JP2019533140A - Segregation analyzer and method - Google Patents

Abstract

A segregation analyzer according to one technical aspect of the present invention is an image acquisition unit that acquires a captured image of a mold test piece including a plurality of test pieces, and extracts a plurality of test piece images for the plurality of test pieces from the captured image, respectively. And a segregation information generation unit that detects a segregation region from each of the plurality of test piece images and quantifies the segregation region to generate segregation analysis information.

Description

  The present invention relates to a segregation analysis apparatus and method that can automatically determine the center segregation grade of a wire product using an image.

  The wire is a linear steel and can be used in various fields such as tire cords and construction. Since such wire is generally required to have higher strength than ordinary steel, there is a high demand for quality control.

  When center part segregation exists in the inside of such a wire, there is a very high possibility that disconnection or the like is caused. Therefore, various techniques for detecting segregation of such a wire have been developed.

  In the conventional segregation detection technique for a wire, whether or not segregation exists in the center of the wire is determined based on an image of a cross section of the wire. On the other hand, in the case of such a conventional technique, there is a problem that when the captured image is changed due to illumination, camera performance, or the like, the detection accuracy of segregation is lowered.

  Such prior art can be easily understood by referring to Korean Published Patent No. 2009-0046920, Korean Published Patent No. 2010-0078590, Korean Published Patent No. 2012-0068635, etc. Can do.

  The present invention is for solving the above-mentioned problems of the prior art, and one technical aspect of the present invention is to detect segregation for each of a plurality of test pieces and provide segregation analysis information for the segregation. Thus, it is possible to provide a segregation analysis apparatus and method capable of automating test piece inspection and quality control and providing accurate analysis.

  One technical aspect of the present invention proposes a segregation analyzer. The segregation analyzer includes an image acquisition unit that acquires a captured image for a mold test piece including a plurality of test pieces, a test piece image extraction unit that extracts a plurality of test piece images for the plurality of test pieces from the captured image, respectively. And a segregation information generation unit that detects a segregation area from each of the plurality of test piece images and quantifies the segregation area to generate segregation analysis information.

  In one embodiment, the image acquisition unit may include a ring illumination having a hollow corresponding to a shape of a camera and a lens of the camera.

  In one embodiment, the test piece image extraction unit includes a matching unit that matches a correlation coefficient between a template image and the captured image, a binarizer that performs binarization on an output from the matching unit, and the above An image extractor that extracts the plurality of test piece images from the captured image based on an output from the binarizer may be included.

  In one embodiment, the segregation information generation unit sets a peripheral region of a central region in a test piece image as a reference region, and variable binary that variably binarizes the central region using the average brightness of the reference region. And a segregation area determiner that determines, in the output from the variable binarizer, the area where the pixel value histogram of the reference area that has been variably binarized is equal to or less than a preset threshold value as the segregation area. be able to.

  In one embodiment, the segregation information generation unit calculates at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation region to generate the segregation analysis information. A generator can further be included.

  Another technical aspect of the present invention proposes a segregation analysis method. The segregation analysis method includes obtaining a captured image for a mold test piece including a plurality of test pieces, extracting a plurality of test piece images for the plurality of test pieces from the captured image, and a plurality of test piece images. Detecting a segregation region from each of them, and quantifying the segregation region to generate segregation analysis information.

  In one embodiment, acquiring the captured image may include irradiating the plurality of specimens with light using a ring illumination having a hollow corresponding to the shape of the lens of the camera.

  In one embodiment, the step of extracting the plurality of test piece images includes a step of matching a correlation coefficient between the template image and the captured image, a step of binarizing the result of matching the correlation coefficient, And extracting the plurality of test piece images from the captured image based on the binarized result.

  In one embodiment, the step of generating the segregation analysis information includes setting a peripheral region of the center region as a reference region in the test piece image, and variable binarizing the center region using the average brightness of the reference region. And a step of determining, as the segregation region, a region where the pixel value histogram of the reference region that has been binarized is equal to or less than a preset threshold value with respect to the result of variable binarization.

  In one embodiment, the step of generating the segregation analysis information includes calculating the segregation analysis information by calculating at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation region. Generating may be included.

  The means for solving the problems described above do not enumerate all the features of the present invention. Various means for solving the problems of the present invention can be understood in more detail by referring to the specific embodiments in the following detailed description.

  According to one embodiment of the present invention, segregation is detected for each of a plurality of test pieces, and the segregation analysis information for the segregation is provided, thereby automating the inspection and quality control of the test pieces and performing accurate analysis. There is an effect that it can be provided.

