JP5951867B2 - Absorbent article inspection method - Google Patents

Description

  TECHNICAL FIELD The present invention relates to an absorbent article inspection method for inspecting the state of attachment of a translucent material such as a nonwoven fabric of an absorbent article.

In the absorbent article, a back sheet (back sheet) is attached to the outermost part of the absorbent body to prevent liquid leakage. When the back sheet is arranged in an exposed state, the back sheet is easily torn during use, and therefore a non-woven fabric (outer layer non-woven fabric) may be attached to the outside of the back sheet.
In the manufacturing process of absorbent articles, if the outer nonwoven fabric is peeled off for some reason or part of it is turned up and the back sheet is exposed, the back sheet may be torn during product use and may cause liquid leakage There is sex.
When the peeling of the outer nonwoven fabric is inspected by image processing during the manufacturing process of the absorbent article, it is difficult to inspect the turned-up state of the outer nonwoven fabric by imaging the outer nonwoven fabric using transmitted illumination and processing the image. . In general, since the outer layer nonwoven fabric is thin and translucent, the contrast between the image where the outer layer nonwoven fabric is peeled off and the image where the outer layer nonwoven fabric is affixed cannot be obtained, and stable inspection cannot be performed.
It is also difficult to inspect the turned-up state of the exterior nonwoven fabric by imaging the outer nonwoven fabric using reflected illumination and processing the image. In general, both the back sheet and the outer nonwoven fabric are often white, and it is difficult to inspect the presence or absence of the same color by image processing based on the captured image.

  As a method for inspecting the peeling of the sheet, a method for inspecting the peeling of the sheet of the crimped postcard (overlapping sending body) is disclosed. A crimped postcard is a postcard in which another sheet is attached to a printed sheet. In the method for inspecting a crimped postcard, the color of the corner of the superimposed sending body is specified based on the imaging data obtained by imaging the corner of the postcard. A method of determining whether or not the corner portion of the overlap sending body is peeled off by comparing the color of the corner portion of the overlapping sending body and the color determination pattern stored in the storage unit in advance. It is. The color determination pattern stored in the storage unit includes a color itself, a color pattern, a determination pattern mark, and the like, which are stored according to the sending object to be inspected (see Patent Document 1).

JP 2012-052983 A

In the inspection method described in Patent Document 1, it is possible to determine the presence or absence of a sheet when colors, pattern shapes, determination pattern marks, and the like are clearly different. However, when the sheet is translucent, it was impossible to inspect the attached state of the sheet.
The present invention relates to a method for inspecting an absorbent article that enables inspection of the presence or absence of a translucent nonwoven fabric.

  The present invention relates to an absorbent article inspection method for inspecting the arrangement state of a translucent sheet placed on a component part constituting a printed absorbent article, wherein the translucent sheet is placed on the component part Illuminating with illumination light, imaging the inspection area including the translucent sheet on the printed area, acquiring color information from the captured image, and acquiring in advance Comparing the color range of the translucent sheet with the acquired color information and extracting the existing area of the translucent sheet; and based on the extracted existing area of the translucent sheet A method for determining whether the arrangement state of the absorbent article is good or bad.

  The inspection method for absorbent articles according to the present invention makes it possible to detect the arrangement state of the translucent sheet and determine whether the arrangement state is good or bad.

