamber boundary and the blood vessel distribution region (chorioallantoic space).
When the contrast ratio is equal to or more than the set value, the determination result is output as bleeding near the air chamber boundary.
As a result of the test, when it is determined that there is bleeding near the air chamber boundary, the test target egg is determined as a dead egg at that time.
As a result of the examination, when it is determined that there is no bleeding near the air chamber boundary, the change in concentration near the air chamber is measured.
Based on the projection result of the density information on the X axis using the image after noise suppression, the range on the X axis where the air chamber exists is specified.
[0026]
FIG. 16 shows a simple example of projection of density information on the X axis of a normal egg and a dead egg.
The concentration of normal eggs decreases from the air chamber to the fetus, but the gradient often steeps at the border of the air chamber. May decrease gradually in a gentle gradient.
The leftmost range in which the density information on the X axis continuously exceeds a certain set value by a specified number or more is specified as the range of the air chamber.
The rightmost end of the air chamber range is defined as the air chamber boundary point, the average of the projected value attenuation rate at regular intervals is measured from the air chamber boundary for a set number, and the average value of the attenuation rates of all the measurement results is used as a judgment material for the normal egg judgment. Used for judgment.
Next, an examination area necessary for acquiring blood vessel information is specified using the color information inside the egg.
[0027]
R and G in the color image (RGB) are converted into monochrome images by respective components, and are binarized by a set threshold value.
When the R image is binarized at a set value or more, the whole egg centering on the air chamber and the blood vessel distribution region (chorioallantoic cavity) is extracted from the surrounding dark part.
When the G image is binarized at a set value or less, an image is extracted from which a high-brightness blood vessel distribution region that is likely to occur in an air chamber or a dead egg is removed.
In the two obtained binary images, a region existing in both is specified as an inspection region.
[0028]
The above-described method has an effect that only a blood vessel distribution region having a normal hue can be specified as an inspection region.
The total number of pixels of the specified inspection region is measured as an area, and is used at the time of primary determination as a material for determining a normal egg.
The color information of the inspection area portion is extracted from the captured image.
As the pre-processing of blood vessel extraction, spots scattered on the extracted inspection area image are removed by using the spot removal technique of the present invention.
FIG. 17 shows the flow of the speckle removal method.
Since the speckles have a higher luminance value than other parts, the L * a * b * color system is used to suppress the influence.
After converting the inspection area image into the L * a * b * color system, a monochrome image is created by extracting the a * component.
The a * image is inverted to emphasize low density values in order to remove remaining high density spots. After that, smoothing for smoothing shading is performed as noise removal.
After removing the spots, the presence or absence of bleeding in the examination area (in the chorioallantoic space) is confirmed.
FIG. 18 shows a flow of the bleeding confirmation processing.
The density information of the inspection area after the removal of speckles is measured, and the presence or absence of bleeding is determined.
In a normal state, the blood vessel portion is dark and the other portions are brighter than the blood vessel in the image after the speckle removal. However, when bleeding is observed inside the chorioallantoic space, the blood vessel and other parts have the same color, so that the density value is low overall and the contrast width is narrow.
The egg image determined to be a bleeding egg is increased in brightness and enhanced in contrast to clarify the difference in shading between the blood vessel and other parts.
After removing speckles, confirming bleeding in the inspection area, and taking countermeasures, blood vessel extraction is performed.
[0029]
FIG. 19 is a diagram for briefly explaining a blood vessel extraction method.
In the blood vessel extraction, a blood vessel candidate area is extracted using the method of FIG. 19, noise removal and thinning are performed, and a blood vessel is extracted as a line segment having a width of one pixel.
For the extracted blood vessels, the average value of the hue of a * and G is measured for each blood vessel using the a * image and the G image.
An extremely thin blood vessel is determined and removed by comparing each color tone average value with a set threshold value.
The remaining extracted blood vessels are used as normal blood vessels of the test egg, and the total blood vessel length (total number of pixels) is calculated. Further, the number of distribution blocks of blood vessels on the inspection area divided into a plurality of blocks is measured.
In the primary determination, a normal egg is determined based on a balance between the total blood vessel length of the four images and the total area of the inspection region.
The secondary determination is performed only on the test egg determined to be dead at the time of the primary determination. Here, a normal egg is determined by comparing the amount of change in density near the air chamber (density decay rate) of the four images and the number of blood vessel distribution blocks with a set threshold value. Eggs with a sharp decrease in concentration and those with a large number of blood vessel distribution blocks are regarded as normal eggs.
For a dead egg, the determination result is determined when detecting bleeding near the air chamber boundary and at the time of the secondary determination, and for a normal egg, the determination result is determined at the time of the primary determination and at the time of the secondary determination.
The present invention provides a primary inspection after processing such as inspection for presence or absence of bleeding near the air chamber boundary, measurement of concentration change near the air chamber, identification of an inspection area / area measurement, removal of spots, confirmation / measures of bleeding in a blood vessel inspection area, extraction of blood vessels, and the like. By performing the secondary two-stage normal egg determination, it is effective for highly accurate inspection of fertilized eggs.
The result of the normal egg determination is output from the image processing device to the sorting conveyor, and sorted according to the result.
A plurality of eggs that have been judged as normal eggs are simultaneously adsorbed from the sorting line, transferred from the foil to a dedicated tray, and discharged. A plurality of the selected dead eggs (including poorly-grown eggs and bleeding eggs near the air chamber border) are simultaneously adsorbed, transferred from the foil to a dead egg collection facility, and collected.
[0030]
【The invention's effect】
Based on the internal color information of the fertilized egg, its development status can be inspected to determine non-destructively the normal and dead eggs (including those with poor growth and bleeding around the air chamber border). it can.
[Brief description of the drawings]
FIG. 1 is a schematic diagram illustrating the structure of a fertilized egg when it has grown normally for about ten days.
FIG. 2 is a schematic view of a picked-up image for explaining typical characteristics of a normal egg of about a dozen days old.
FIG. 3 is a schematic view of a picked-up image for explaining typical characteristics of dead eggs aged ten and several days.
FIG. 4 is a schematic view of a picked-up image for explaining typical characteristics of a poorly-developed egg of about a dozen days old.
FIG. 5 is a schematic view of a captured image for explaining typical characteristics of an allantoic hemorrhagic egg of a dozen days old.
FIG. 6 is a schematic view of a picked-up image for explaining typical characteristics of a bleeding egg near the air chamber boundary at the age of ten and several days.
FIG. 7 is a schematic view of a picked-up image for explaining typical characteristics of a spotted egg of about a dozen days old.
FIG. 8 is a schematic diagram showing a schematic configuration of an embodiment of a non-destructive inspection apparatus for a fertilized egg of the present invention.
FIG. 9 is a schematic elevational view 1 of the image pickup section of FIG. 8;
FIG. 10 is a schematic elevational view 2 of the image pickup section of FIG. 8;
FIG. 11 is a schematic plan view of the image pickup unit of FIG. 8;
FIG. 12 is a flowchart of a determination of a normal egg of a sperm egg by image processing in the inspection device of FIG. 8;
FIG. 13 is a flowchart of a bleeding test method near the air chamber boundary based on density information around the air chamber inside the egg in the test apparatus of FIG. 8;
14 is a schematic diagram showing an example of setting a bleeding test area in the bleeding test method near the air chamber boundary in FIG. A to D represent areas, and a to d represent average values of the gray levels in the area.
FIG. 15 is a schematic diagram for explaining an example of a difference in the average gray value measured for each test area of a normal egg a) and a bleeding egg b) near the air chamber boundary in the method for testing bleeding near the air chamber boundary in FIG. . a1 to d1 represent the average density of the normal eggs, and a2 to d2 represent the in-area average values of the test areas of the bleeding eggs near the air chamber boundary.
FIG. 16 is a schematic diagram illustrating a difference in the concentration decay rate between a normal egg and a dead egg by an example of a maximum density value projection on the X-axis of the normal egg a) and the dead egg b) in the air chamber vicinity concentration change measurement method. It is.
FIG. 17 is a flowchart of a speckle removal method based on L * a * b * color system information inside an egg in the inspection apparatus of FIG. 8;
FIG. 18 is a flowchart of a method for determining and taking measures for the presence or absence of bleeding in the chorioallantoic space in the test apparatus of FIG.
FIG. 19 is a diagram briefly describing conditions for extracting blood vessels in the inspection apparatus of FIG. 8;
FIG. 20 is a diagram briefly describing conditions for extracting a bleeding confirmation test region near the air chamber boundary in the test device of FIG. 8;