  According to an embodiment of the present invention, since the presence / absence of segregation is determined using an average value of brightness based on variable binarization, it is possible to accurately detect segregation even if illumination or the like changes. Play.

It is a block block diagram explaining the segregation analyzer by one Embodiment of this invention. FIG. 2 is a reference diagram illustrating an embodiment of an image acquisition unit illustrated in FIG. 1. It is a block block diagram explaining one Embodiment of the test piece image extraction part shown by FIG. FIG. 4 is a reference diagram for explaining how a test piece image is extracted by a test piece image extraction unit shown in FIG. 3. FIG. 3 is a reference diagram for explaining how a captured image is corrected by a test piece image extraction unit shown in FIG. 1. It is a block block diagram explaining one Embodiment of the segregation information generation part shown by FIG. It is a reference figure explaining a mode that a segregation is extracted by the segregation information generation part shown by FIG. It is a graph for demonstrating the segregation determination performed by the segregation information generation part shown by FIG. It is a reference graph for demonstrating the variable binarization by the variable binarizer shown by FIG. It is a reference figure which shows the result of having performed the fixed value binarization with respect to the test piece image which has mutually different brightness. It is a reference figure which shows the result of having performed variable binarization with respect to the test piece image which has mutually different brightness. It is a flowchart explaining the segregation analysis method by one Embodiment of this invention.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

  However, the embodiments of the present invention can be modified in various other forms, and the scope of the present invention is not limited to the embodiments described below. Also, the embodiments of the present invention are provided to more fully explain the present invention to those having average knowledge in the art.

  FIG. 1 is a block diagram illustrating a segregation analyzer according to an embodiment of the present invention.

  Referring to FIG. 1, the segregation analyzer 100 may include an image acquisition unit 110, a test piece image extraction unit 120, and a segregation information generation unit 130.

  The image acquisition unit 110 can acquire a captured image for a mold test piece including a plurality of test pieces.

  The test strip image extraction unit 120 can extract a plurality of test strip images for a plurality of test strips from the captured image.

  The test strip image extraction unit 120 can classify the extracted plurality of test strip images by assigning numbers.

  When the captured image is rotated, the test piece image extraction unit 120 can detect the image and correct it in the forward direction.

  The segregation information generation unit 130 can detect a segregation area from each of a plurality of test piece images, and can generate segregation analysis information by quantifying the segregation area.

  The segregation information generation unit 130 can detect the segregation area from the center area of the test piece image. For example, the segregation information generation unit 130 can perform variable binarization using the average brightness of the peripheral region of the central region, and can detect the segregation region from the central region of the test piece image.

  Hereinafter, various embodiments of the segregation analyzer 100 will be described in more detail with reference to FIGS.

  FIG. 2 is a reference diagram for explaining an embodiment of the image acquisition unit shown in FIG.

  Referring to FIG. 2, the image acquisition unit 110 may include a camera 210 and a lighting device 220 including a plurality of light source elements. In the illustrated example, the illumination device 220 is shown as a ring illumination device, but this is exemplary. Therefore, according to the embodiment, the lighting device can be modified in various forms including a plurality of light source elements.

  The camera 210 includes a lens 211, and can capture an image by photographing the mold test piece 10 including a plurality of test pieces.

  In one embodiment, the camera 210 can comprise a telecentric lens. Therefore, it is possible to eliminate the segregation size distortion that occurs in normal camera photography.

  The illumination device 220 may be a ring-shaped illumination device including a plurality of light source elements, and may include a hollow corresponding to the shape of the lens 211 of the camera. Therefore, as in the illustrated example, when the illumination device is located, a plurality of light source elements are arranged around the camera, so that the mold test piece can be irradiated with light uniformly. Therefore, the change in brightness can be minimized, and the generation of noise due to the unevenness on the surface of the test piece can be minimized.

  In one embodiment, the light irradiated on the mold specimen 10 by the lighting device 220 has a light uniformity (Uniformity) on the horizontal axis and a light uniformity (Uniformity) on the vertical axis of 96% or more, The difference between the light uniformity and the light uniformity on the vertical axis may be within 2%. Table 1 below shows the light uniformity for one such embodiment.

  Referring to Table 1, the light irradiated by the lighting device 220 has a light uniformity of 96% or more on the horizontal axis and a light uniformity of the vertical axis at a distance of 110 mm to 300 mm from the mold test piece. It can be seen that the condition that the difference in axial light uniformity is within 2% is satisfied.