It is a principal part block diagram which showed a preferable example of the test | inspection apparatus which implements the test | inspection method of the absorbent article of this invention. It is the top view which showed a preferable example of the test | inspection area | region of the inspection method of the absorbent article of this invention. It is the flowchart which showed a preferable example (embodiment) of the inspection method of the absorbent article of this invention. It is a figure which shows the sample of the printing density which showed the reference color of color range designation | designated in the flowchart which implements the test | inspection method of embodiment based on the CMYK color system. It is the flowchart which showed a preferable example of the color range specification in the flowchart which implements the test | inspection method of embodiment. It is an HSV histogram in the presence or absence of a translucent nonwoven fabric for one reference color, (a) is an example when one translucent nonwoven fabric is arranged, and (b) is an example when no translucent nonwoven fabric is arranged. It is. FIG. 6 is a relationship diagram between saturation and printing density obtained from each histogram when there is no translucent nonwoven fabric and one sheet when the reference color is dark blue (C100M90). FIG. 6 is a relationship diagram between saturation and printing density obtained from each histogram when the reference color is green (C80Y100) when there is no translucent nonwoven fabric and when there is one sheet. FIG. 6 is a relationship diagram between saturation and printing density obtained from each histogram when the reference color is orange (M75Y100) when there is no translucent nonwoven fabric and when there is one sheet. It is a relationship figure of the saturation and printing density which are obtained from each histogram when there is no translucent nonwoven fabric when the reference color is blue (C100M40) and when there is one sheet. FIG. 10 is a relationship diagram between saturation and printing density obtained from each histogram when the reference color is purple (C55M100) when there is no translucent nonwoven fabric and when there is one sheet. It is the flowchart which showed a preferable example of the color extraction process in the flowchart which implements the test | inspection method of embodiment. It is the flowchart which showed a preferable example of the result discrimination | determination process in the flowchart which implements the test | inspection method of embodiment. It is drawing explaining the discrimination | determination method in the case of making the case where one piece of semi-transparent nonwoven fabric is arranged in the printing area | region of a back surface sheet into a normal good article. (A) is a drawing-substituting photograph showing a state in which one non-woven fabric is arranged on the printing region, (b) is a histogram showing the relationship between the number of pixels and saturation, and (c) is after saturation extraction. (D) is an image after contraction and expansion processing. It is drawing explaining the discrimination | determination method when the case where the translucent nonwoven fabric is not arranged in the printing area | region of a back surface sheet is made into a defective article. (A) is a drawing-substituting photograph showing a state in which a non-woven fabric is not arranged on the print region, (b) is a histogram showing the relationship between the number of pixels and saturation, and (c) is a graph after saturation extraction. (D) is an image after contraction and expansion processing. It is drawing explaining the discrimination | determination method in the case of making the case where two translucent nonwoven fabrics are distribute | arranged in the printing area | region of a back surface sheet into a defective article. (A) is a drawing-substituting photograph showing a state in which two non-woven fabrics are arranged on the printing region, (b) is a histogram showing the relationship between the number of pixels and saturation, and (c) is after saturation extraction. (D) is an image after contraction and expansion processing. It is drawing which showed the detection algorithm of the translucent nonwoven fabric. (A) is the top view which showed the test | inspection area | region when a translucent nonwoven fabric is distribute | arranged on the back surface sheet, (b) is a translucent nonwoven fabric distribute | arranged in the state where some corners turned over on the back surface sheet. (C) is a histogram showing the relationship between the number of pixels obtained by processing an image obtained by imaging the inspection area and saturation, and (d) is binarized. It is the figure which showed the process, (e) is the figure which showed the determination result of a good article and inferior goods. When a white translucent non-woven fabric is arranged on a white back sheet, in the case of a groove of the white back sheet, when a white translucent non-woven fabric is arranged on the white back sheet, It is the graph which showed HSV (hue, saturation, brightness) of 100% standard color of green (C80Y100) about the case where a printing field is arranged on a white back sheet. (A) shows hue, (b) shows saturation, and (c) shows lightness. When a white translucent non-woven fabric is arranged on a white back sheet, in the case of a groove of the white back sheet, when a white translucent non-woven fabric is arranged on the white back sheet, It is the graph which showed HSV (hue, saturation, brightness) of 100% standard color of orange (M75Y100) about the case where the printing area is arranged on the white back sheet. (A) shows hue, (b) shows saturation, and (c) shows lightness. When a white translucent non-woven fabric is arranged on a white back sheet, in the case of a groove of the white back sheet, when a white translucent non-woven fabric is arranged on the white back sheet, It is a graph which showed HSV (hue, saturation, brightness) of 100% standard color of blue (C100M40) about the case where a printing field is arranged on a white back sheet. (A) shows hue, (b) shows saturation, and (c) shows lightness. When a white translucent non-woven fabric is arranged on a white back sheet, in the case of a groove of the white back sheet, when a white translucent non-woven fabric is arranged on the white back sheet, It is the graph which showed HSV (hue, saturation, lightness) of the 100% standard color of purple (C55M100) about the case where the printing area is arranged on the white back sheet. (A) shows hue, (b) shows saturation, and (c) shows lightness. It is drawing which shows the relationship between a printing area | region and a test | inspection area | region. (A) is a top view, (b) is an A-A 'sectional view. It is drawing which showed the example of printing of the printing area | region of an inspection area | region. (A), (b) is a plan view showing a preferred example, (c) to (e) are plan views showing an unfavorable example, and (f), (g) are "previous mark" It is the top view which showed the example contained. It is the schematic block diagram which showed a preferable example of the measuring apparatus which measures the transmittance | permeability of a semi-transparent nonwoven fabric.