Claims (9)

A method for inspecting a fertilized egg, which comprises irradiating light inside the fertilized egg to capture a color image of the inside of the egg, measuring blood vessel information in the inspection area, and determining a normal egg based on the information. 2. The method for testing a fertilized egg according to claim 1, wherein the presence / absence, thickness, distribution state, and the like of blood vessels in the captured color image of the inside of the egg are ascertained, and a dead egg or a poorly-grown egg is automatically detected based on the information. A method for inspecting a fertilized egg, which comprises irradiating light inside the fertilized egg to capture a color image of the inside of the egg, measuring internal color information in the inspection area, and determining a normal egg based on the information. 4. The method for inspecting a fertilized egg according to claim 3, wherein the density distribution information in the vicinity of the air chamber boundary in the captured color image of the inside of the egg is grasped, and the presence or absence of a dead egg or bleeding near the air chamber boundary is automatically detected based on the information. Inspection method for any of the fertilized eggs according to any one of 1 to 4, which is performed after removing the spot-like pattern of the eggshell based on the color information of the L * a * b * color system. 6. The method according to claim 1, wherein, when irradiating light inside the fertilized egg to capture a color image of the inside of the egg, a plurality of color images of the inside of the fertilized egg are captured by changing the imaging direction and performed based on the images. Inspection method of the sperm egg. 7. The method according to claim 1, further comprising arranging the fertilized egg in a light-shielding structure and irradiating the inside of the fertilized egg with light to capture a color image of the inside of the egg. Test method for sperm. A nondestructive inspection device for realizing the method for inspecting a fertilized egg according to any one of claims 1 to 7, wherein the inspection foil for arranging the fertilized egg, means for irradiating light inside the fertilized egg, and the egg An inspection apparatus for a fertilized egg, comprising: an image capturing section having a color CCD camera for capturing an internal color image; and a non-destructive inspection section for determining a normal egg using the captured egg image. 9. The inspection apparatus for fertilized eggs according to claim 8, further comprising a sorting conveyor for sorting eggs to be inspected according to a result of the determination.

Images