  That is, when the light uniformity for each of the horizontal axis and the vertical axis itself has a numerical value of 96% or more, and the condition that the difference in light uniformity between the two axes is 2% or less is satisfied, the difference in light uniformity is The segregated image is not distorted by the above-mentioned or can be offset by the operation of the segregation information generation unit 130 described below.

  FIG. 3 is a block diagram for explaining an embodiment of the test piece image extracting unit shown in FIG. 1, and FIG. 4 is a diagram showing that a piece image is extracted by the test piece image extracting unit shown in FIG. It is a reference figure explaining a mode.

  Referring to FIG. 3, the test piece image extraction unit 120 may include a matching unit 310, a binarizer 320, and an image extractor 330.

  The matching unit 310 can match the correlation coefficient between the template image and the captured image.

  The binarizer 320 can binarize the output from the matching unit 310.

  The image extractor 330 can extract a plurality of test piece images from the captured image based on the output from the binarizer 320.

  The image extractor 330 can number each test strip image while extracting a plurality of test strip images.

  With further reference to FIG. 4, FIG. 4 (a) shows a captured image of the mold specimen and FIG. 4 (b) shows a template image.

  FIG. 4C shows an example of a result obtained by matching the correlation coefficient between the template image and the captured image by the matching unit 310. It can be seen that a point having the most similar form to the template image is detected from the captured image by matching the correlation coefficient by the matching unit 310.

  FIG. 4D shows an example of a result obtained by binarizing FIG. 4C by the binarizer 320.

  FIG. 9E shows an example in which a plurality of test piece images are extracted from the captured image by the image extractor 330 based on FIG.

  FIG. 5 is a reference diagram for explaining how the captured image is corrected by the test piece image extracting unit shown in FIG.

  If the mold is rotated by the operator when the mold test piece is placed on the segregation analyzer, the recognition rate may decrease. Therefore, when the mold is rotated and placed, the test piece image extraction unit 120 can rotationally correct the captured image without the operator placing the mold again.

  That is, the test piece image extraction unit 120 detects the center point for the plurality of test piece images, and uses the at least part of the plurality of points belonging to the highest line among the plurality of center points to determine the array line 510. Can be formed.

  The test piece image extraction unit 120 can rotate the captured image in accordance with the angle difference between the reference line 520 preset in the horizontal direction and the array line 510.

  FIG. 6 is a block diagram illustrating an embodiment of the segregation information generation unit shown in FIG.

  First, referring to FIG. 6, the segregation information generation unit 130 may include a variable binarizer 510 and a segregation region determiner 520. According to the embodiment, the segregation information generation unit 130 may further include an information generator 530.

  The variable binarizer 510 can set the peripheral region of the central region in the test piece image as a reference region, and can variably binarize the central region using the average brightness of the reference region.

  The segregation area determiner 520 can determine, in the output from the variable binarizer 510, an area where the pixel value histogram of the reference area that has been binarized is equal to or less than a preset threshold value as a segregation area.

  The information generator 530 calculates the segregation analysis information by calculating at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation area detected from the segregation area determiner 520. can do.

  That is, the information generator 530 can generate quantitative numerical information with the degree of segregation as a simple order. For example, the information generator 530 may determine the size, length, brightness information, amount of segregation, and the like of the segregation using image processing techniques on the segregation area, and quantify it as quantitative data. it can.

  FIG. 7 is a reference diagram for explaining the state in which segregation is extracted by the segregation information generation unit shown in FIG. 6. Further referring to FIG. 7, (a) in FIG. Show.

  Referring to FIG. 7B, the variable binarizer 510 sets a peripheral region 720 of the center region 710 as a reference region in the test piece image, and the center region 710 based on the average brightness value of the reference region 720. Can be binarized.

  The result of the variable binarization is as shown in FIG. 7C, and the segregation region determiner 520 can determine the region 731 having an area of a certain area or more as the segregation region. That is, the segregation region determiner 520 can remove noise by deleting the binarized value 732 that is less than a certain size in the variable binarization result.

  FIG. 8 is a graph for explaining the segregation determination performed by the segregation information generation unit shown in FIG.

  As described above, the segregation area determiner 520 can determine an area where the pixel value histogram of the variable binarized reference area is equal to or less than a preset threshold value as the segregation area.

  The graph of FIG. 8 shows the relationship between the pixel value (Gray level) and the histogram (Histogram) of the reference region that is variably binarized.

  In the example shown in the figure, it can be seen that the lower 2% is set as the threshold, and the region corresponding to the lower 2% as the threshold in the pixel value histogram is determined as the segregation region. However, depending on the embodiment, the lower% value as the threshold value may be changed and set.

