JP3047168B2 - Inspection method of chicken eggs - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image of a transmitted light which is condensed on the surface of a shell and scattered inside, and is transmitted again to the outside for use in an egg egg inspection process. The present invention relates to a method and an apparatus for inspecting a chicken egg, wherein defective products are detected by processing a captured image.

[0002]

2. Description of the Related Art In recent years, complaints against defective products such as blood eggs and broken eggs in chicken egg products have become severe. For this reason, strict inspections are performed to eliminate defective products. As an inspection method, when hen eggs are irradiated with light, blood eggs are generally redder or have a reddish mass than normal eggs, and egg breaks are simple, taking advantage of the fact that the broken part looks brighter. In a place where disturbance light is shielded, lighting is performed by fluorescent light or the like from the lower part of the transfer device, and humans make judgments by looking from the top, and the difference between normal eggs and defects is not clear, and quantitative Due to lack of standards, it relies on humans' high discrimination ability and is difficult to use for automation.

[0003]

When light is simply projected onto a chicken egg, the reflected light on the shell surface due to the wraparound of the light is more strongly affected than the transmitted light scattered inside. The intensity difference in the specific wavelength region of the light spectrum between the nondefective egg and the defective egg and the difference in shading with respect to breakage are not clear and cannot be identified. Therefore, it is desired to provide a chicken egg inspection method and a chicken egg inspection device capable of solving these problems.

[0004]

Therefore, according to the first invention of this application, there is provided a method and an apparatus for testing a blood egg of a hen egg, wherein the light projecting means is used for emitting a light, such as a fiber lighting apparatus. By focusing the light so that it focuses on the surface of the egg shell, the light is prevented from wrapping around the shell surface, and the inside blood egg state becomes clear by observing only the transmitted light scattered inside. It is characterized by. Here, FIG. 2 (a) shows the light spectrum of the transmitted light with respect to the normal egg by the light emitting means, and FIG. 2 (b) shows the light spectrum of the transmitted light with respect to the blood egg. Fig. 2 (a) and (b)
Can be clearly confirmed in the wavelength region of about 550 nm to about 650 nm, and the difference in light intensity in this specific wavelength region can be quantified by measuring and digitizing an image captured by the imaging means, Judgment becomes possible.

Further, according to the second invention of this application, there is provided a method and an apparatus for inspecting a broken egg shell, wherein an example of a result obtained by projecting light by the light projecting means and photographing by a television camera is described as a third example. As shown in the figure, it becomes possible to clearly identify the pattern due to the shade of the broken shell, and by performing processing such as the difference of the texture of the surface due to the broken shell and the connected components, distribution, area count, etc. Egg breaking judgment becomes possible. Further, in the egg inspection apparatus according to the present invention, since the light emitting means and the imaging means are not in contact with the eggs, it can be easily incorporated into an automated line. The above is the feature of the present invention.

[0006] A chicken egg test according to the present invention will be described. In an environment where there is no disturbance light, a light emitting means that emits light so as to converge it on the shell surface so that light on the egg surface does not wrap around, and blood eggs and egg breaks by receiving transmitted light due to scattering inside the shell It comprises an imaging unit for determining a defective product by the processing unit and a determination result output unit. In the blood egg test, either the light emitting unit or the imaging unit for measuring the intensity of the wavelength region of the transmitted light. An optical filter that transmits light of about 550 nm to about 650 nm is inserted in one or both, or the processing means has the ability to measure the intensity difference in the above-mentioned wavelength region. However,
Since the light spectrum intensity at a wavelength of 550 nm or less does not depend much on the state of the egg, the lower limit can be set to an appropriate value if the disturbance light is small and the light shielding conditions are good, if the lower limit is about 550 nm or less. In the egg breaking test, the imaging means obtains density information, and has the ability to measure a pattern of a high brightness portion due to a difference in gray value due to a broken shell, an area and a distribution of connected components, and the like.