  A preferred embodiment of an inspection method for absorbent articles according to the present invention will be described below with reference to the drawings. First, a preferred example of an inspection apparatus that performs an inspection method for absorbent articles will be described with reference to FIG.

As shown in FIG. 1, the absorbent article inspection apparatus 100 of the present invention irradiates illumination light L from an illumination unit 110, and reflects reflected light Lr of an inspection region (not shown) illuminated by the illumination light L. The imaging unit 120 receives light and images.
The illumination unit 110 includes a light emitting unit 111 that emits light that illuminates the inspection area, and a power supply unit 112 that emits light from the light emitting element of the light emitting unit 111. Although not shown, the light emitting unit 111 uses a white light emitting element such as an LED light emitting element, an organic electroluminescence (organic EL) element, a white fluorescent tube, or an incandescent bulb.

  The illumination unit 110 preferably uses an illumination device that emits normal white illumination light (color temperature is, for example, 4200K to 6500K). Although not shown, a plurality of white LED light emitting elements are arranged in a plurality of rows and a plurality of columns so as not to cause unevenness in the illumination light L irradiated to the absorber 10. Furthermore, in order to obtain illumination light with uniform illuminance without unevenness, it is preferable that a reflecting plate is provided on the side surface in the illumination device. The illuminance may be illuminance at which a reflected image can be obtained, and is preferably equal to or less than the illuminance that does not cause overexposure, although it depends on the sensitivity of the image sensor. Moreover, the irradiation angle with respect to the translucent nonwoven fabric 21 of the illumination light L is 10 degrees or more and 80 degrees or less, Preferably it is 30 degrees or more and 60 degrees or less from a viewpoint of irradiating an imaging range uniformly. In the illustrated example, the irradiation angle is 45 °.

  The imaging unit 120 preferably uses an imaging device having a camera body 121 and an imaging lens 122. By using a solid-state image sensor as the image sensor of the camera body 121, image processing by digital processing is facilitated. As the solid-state imaging device, a color imaging device may be used as long as the imaging device can express, for example, 128 gradations or more, and may be a charge coupled device (CCD) or a CMOS sensor. The number of pixels of the image sensor is preferably 400 pixels or more, more preferably 600 pixels or more in order to obtain a sufficient resolution. Note that since the image processing speed is slow, it is not necessary to have a higher image quality (higher pixels) than necessary. Further, gradation expression may be enhanced by setting two pixels, four pixels, or more pixels of the image sensor as one pixel. Further, the imaging lens 122 may be a lens having a focal length capable of imaging the inspection region, and a long focal lens is preferable to a short focal lens in order to reduce image distortion. A long focal lens is a lens having a longer focal length than a standard lens. The standard lens means a lens having a focal length of approximately 43 mm to 58 mm in terms of a 35 mm format of a normal film camera.