  This shows that in one embodiment of the present invention in which uniform light is emitted, the pixel values are shown with a relatively uniform distribution, similar to the example shown in the histogram for the image. Therefore, the region corresponding to the dark region of 2% or less of the lower level is a region having a dark value higher than the average brightness of the peripheral region, and this corresponds to the segregation region.

  FIG. 9 is a reference graph for explaining variable binarization by the variable binarizer shown in FIG. The variable binarization will be described with further reference to FIG. 9. The variable binarization means that the binarization reference varies as the average brightness value of the reference region 720 varies.

  That is, in the conventional fixed value binarization, the binarization is performed by applying the image brightness of the center area with a preset reference value as a reference. Since the binarization is performed on the central region with reference to the average brightness value, even when the test piece images have different overall brightness, the binarization can be performed accurately.

  FIG. 10 shows the result of binarization of fixed values for the test piece images of the same test piece having different brightness, and FIG. 11 shows the test pieces of the same test piece having different brightness. It is a reference figure showing the result of having performed variable binarization by the present invention to the image.

  Since fixed-value binarization is used in FIG. 10, it can be seen that in FIG. 10A, a region that is not segregated is also detected as segregation, and in FIG. 10B, segregation is not detected.

  On the other hand, since variable binarization is used in FIG. 11 according to the present invention, it can be seen that segregation is detected in the same manner even when the brightness in FIGS. 11A and 11B is different from each other.

  The various embodiments of the segregation analyzer have been described above with reference to FIGS.

  In the following, various embodiments of the segregation analysis method will be described with reference to FIG. However, the segregation analysis method described below can be more easily understood by referring to the description of the segregation analysis apparatus described in advance with reference to FIGS.

  FIG. 12 is a flowchart illustrating a segregation analysis method according to an embodiment of the present invention.

  Referring to FIG. 12, the segregation analyzer can acquire a captured image of a mold test piece including a plurality of test pieces (S1210).

  The segregation analyzer can extract a plurality of test piece images for the plurality of test pieces from the captured image (S1220).

  The segregation analyzer can detect a segregation area from each of a plurality of test piece images (S1230), and can generate segregation analysis information by quantifying the segregation area (S1240).

  In one embodiment for step S1210, the segregation analyzer may irradiate the plurality of specimens with light using a ring illumination having a hollow corresponding to the shape of the camera lens.

  In one embodiment with respect to step S1220, the segregation analyzer can match the correlation coefficient between the template image and the captured image, and perform binarization on the result of matching the correlation coefficient. Thereafter, based on the binarized result, the plurality of test piece images can be extracted from the captured image.

  In one embodiment for step S1230, the segregation analyzer may set a peripheral region of the center region as a reference region in the test piece image, and variably binarize the center region using the average brightness of the reference region. it can. Thereafter, a region where the pixel value histogram of the reference region that has been binarized with respect to the variable binarized result value is less than or equal to a preset threshold value can be determined as the segregation region.

  In one embodiment for step S1240, the segregation analyzer generates at least one of the size, length, brightness information, angle, amount of segregation, and segregation ratio of the segregation region to generate the segregation analysis information. be able to.

The present invention described above is not limited by the above-described embodiment and the attached drawings, but is limited by the scope of claims described later, and the configuration of the present invention does not depart from the technical idea of the present invention. It will be readily apparent to those skilled in the art to which the present invention pertains that various changes and modifications can be made to the configuration within the scope.

  The present invention relates to a segregation analysis apparatus and method that can automatically determine the center segregation grade of a wire product using an image.

  The wire is a linear steel and can be used in various fields such as tire cords and construction. Since such wire is generally required to have higher strength than ordinary steel, there is a high demand for quality control.

  When center part segregation exists in the inside of such a wire, there is a very high possibility that disconnection or the like is caused. Therefore, various techniques for detecting segregation of such a wire have been developed.

  In the conventional segregation detection technique for a wire, whether or not segregation exists in the center of the wire is determined based on an image of a cross section of the wire. On the other hand, in the case of such a conventional technique, there is a problem that when the captured image is changed due to illumination, camera performance, or the like, the detection accuracy of segregation is lowered.

  Such prior art can be easily understood by referring to Korean Published Patent No. 2009-0046920, Korean Published Patent No. 2010-0078590, Korean Published Patent No. 2012-0068635, etc. Can do.

  The present invention is for solving the above-mentioned problems of the prior art, and one technical aspect of the present invention is to detect segregation for each of a plurality of test pieces and provide segregation analysis information for the segregation. Thus, it is possible to provide a segregation analysis apparatus and method capable of automating test piece inspection and quality control and providing accurate analysis.