[0007]

According to the invention of the present application, a clear judgment reference value can be set when a defective egg is judged, and an ambiguous judgment based on a subjective standard for each worker can be eliminated. This enables quality control of the product.

[0008]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a perspective view of a chicken egg inspection method according to an embodiment of the present invention;
However, the drawings used in the description are only schematically shown to the extent that these inventions can be understood. Also, in each of the drawings, the same components are denoted by the same reference numerals, and duplicate description thereof may be omitted.

FIG. 1 is a diagram showing the configuration of a chicken egg inspection apparatus 10 according to the present invention. Egg inspection apparatus 10 of this embodiment
Has light emitting means 12, imaging means 14, and processing means 16, and has a blood egg and egg breaking test means in an information processing means which is a part of the processing means. At this time, the blood egg and the egg breaking test apparatus may be separately provided, or a part thereof may be shared.

Here, the light projecting means 12 is a known light projecting device such as a light projecting device which uses a halogen bulb having a full wavelength range as a light source and irradiates the surface of the chicken egg 20 as an object with an optical fiber or the like so as to be focused. It can be composed of light emitting means.

When only the blood egg test is taken into consideration, the imaging means 14 has an area of about 550 n in which the difference in the received light intensity between the blood egg and the normal egg is clear from the wavelength area emitted by the light emitting means.
Since it is sufficient to selectively capture the range of m to about 650 nm, when light in the entire wavelength range is irradiated as in the light projecting means, for example, for simplicity, an image is taken through a green transmission light filter or the like. Thereby, it can be constituted by a simple television camera or the like. In this embodiment, an image input device is provided which can separate a signal from a color television camera into RGB signals and use the G screen. Of course, it may be configured using a transmission light filter for another wavelength. Further, if the imaging means is limited to an egg breaking test, it can be constituted by a known image input means capable of taking in, as multi-valued information, a grayscale distribution of an image from a television camera, a video camera, or the like. In this embodiment, since it is also used as the light receiving unit for blood egg test, as a color television camera, information indicating the grayscale distribution when it is captured can be obtained as multi-level information of 256 gradations.

The processing means 16 is an information processing means according to the present invention, and constitutes means for acquiring input information from the imaging means and means for outputting judgment result output information, respectively, and realizes these means. Any suitable program incorporating a program for performing For example, when the imaging unit is configured by a television camera, it can be configured by a computer that has an image input unit for capturing an image of the television camera and has a program for realizing the information processing unit. The information processing means that is part of this computer differs from the blood egg test and the egg breaking test.
In the blood egg test, when the input device is a color television camera as in the embodiment, all or a part of the egg portion of the G screen is extracted, and the grayscale information is binarized with an appropriate threshold value. The judgment is made using a frequency distribution or the like. At this time, by performing image improvement by density gradation conversion processing as pre-processing of the binarization processing, processing that is resistant to changes in environmental conditions can be performed. In addition, the egg breaking inspection examines the distribution state of high-brightness broken spots by performing appropriate binarization processing or edge extraction processing on the grayscale information from the TV camera, and also obtains the pixel connection state, The number and size of chunks are determined, and a comprehensive judgment is made based on the information based on the criteria. The setting of the judgment conditions such as the threshold value and the number of pixels to be used as the judgment criterion in the information processing means cannot be clarified because it changes depending on the imaging conditions, the type of eggs, and the like. For this reason, the setting is performed by a method of selecting and incorporating a plurality of types of condition settings in advance, or performing a test at a work site and inputting the settings.

The judgment result output means 18 is a means for outputting a signal to a sorting machine for defective products, and needs to be configured in accordance with the sorting machine. A good / bad condition is represented by turning on / off the contacts by the output of individual contacts.

Next, the inspection procedure will be described. This description will be made with reference to FIGS. 4 to 7 in addition to FIG. 1 referred to above. Here, FIG. 4 is a flowchart showing a processing procedure in the embodiment of the blood egg testing method, and FIG. 5 is a flowchart of a processing procedure in the embodiment of the egg breaking test method.