  An image processing unit 130 that processes the captured image signal is connected to the imaging unit 120. As the image processing unit 130, for example, a dedicated image sensor controller may be used, or a personal computer may be used. The image processing described below may be executed, and an “inspection start signal” may be input to start the inspection.

  Next, an embodiment of an inspection method including image processing of an absorber image acquired by the imaging unit 120 will be described below.

  As shown in FIG. 2, the inspection region 31 is set in advance on the back sheet 2 that is hardly transparent and is arranged on the non-skin surface of the absorbent body in the absorbent article. Difficult transparency refers to a state in which the contours of shapes such as characters, symbols, and figures printed on an object covered with a back sheet cannot be easily visually recognized. For example, the translucent nonwoven fabric 21 which is a translucent sheet is used for the nonwoven fabric under the absorber disposed on the outermost portion of the back sheet 2 (the non-skin surface side and the outermost surface on the skin surface side). Semi-transparent means a state in which the outline of a printed character, symbol, figure, or the like of a printed object placed on the back side of the semi-transparent sheet can be visually confirmed. An inspection region 31 is set near each corner of the translucent nonwoven fabric 21. The inspection region 31 is about a half of the area of the inspection region 31 on the translucent nonwoven fabric 21 on the print region 33 that is printed on the back sheet 2 and has a design of colored figures, characters, symbols, and the like. For example, it is preferable to set a rectangular shape. More specifically, the inspection regions 31 are set at a total of four locations in the width direction ends of both sides facing each other in the longitudinal direction of the translucent nonwoven fabric 21. Moreover, the inspection region 31 is set so that, for example, half of the inspection region 31 is applied to the translucent nonwoven fabric 21 and the remaining half is separated from the translucent nonwoven fabric 21 in the longitudinal direction and exists on the back sheet 2. This setting is not limited to half of the inspection area 31.

In the inspection apparatus 100 shown in FIG. 1, first, as shown in FIG. 3, “color range designation” S1 is performed. Next, it is determined whether or not the inspection start signal Sg is input in “existence of input of inspection start signal” S2.
The inspection start signal Sg is input by an external device (not shown) or an inspector. If the presence / absence of inspection start signal input “S2” is “No”, that is, if there is no input of inspection start signal Sg, input of inspection start signal Sg is waited.
In the case of “Yes” in the above determination, that is, when the inspection start signal Sg is input, the imaging region 31 described with reference to FIG. 2 is imaged from the translucent nonwoven fabric 22 side in “imaging” S3. I do. Then, the captured image including the inspection region 31 is acquired as an RGB image.
Next, in the “extraction region extraction” S4, the image processing device 130 shown in FIG. 1 extracts the image of the inspection region 31 (see FIG. 3) from the captured image.
Next, in “color extraction process” S5, the captured image in the inspection area is converted from the RGB color system to the HSV color system, and the color information of the inspection area 31 is acquired. This conversion may be converted to the HSL color system. Then, pixel binarization processing is performed. That is, an image is created by converting pixels in the inspection area and within the non-defective range into white (255, 255, 255) and pixels outside the non-defective range into black (0, 0, 0).
Next, a “result determination process” S6 is performed. Details thereof will be described later.
Then, “determining non-defective product and outputting the result” S7 determines whether the color of the inspection area is non-defective or defective and outputs the result.
In the case of “Yes”, the non-defective product determination result flag F is turned ON and “non-defective product signal output” S8 is performed. On the other hand, in the case of “No”, the non-defective product discrimination result flag F is turned OFF and “defective product signal output” S9 is performed.