  One technical aspect of the present invention proposes a segregation analyzer. The segregation analyzer includes an image acquisition unit that acquires a captured image for a mold test piece including a plurality of test pieces, a test piece image extraction unit that extracts a plurality of test piece images for the plurality of test pieces from the captured image, respectively. And a segregation information generation unit that detects a segregation area from each of the plurality of test piece images and quantifies the segregation area to generate segregation analysis information.

  In one embodiment, the image acquisition unit may include a ring illumination having a hollow corresponding to a shape of a camera and a lens of the camera.

  In one embodiment, the test piece image extraction unit includes a matching unit that matches a correlation coefficient between a template image and the captured image, a binarizer that performs binarization on an output from the matching unit, and the above An image extractor that extracts the plurality of test piece images from the captured image based on an output from the binarizer may be included.

  In one embodiment, the segregation information generation unit sets a peripheral region of a central region in a test piece image as a reference region, and variable binary that variably binarizes the central region using the average brightness of the reference region. And a segregation area determiner that determines, in the output from the variable binarizer, the area where the pixel value histogram of the reference area that has been variably binarized is equal to or less than a preset threshold value as the segregation area. be able to.

  In one embodiment, the segregation information generation unit calculates at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation region to generate the segregation analysis information. A generator can further be included.

  Another technical aspect of the present invention proposes a segregation analysis method. The segregation analysis method includes obtaining a captured image for a mold test piece including a plurality of test pieces, extracting a plurality of test piece images for the plurality of test pieces from the captured image, and a plurality of test piece images. Detecting a segregation region from each of them, and quantifying the segregation region to generate segregation analysis information.

  In one embodiment, acquiring the captured image may include irradiating the plurality of specimens with light using a ring illumination having a hollow corresponding to the shape of the lens of the camera.

  In one embodiment, the step of extracting the plurality of test piece images includes a step of matching a correlation coefficient between the template image and the captured image, a step of binarizing the result of matching the correlation coefficient, And extracting the plurality of test piece images from the captured image based on the binarized result.

  In one embodiment, the step of generating the segregation analysis information includes setting a peripheral region of the center region as a reference region in the test piece image, and variable binarizing the center region using the average brightness of the reference region. And a step of determining, as the segregation region, a region where the pixel value histogram of the reference region that has been binarized is equal to or less than a preset threshold value with respect to the result of variable binarization.

  In one embodiment, the step of generating the segregation analysis information includes calculating the segregation analysis information by calculating at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation region. Generating may be included.

  The means for solving the problems described above do not enumerate all the features of the present invention. Various means for solving the problems of the present invention can be understood in more detail by referring to the specific embodiments in the following detailed description.

  According to one embodiment of the present invention, segregation is detected for each of a plurality of test pieces, and the segregation analysis information for the segregation is provided, thereby automating the inspection and quality control of the test pieces and performing accurate analysis. There is an effect that it can be provided.

  According to an embodiment of the present invention, since the presence / absence of segregation is determined using an average value of brightness based on variable binarization, it is possible to accurately detect segregation even if illumination or the like changes. Play.

It is a block block diagram explaining the segregation analyzer by one Embodiment of this invention. FIG. 2 is a reference diagram illustrating an embodiment of an image acquisition unit illustrated in FIG. 1. It is a block block diagram explaining one Embodiment of the test piece image extraction part shown by FIG. FIG. 4 is a reference diagram for explaining how a test piece image is extracted by a test piece image extraction unit shown in FIG. 3. FIG. 3 is a reference diagram for explaining how a captured image is corrected by a test piece image extraction unit shown in FIG. 1. It is a block block diagram explaining one Embodiment of the segregation information generation part shown by FIG. It is a reference figure explaining a mode that a segregation is extracted by the segregation information generation part shown by FIG. It is a graph for demonstrating the segregation determination performed by the segregation information generation part shown by FIG. It is a reference graph for demonstrating the variable binarization by the variable binarizer shown by FIG. It is a reference figure which shows the result of having performed the fixed value binarization with respect to the test piece image which has mutually different brightness. It is a reference figure which shows the result of having performed variable binarization with respect to the test piece image which has mutually different brightness. It is a flowchart explaining the segregation analysis method by one Embodiment of this invention.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

  However, the embodiments of the present invention can be modified in various other forms, and the scope of the present invention is not limited to the embodiments described below. Also, the embodiments of the present invention are provided to more fully explain the present invention to those having average knowledge in the art.

  FIG. 1 is a block diagram illustrating a segregation analyzer according to an embodiment of the present invention.