First, the processing procedure in the embodiment of the blood egg test method will be described. The hen egg is captured by the imaging means 14 into the computer 16 on the G screen of the color image (FIG. 4).
Step S1). Since this image has density information of values 0 to 255, a histogram as shown in FIG. 6 is obtained by calculating a histogram indicating the appearance frequency of the density values. FIG.
6A is a histogram of a normal egg, and FIG. 6B is a histogram of a blood egg. If a sufficient density difference can be obtained by the apparatus, the binarization processing may be performed as it is, but if the density difference cannot be sufficiently obtained, the density gradation conversion processing (step S2 in FIG. 4) is performed. By improving the image and improving the image, stable detection can be performed even when the threshold value of the next binarization process is fixed. In the example, linear conversion was performed from the range of density values 5 to 63 to the range of 0 to 255. The value used here changes depending on the type of the egg, the lighting device, the aperture of the camera lens, and the like. FIG. 6 shows a histogram obtained by the density gradation conversion process. FIG. 6C is a histogram for a normal egg, and FIG. 6D is a histogram for a blood egg image. When the histograms are compared, in FIG. 6 (C), the pixel density with respect to the egg region appears at a density value of about 70 or more, but in FIG. 6 (D), the pixel having a density value of about 60 or more has a density value. Absent. Although the number of pixels having a density value of 70 or more in this histogram may be counted, the binarization process (step S3 in FIG. 4) is performed in the embodiment. When processing is performed with the fixed threshold value of the threshold value 70, pixels in the chicken egg region are extracted from normal eggs, but pixels are not extracted from blood eggs. Since a difference appears in the number of pixels to be extracted depending on the type of chicken egg and the degree of blood egg, the number of pixels is counted (step S4 in FIG. 4), and the number of pixels equal to or less than the reference number of pixels can be determined as blood egg (FIG. 4). Step S5). This value depends on the total number of pixels of the image, the type of chicken eggs, the lighting device, the imaging environment such as the camera, and can be determined by experiments or the like.

Next, an embodiment of the egg breaking inspection method will be described. The egg is taken in by the light receiving means 14 to obtain light and shade information. In the present embodiment, processing is performed after the blood egg test, and (R + G + B) / 3 is used as lightness information as a simple lightness value from the R, G, and B screens captured during the blood egg test. If a sufficient luminance is obtained in (Step S6 in FIG. 5), a specific one screen such as the G screen can be used as it is. If a sufficient density difference cannot be obtained, a density gradation conversion process is performed to improve the image (step S7 in FIG. 5). Step S8) is performed. next,
In the embodiment, in order to detect the break, the break is extracted by the edge detection process (step S9 in FIG. 5).
Detection may be performed by a process of extracting a density difference by binarization processing, texture analysis, or the like.

The edge detection result is binarized to extract a break. If the edge detection result does not have a value sufficient to perform the binarization process, if the edge detection result cannot be obtained, the density gradation is used for clarity. By performing the binarization processing after the conversion, the accuracy of the tear extraction is improved. In the embodiment, density gradation conversion is performed from the density values 0 to 63 to the range of 0 to 255 (step S10 in FIG. 5), and binarization processing is performed with the threshold value 32 (step S11 in FIG. 5). Further, expansion processing and reduction processing are performed for repairing the connected component (step S12 in FIG. 5), and if noise removal processing is desired, processing such as isolated point removal is performed (step S13 in FIG. 5).

Next, in the edge detection processing, since the contour of the egg is also extracted, the contour is removed (step S14 in FIG. 5). For example, an image is prepared in which an image that has been subjected to the binarization processing with a threshold value that separates the chicken egg from the background has been subjected to the degeneration processing a plurality of times corresponding to the contour appearance pixels in the edge processing. At this time, the concentration value of the egg region was set to 255.
By calculating the logical product of this image and the image subjected to the edge processing, an edge-extracted image from which the contour has been removed can be obtained. Binary processing by the p-tile method is effective for separating the egg from the background because the size of the egg can be assumed to some extent. In this embodiment, the number of times of degeneration is five times in consideration of the noise component. Of course, the contour line removal processing is not limited to this example. FIG. 7 shows an example of the processing results so far.