The “color range designation” S1 will be described in detail.
“Color range designation” S1 is a designation of the color of the design used when determining whether or not the arrangement of the translucent nonwoven fabric is within a specified range for all colors used in the design of the back sheet 2. For example, as shown in FIG. 4, it is assumed that five colors of amber, green, orange, blue, and purple are used. In this case, for example, a reference color that is 100% is determined based on the CMYK color system. For example, the dark blue color is C100M90, the green color is C80Y100, the orange color is M75Y100, the blue color is C100M40, and the purple color is C55M100. Then, for each 100% reference color, a color whose transparency is reduced by 10% from 100% to 10% is obtained, and these are used as reference colors for each transparency. The transparency can be a printing density. Each sample can be, for example, 10 mm × 10 mm. The size of the sample is not limited to the above, and may be larger, or may be smaller when the resolution of the imaging unit 110 (see FIG. 1) is sufficiently high. The size in the case of a square is preferably 20 mm square, more preferably 10 mm square.

Specifically, as shown in FIG. 5, “imaging” S11 for designating a color range is performed on the sample. In this imaging, the 100% reference color and each reference color whose transparency (printing density) has been sequentially lowered are photographed to determine an allowable range when the printing density of the back sheet varies. Imaging is performed for each reference color when there is no translucent nonwoven fabric and when there is one translucent nonwoven fabric, and for each, saturation is obtained from the RGB image of the captured image in the inspection area. That is, the saturation is obtained based on the existing area of the translucent nonwoven fabric.
As the translucent nonwoven fabric 21, a spunbond nonwoven fabric made of white polypropylene fibers having a basis weight of 17 g / m ² is used.
The shooting conditions are: aperture f = 1.4, shutter speed = 1/5000 s (electronic shutter), and illumination illuminance 8000 lx.

Subsequently, in each “shoot from RGB to HSV” S12, an image obtained by the above shooting is changed from the RGB color system (hereinafter also referred to as RGB) to the HSV color system (hereinafter also referred to as HSV). Convert. Specifically, the RGB color space is converted to the HSV color space. This conversion is based on a known conversion method.
Conversion from RGB to HSV is performed as follows.
The maximum M = max (R, G, B) and the minimum m = min (R, G, B) are obtained from R, G, Bε [0, 1] in the RGB color space. That is, the maximum value and the minimum value of each color R, G, B are obtained.
The saturation S _{HSV and} saturation S _HSL of the HSV and HSL color spaces are obtained from two values of M and m.
When M = m = 0, S _HSV = 0.
In cases other than M = m = 0, S _HSV = (M−m) / M.
In the above equation, it can be seen that when M = m, that is, R = G = B, the saturation is 0, and when m = 0, that is, when any of R, G, B is 0, the saturation is 1.
However, in the case of black in HSV, the saturation is undefined as it is, so the saturation in that case is 0.
The HSL color space is the same as the HSV color space.
When M = m = 0, S _HSL = 0.
In cases other than M = m = 0, S _HSL = (M−m) / (1− | M + m−1 |).
Also in this case, it is understood that the saturation is 0 when M = m, that is, R = G = B, and the saturation is 1 when m = 0, that is, any of R, G, and B is 0.
However, in the case of HSL in white or black, the saturation is indefinite as it is, so the saturation in that case is 0.

Next, in “Specify non-defective range in inspection region” S13, an HSV histogram of an HSV image in the inspection region is obtained, and a non-defective range of HSV is specified based on the histogram. As shown in FIG. 6, the non-defective product range is a region that is a mountain that includes the saturation peak of the reference print portion.
First, an HSV histogram is obtained for each color of the reference colors. For example, regarding the 100% reference color of the amber color, when one translucent nonwoven fabric is coated, a histogram as shown in FIG. 6A is obtained. When no translucent nonwoven fabric is coated, a histogram as shown in FIG. 6B is obtained. Such a histogram is obtained for each reference color density from 100% to 0%.