  Referring to FIG. 1, the segregation analyzer 100 may include an image acquisition unit 110, a test piece image extraction unit 120, and a segregation information generation unit 130.

  The image acquisition unit 110 can acquire a captured image for a mold test piece including a plurality of test pieces.

  The test strip image extraction unit 120 can extract a plurality of test strip images for a plurality of test strips from the captured image.

  The test strip image extraction unit 120 can classify the extracted plurality of test strip images by assigning numbers.

  When the captured image is rotated, the test piece image extraction unit 120 can detect the image and correct it in the forward direction.

  The segregation information generation unit 130 can detect a segregation area from each of a plurality of test piece images, and can generate segregation analysis information by quantifying the segregation area.

  The segregation information generation unit 130 can detect the segregation area from the center area of the test piece image. For example, the segregation information generation unit 130 can perform variable binarization using the average brightness of the peripheral region of the central region, and can detect the segregation region from the central region of the test piece image.

  Hereinafter, various embodiments of the segregation analyzer 100 will be described in more detail with reference to FIGS.

  FIG. 2 is a reference diagram for explaining an embodiment of the image acquisition unit shown in FIG.

  Referring to FIG. 2, the image acquisition unit 110 may include a camera 210 and a lighting device 220 including a plurality of light source elements. In the illustrated example, the illumination device 220 is shown as a ring illumination device, but this is exemplary. Therefore, according to the embodiment, the lighting device can be modified in various forms including a plurality of light source elements.

  The camera 210 includes a lens 211, and can capture an image by photographing the mold test piece 10 including a plurality of test pieces.

  In one embodiment, the camera 210 can comprise a telecentric lens. Therefore, it is possible to eliminate the segregation size distortion that occurs in normal camera photography.

  The illumination device 220 may be a ring-shaped illumination device including a plurality of light source elements, and may include a hollow corresponding to the shape of the lens 211 of the camera. Therefore, as in the illustrated example, when the illumination device is located, a plurality of light source elements are arranged around the camera, so that the mold test piece can be irradiated with light uniformly. Therefore, the change in brightness can be minimized, and the generation of noise due to the unevenness on the surface of the test piece can be minimized.

  In one embodiment, the light irradiated on the mold specimen 10 by the lighting device 220 has a light uniformity (Uniformity) on the horizontal axis and a light uniformity (Uniformity) on the vertical axis of 96% or more, The difference between the light uniformity and the light uniformity on the vertical axis may be within 2%. Table 1 below shows the light uniformity for one such embodiment.

  Referring to Table 1, the light irradiated by the lighting device 220 has a light uniformity of 96% or more on the horizontal axis and a light uniformity of the vertical axis at a distance of 110 mm to 300 mm from the mold test piece. It can be seen that the condition that the difference in axial light uniformity is within 2% is satisfied.

  That is, when the light uniformity for each of the horizontal axis and the vertical axis itself has a numerical value of 96% or more, and the condition that the difference in light uniformity between the two axes is 2% or less is satisfied, the difference in light uniformity is The segregated image is not distorted by the above-mentioned or can be offset by the operation of the segregation information generation unit 130 described below.

  FIG. 3 is a block diagram for explaining an embodiment of the test piece image extracting unit shown in FIG. 1, and FIG. 4 is a diagram showing that a piece image is extracted by the test piece image extracting unit shown in FIG. It is a reference figure explaining a mode.

  Referring to FIG. 3, the test piece image extraction unit 120 may include a matching unit 310, a binarizer 320, and an image extractor 330.

  The matching unit 310 can match the correlation coefficient between the template image and the captured image.

  The binarizer 320 can binarize the output from the matching unit 310.

  The image extractor 330 can extract a plurality of test piece images from the captured image based on the output from the binarizer 320.

  The image extractor 330 can number each test strip image while extracting a plurality of test strip images.

  With further reference to FIG. 4, FIG. 4 (a) shows a captured image of the mold specimen and FIG. 4 (b) shows a template image.

  FIG. 4C shows an example of a result obtained by matching the correlation coefficient between the template image and the captured image by the matching unit 310. It can be seen that a point having the most similar form to the template image is detected from the captured image by matching the correlation coefficient by the matching unit 310.

  FIG. 4D shows an example of a result obtained by binarizing FIG. 4C by the binarizer 320.

  FIG. 9E shows an example in which a plurality of test piece images are extracted from the captured image by the image extractor 330 based on FIG.

  FIG. 5 is a reference diagram for explaining how the captured image is corrected by the test piece image extracting unit shown in FIG.