A labeling process is performed on the obtained edge extracted image (step S15 in FIG. 5), and labeling is performed for each connected component. At this time, labels having a certain area (the number of pixels) or less are removed (step S16 in FIG. 5). As a result, the total number of remaining labels and the number of pixels for each label are counted, and, for example, if at least one label having a large area is defective, if there are at least several labels having a certain area or more, it is determined to be defective. By setting conditions, good / bad judgment can be made (step S17 in FIG. 5). Since this criterion differs depending on the measurement environment and the like, it is determined by an experiment at a site or the like. Finally, the determination result is output to the output means.
In this embodiment, after the blood egg test is performed, the egg breaking test is performed. However, the procedure of the processing method is limited, such as changing the order of these tests or performing the test processing in parallel using two computers or the like. Not something.

[0020]

As is apparent from the above description, the present invention enables the automation of the egg egg inspection process conventionally performed by humans, enables quantitative management of products, and stabilizes product quality. Significantly contribute to the maintenance of

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of an embodiment of the first invention and the second invention.

FIG. 2 is a diagram showing an example of optical spectra of a blood egg and a normal egg.

FIG. 3 is a diagram showing an example of egg breaking when light is emitted by the present light emitting means.

FIG. 4 is a flowchart showing a procedure of a blood egg test method according to the embodiment.

FIG. 5 is a flowchart illustrating a procedure of an egg breaking inspection method according to the embodiment.

FIG. 6 is a diagram showing histograms of normal eggs and blood eggs.

FIG. 7 is a diagram showing a state where an edge portion of a crack is extracted by an egg breaking inspection.

[Explanation of symbols]

 10: Egg inspection apparatus of the embodiment 12: Light emitting means 14: Light receiving means 16: Processing means 18: Judgment result output means 20: Target egg

──────────────────────────────────────────────────の Continued on the front page (72) Shinichi Nagao, Inventor Shinichi Nagano, 1-11, Kita 19-Jo Nishi, Kita-ku, Sapporo, Hokkaido (72) Inventor Isao Omura 11 Kita 19-Jo Nishi, Kita-ku, Sapporo, Hokkaido 1-chome, Hokkaido Industrial Testing Center (72) Inventor Teruo Hamada 4-14-17 Mae, Joyo-ku, Toyohira-ku, Sapporo, Hokkaido In-Hokkaido Static Co., Ltd. No. 2-15 Elm Data Co., Ltd. Examiner Keiko Nagai (58) Field surveyed (Int. Cl. ⁷ , DB name) A01K 43/00 G01N 21/27 G01N 21/84 G01N 33/08 G06T 1/00 JICST File (JOIS)

Claims (2)

(57) [Claims] (1) Disturbance in inspecting defective eggs.
Focus light on the egg surface in a light-shielded environment
The light from the surface of the shell using the light emitting means
Is obtained by imaging transmitted light scattered inside.
Optical spectroscopy that clearly shows the difference between blood and normal eggs
In the range of about 550 nm to about 650 nm,
An optical filter that transmits light in the wavelength range
Depending on the device provided on one or both of the imaging side
And perform measurement processing to quantify the difference
Detects blood, blood clots, and blood eggs mixed with meat fragments
A method for inspecting chicken eggs, characterized in that: 2. In the inspection of defective eggs, there is a disturbance.
Focus light on the egg surface in a light-shielded environment
Transmission of Eggs Imaged by Light Emitting Means
For images, based on grayscale image information indicating brightness, etc.
Extract bright spots caused by shell tearing
Measure the area and distribution state of connected components of the pattern and shell
Method of inspecting chicken eggs characterized by detecting broken egg breaking
Law.