Next, the color range of each color is obtained from each histogram. In each color range, the saturation due to the print density when there is no translucent nonwoven fabric of each reference color and when there is one sheet is obtained. The result is as shown in FIGS.
As shown in FIG. 7, in the case of dark blue (C100M90), a case where the print density is 60% or more is recognized as a good product. The gradation difference between the white color and the print area was 159 gradations.
As shown in FIG. 8, in the case of green (C80Y100), a case where the print density is 80% or more is recognized as a good product. Note that the gradation difference between white and the printing area was 108 gradations.
As shown in FIG. 9, in the case of orange (M75Y100), a case where the print density is 70% or more is recognized as a non-defective product. The gradation difference between the white color and the print area was 158 gradations.
As shown in FIG. 10, in the case of blue (C100M40), a case where the print density is 70% or more is recognized as a non-defective product. The gradation difference between the white color and the print area was 175 gradations.
As shown in FIG. 11, in the case of purple (C55M100), a case where the print density is 70% or more is recognized as a good product. Note that the gradation difference between the white color and the print area was 93 gradations.
Recognizing a good product is recognized as a good product when there is a difference of, for example, 50 gradations or more between the saturation when there is no translucent nonwoven fabric and the saturation when there is one sheet. Thus, if the gradation difference of saturation is 50 gradations or more, the presence or absence of the translucent nonwoven fabric can be reliably recognized.

  Next, in “storage of good product range” S14, the good product range of each color is stored in a storage unit (not shown) in the image processing unit 130 (see FIG. 1), and the color range designation is completed.

Next, the “color extraction process” S5 shown in FIG. 3 will be described.
The “color extraction process” S5 is performed on the image of the inspection region 31 extracted in the “extraction region extraction” S4 from the image including the inspection region 31 obtained by the “imaging” S3 as described above.
As shown in FIG. 12, the “color extraction process” S5 converts the image in the inspection area 31 from the RGB color system (color space) to the HSV color system (color space) in “convert from RGB to HSV” S21. . This conversion may be converted to the HSL color system. Subsequently, “pixel binarization processing” S22 is performed. That is, the preexisting color range of the translucent sheet determined in the above-mentioned “color range designation” S1 is compared with the acquired color information to extract the existing area of the translucent nonwoven fabric. The existence area of the translucent sheet is extracted by the color range and the saturation of the color information. Then, an image in which pixels in the inspection area and in the non-defective range are converted to white (255, 255, 255) and pixels outside the range are converted to black (0, 0, 0) is created.
In “storage of converted image” S23, the image converted into the HSV is stored in a storage unit (not shown) in the image processing unit 130 (see FIG. 1). This point will be described later with reference to FIG.

Next, the “result determination process” S6 shown in FIG. 3 will be described.
As shown in FIG. 13, the “result discrimination process” S6 counts the white pixels obtained by the “pixel binarization process” S22 in the “counting the number of white pixels in the inspection area” S31. Then, the preset upper limit pixel number TH and lower limit pixel number TL are compared with the number of white pixels n (see S32). As a result, if n is in the range of TL <n <TH, it is determined as a non-defective product, and “determination result flag F is set to ON” S33. On the other hand, if it is outside the above range, it is determined as a defective product, and “determination result flag F is OFF” S34.
The upper limit pixel number TH and the lower limit pixel number TL are determined based on the measurement result of the white pixel number n. That is, n shakes are measured by imaging the same sample a plurality of times, or n shakes are measured by imaging a plurality of samples. Then, an upper limit value n _max and a lower limit value n _min of n are obtained. Here, when a margin for fluctuation of n is α, TL = n _min −α and TH = n _max + α. α is set as an arbitrary positive integer of 0 or more as a margin for fluctuation.