  If the mold is rotated by the operator when the mold test piece is placed on the segregation analyzer, the recognition rate may decrease. Therefore, when the mold is rotated and placed, the test piece image extraction unit 120 can rotationally correct the captured image without the operator placing the mold again.

  That is, the test piece image extraction unit 120 detects the center point for the plurality of test piece images, and uses the at least part of the plurality of points belonging to the highest line among the plurality of center points to determine the array line 510. Can be formed.

  The test piece image extraction unit 120 can rotate the captured image in accordance with the angle difference between the reference line 520 preset in the horizontal direction and the array line 510.

  FIG. 6 is a block diagram illustrating an embodiment of the segregation information generation unit shown in FIG.

First, referring to FIG. 6, the segregation information generation unit 130 may include a variable binarizer 610 and a segregation region determiner 620 . According to the embodiment, the segregation information generation unit 130 may further include an information generator 630 .

The variable binarizer 610 can set the peripheral region of the central region in the test piece image as a reference region, and can variably binarize the central region using the average brightness of the reference region.

The segregation area determiner 620 can determine, in the output from the variable binarizer 610 , an area in which the pixel value histogram of the reference area that has been binarized is equal to or less than a preset threshold value as a segregation area.

The information generator 630 calculates the segregation analysis information by calculating at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation area detected from the segregation area determiner 620. can do.

That is, the information generator 630 can generate quantitative numerical information with the degree of segregation as a simple order. For example, the information generator 630 may determine the size, length, brightness information, amount of segregation, and the like of the segregation using image processing techniques for the segregation area, and quantify it as quantitative data. it can.

  FIG. 7 is a reference diagram for explaining the state in which segregation is extracted by the segregation information generation unit shown in FIG. 6. Further referring to FIG. 7, (a) in FIG. Show.

Referring to FIG. 7B, the variable binarizer 610 sets a peripheral area 720 of the center area 710 as a reference area in the test piece image, and the center area 710 based on the average brightness value of the reference area 720. Can be binarized.

The result of the variable binarization is as shown in FIG. 7C, and the segregation region determiner 620 can determine the region 731 having an area of a certain area or more as the segregation region. That is, the segregation region determiner 620 can remove noise by deleting the binarized value 732 that is less than a certain size in the variable binarization result.

  FIG. 8 is a graph for explaining the segregation determination performed by the segregation information generation unit shown in FIG.

As described above, the segregation area determiner 620 can determine an area where the pixel value histogram of the variable binarized reference area is equal to or less than a preset threshold value as the segregation area.

  The graph of FIG. 8 shows the relationship between the pixel value (Gray level) and the histogram (Histogram) of the reference region that is variably binarized.

  In the example shown in the figure, it can be seen that the lower 2% is set as the threshold, and the region corresponding to the lower 2% as the threshold in the pixel value histogram is determined as the segregation region. However, depending on the embodiment, the lower% value as the threshold value may be changed and set.

  This shows that in one embodiment of the present invention in which uniform light is emitted, the pixel values are shown with a relatively uniform distribution, similar to the example shown in the histogram for the image. Therefore, the region corresponding to the dark region of 2% or less of the lower level is a region having a dark value higher than the average brightness of the peripheral region, and this corresponds to the segregation region.

  FIG. 9 is a reference graph for explaining variable binarization by the variable binarizer shown in FIG. The variable binarization will be described with further reference to FIG. 9. The variable binarization means that the binarization reference varies as the average brightness value of the reference region 720 varies.

  That is, in the conventional fixed value binarization, the binarization is performed by applying the image brightness of the center area with a preset reference value as a reference. Since the binarization is performed on the central region with reference to the average brightness value, even when the test piece images have different overall brightness, the binarization can be performed accurately.

  FIG. 10 shows the result of binarization of fixed values for the test piece images of the same test piece having different brightness, and FIG. 11 shows the test pieces of the same test piece having different brightness. It is a reference figure showing the result of having performed variable binarization by the present invention to the image.

  Since fixed-value binarization is used in FIG. 10, it can be seen that in FIG. 10A, a region that is not segregated is also detected as segregation, and in FIG. 10B, segregation is not detected.

  On the other hand, since variable binarization is used in FIG. 11 according to the present invention, it can be seen that segregation is detected in the same manner even when the brightness in FIGS. 11A and 11B is different from each other.

  The various embodiments of the segregation analyzer have been described above with reference to FIGS.

  In the following, various embodiments of the segregation analysis method will be described with reference to FIG. However, the segregation analysis method described below can be more easily understood by referring to the description of the segregation analysis apparatus described in advance with reference to FIGS.

  FIG. 12 is a flowchart illustrating a segregation analysis method according to an embodiment of the present invention.

  Referring to FIG. 12, the segregation analyzer can acquire a captured image of a mold test piece including a plurality of test pieces (S1210).

  The segregation analyzer can extract a plurality of test piece images for the plurality of test pieces from the captured image (S1220).

  The segregation analyzer can detect a segregation area from each of a plurality of test piece images (S1230), and can generate segregation analysis information by quantifying the segregation area (S1240).

  In one embodiment for step S1210, the segregation analyzer may irradiate the plurality of specimens with light using a ring illumination having a hollow corresponding to the shape of the camera lens.

  In one embodiment with respect to step S1220, the segregation analyzer can match the correlation coefficient between the template image and the captured image, and perform binarization on the result of matching the correlation coefficient. Thereafter, based on the binarized result, the plurality of test piece images can be extracted from the captured image.

  In one embodiment for step S1230, the segregation analyzer may set a peripheral region of the center region as a reference region in the test piece image, and variably binarize the center region using the average brightness of the reference region. it can. Thereafter, a region where the pixel value histogram of the reference region that has been binarized with respect to the variable binarized result value is less than or equal to a preset threshold value can be determined as the segregation region.

  In one embodiment for step S1240, the segregation analyzer generates at least one of the size, length, brightness information, angle, amount of segregation, and segregation ratio of the segregation region to generate the segregation analysis information. be able to.

  The present invention described above is not limited by the above-described embodiment and the accompanying drawings, but is limited by the scope of claims described later, and the configuration of the present invention does not depart from the technical idea of the present invention. It will be readily apparent to those skilled in the art to which the present invention pertains that various changes and modifications can be made to the configuration within the scope.

Claims (10)

An image acquisition unit for acquiring a captured image of a mold test piece including a plurality of test pieces;
A test strip image extraction unit for extracting a plurality of test strip images for the plurality of test strips from the captured image;
A segregation information generating unit that detects a segregation area from each of a plurality of test piece images, quantifies the segregation area, and generates segregation analysis information;
Including a segregation analyzer. The image acquisition unit
A camera,
An illumination device including a plurality of light source elements and having a hollow corresponding to the shape of the lens of the camera;
The segregation analyzer according to claim 1, comprising: The test piece image extraction unit includes:
A matching unit for matching a correlation coefficient between the template image and the captured image;
A binarizer that binarizes the output from the matching unit;
An image extractor for extracting the plurality of test piece images from the captured image based on an output from the binarizer;
The segregation analyzer according to claim 1, comprising: The segregation information generation unit
A variable binarizer that sets a peripheral region of the center region in a test piece image as a reference region, and variably binarizes the center region using an average brightness of the reference region;
A segregation region determiner that determines a region in which the pixel value histogram of the reference region that is variably binarized is equal to or less than a preset threshold value as the segregation region;
The segregation analyzer according to claim 1, comprising: The segregation information generation unit
The information generator according to claim 1, further comprising: an information generator that calculates at least one of a size, a length, brightness information, an angle, a segregation amount, and a segregation ratio of the segregation region to generate the segregation analysis information. Segregation analyzer. Obtaining a captured image for a mold specimen including a plurality of specimens;
Extracting a plurality of test strip images for the plurality of test strips, respectively, from the captured image;
Detecting a segregation area from each of a plurality of test piece images, generating the segregation analysis information by quantifying the segregation area;
Including a segregation analysis method. The step of acquiring the captured image includes
Irradiating light in which the horizontal axis light uniformity and the vertical axis light uniformity are 96% or more, and the difference between the horizontal axis light uniformity and the vertical light uniformity is within 2%. The segregation analysis method according to claim 6. Extracting the plurality of specimen images;
Matching a correlation coefficient between a template image and the captured image;
Performing binarization on the result of matching the correlation coefficient;
Extracting the plurality of test piece images from the captured image based on the binarized result;
The segregation analysis method according to claim 6, comprising: The step of generating the segregation analysis information includes:
Setting a peripheral region of the central region as a reference region in the specimen image;
Variable binarization of the central area using the average brightness of the reference area;
Determining, as the segregation area, an area in which the pixel value histogram of the reference area that has been binarized is equal to or less than a preset threshold for the result of variable binarization;
The segregation analysis method according to claim 6, comprising: The step of generating the segregation analysis information includes:
The segregation analysis method according to claim 6, further comprising: calculating at least one of the size, length, brightness information, angle, segregation amount, and segregation ratio of the segregation region to generate the segregation analysis information. .

Images