For example, in a portion where something is printed on the back sheet 2, a case where one translucent non-woven fabric is arranged is regarded as a normal non-defective product, and no translucent non-woven fabric and two translucent non-woven fabrics are arranged. If it is, it was regarded as defective. In this case, in the above inspection method, the determination is made as follows.
As shown to Fig.14 (a), in the part by which something was printed on the back surface sheet 2, one piece of semi-transparent nonwoven fabric is distribute | arranged. In this case, when the color analysis of the print area is performed, a histogram as shown in FIG. 14B is obtained. That is, there is a peak indicating white characters in a low-saturation region, and a saturation peak indicating a reference print region. Each histogram has the number of pixels on the vertical axis and the saturation on the horizontal axis. Furthermore, by performing color (saturation) extraction processing, an image of a print area as shown in FIG. 14C is obtained. Further, by performing the contraction and expansion processing, an image of the print area as shown in FIG. 14D is obtained.

  As shown to Fig.15 (a), the translucent nonwoven fabric is not distribute | arranged in the part by which something was printed on the back surface sheet 2. FIG. In this case, when the color analysis of the print area is performed, a histogram as shown in FIG. 15B is obtained. That is, the peak which shows a white character exists in the area | region with low saturation similar to the case where one piece of translucent nonwoven fabric is arranged. Furthermore, a saturation peak indicating a printing area serving as a reference appears in an area where the saturation is high as much as there is no translucent nonwoven fabric. Further, by performing color (saturation) extraction processing, an image of a print area as shown in FIG. 15C is obtained. Further, by performing contraction and expansion processing as the filter processing, an image of the print area as shown in FIG. 15D is obtained.

  As shown in FIG. 16A, two translucent nonwoven fabrics are arranged in a portion where something is printed on the back sheet 2. In this case, when the color analysis of the print area is performed, a histogram as shown in FIG. 16B is obtained. That is, there is a peak indicating white characters in a region with low saturation similar to the case where one translucent nonwoven fabric is arranged. Furthermore, so as to overlap with this peak, a saturation peak indicating a reference printing area appears in a low saturation area by the amount of two translucent nonwoven fabrics. In addition, the small saturation peak at a position higher than the saturation peak is not completely covered with the two translucent nonwoven fabrics, and a part thereof is covered with one translucent nonwoven fabric. . Therefore, it is the peak of the saturation of the part coat | covered with one translucent nonwoven fabric. Further, by performing color (saturation) extraction processing, an image of a print area as shown in FIG. 16C is obtained. Further, by performing contraction and expansion processing as the filter processing, an image of the print area as shown in FIG.

The filter processing in the image processing generally uses a filter of pixels in 3 rows and 3 columns. Then, for an image of 3 × 3 pixels, an image with 0 to 255 gradations is processed as follows.
In the contraction process in the filter process, the luminance of the center pixel of 3 × 3 is replaced with the minimum luminance among the total of the central pixel and the surrounding pixels. By this process, white noise components are removed.
In the expansion process in the filter process, the luminance of the 3 × 3 central pixel is replaced with the maximum luminance among the total of the central pixel and the surrounding pixels. By this process, the black noise component is removed.

  Thus, every time the semi-transparent nonwoven fabric overlaps on the back sheet 2, the saturation of the printing region is lowered. Accordingly, by extracting the saturation when there is one translucent nonwoven fabric, it is possible to detect whether the translucent nonwoven fabric is turned or not.

Next, the detection algorithm of a translucent nonwoven fabric will be described below according to the above detection method.
As shown to Fig.17 (a), (b), the test | inspection area | region 31 when the translucent nonwoven fabric 21 is distribute | arranged on the back surface sheet 2 (or absorber (not shown)) is set. This setting method is as described above with reference to FIG. FIG. 17A shows a case of a non-defective product, and FIG. 17B shows a case of a defective product in which the translucent nonwoven fabric 21 is partially turned.

  Next, the inspection area 31 is imaged by the “imaging” S3, and the image of the inspection area is extracted from the captured image including the inspection area by the “extraction area extraction” S4. Next, the “color extraction process” S6 is performed. As described above, the color printed on the back sheet 2 in a state where one semi-transparent non-woven fabric is arranged in advance is measured. Then, as shown in FIG. 17C, a non-defective range G of saturation is set from the obtained histogram in the same manner as described above.

  At this time, the color of the photographed inspection area is measured by “Conversion from RGB to HSV” S21. Then, by “binarization processing” S22, as shown in FIG. 17D, the portion where the saturation is in the non-defective range G is converted to white, and the area of the white portion is calculated. Next, in the “non-defective product determination” S8, as shown in FIG. 17 (e), if the calculated area is within the range of the non-defective product, it is determined that the product is “OK”. NG ".

Next, in the case where a white translucent nonwoven fabric is arranged on a white backsheet, the above-mentioned 100% reference color HSV of green (C80Y100), orange (M75Y100), blue (C100M40), and purple (C55M100) (Hue, saturation, brightness) were examined. Similarly, in the case of only the white back sheet, when the print area is arranged on the white back sheet and the white translucent nonwoven fabric is arranged thereon, and when the print area 33 is arranged on the white back sheet, The HSV of 100% reference colors of green, orange, blue and purple was examined. The print area has a so-called solid coating in which only the reference colors are printed without characters. In addition, it becomes possible to distinguish the kind of absorbent article by changing each color according to the kind of absorbent article.
The results are shown in FIGS.

As shown in FIG. 18, in the case of green color (C80Y100), a difference larger than the error range in the print area was generated in the saturation depending on the presence or absence of the translucent nonwoven fabric (see (b)). There was no significant difference in hue (see (a)), and in brightness, a difference was found when a white translucent nonwoven fabric was placed on a white backsheet. On the other hand, although there was a difference on the print area, it was within the error range (see (c)).
Therefore, it was demonstrated that the presence or absence of a translucent nonwoven fabric can be determined by detecting the saturation on the print area.

As shown in FIG. 19, in the case of orange (M75Y100), a difference larger than the error range in the print area was produced in the saturation depending on the presence or absence of the translucent nonwoven fabric (see (b)). In the hue, when there was no translucent nonwoven fabric in the printing region, the hue was 0, and when there was a translucent nonwoven fabric, there was a slight hue, and there was a slight difference between them (see (a)). Further, in lightness, a difference was found when a white translucent nonwoven fabric was arranged on a white back sheet. On the other hand, although there was a difference on the print area, it was within the error range (see (c)).
Therefore, it was demonstrated that the presence or absence of a translucent nonwoven fabric can be determined by detecting the saturation on the print area.

As shown in FIG. 20, in the case of blue (C100M40), a difference larger than the error range in the printing region was produced in the saturation depending on the presence or absence of the translucent nonwoven fabric (see (b)). There was no significant difference in hue (see (a)), and in brightness, a difference was found when a white translucent nonwoven fabric was placed on a white back sheet. On the other hand, there was a difference on the print area, but it was within the error range (see (c)).
Therefore, it was demonstrated that the presence or absence of a translucent nonwoven fabric can be determined by detecting the saturation on the print area.

As shown in FIG. 21, in the case of purple (C55M100), the difference in saturation in the printing region was larger than the error range depending on the presence or absence of the translucent nonwoven fabric (see (b)). There was no significant difference in hue (see (a)), and in brightness, a difference was found when a white translucent nonwoven fabric was placed on a white back sheet. On the other hand, there was a difference on the print area, but it was within the error range (see (c)).
Therefore, it was demonstrated that the presence or absence of a translucent nonwoven fabric can be determined by detecting the saturation on the print area.

The shape of the print area is preferably a rectangular inspection area, and any design may be used as long as it is within the inspection area.
For example, as shown in FIGS. 22A and 22B, a part of the side sheet 5 is folded on the outermost side in the width direction of the back sheet 2, and the translucent nonwoven fabric 21 is further formed in the longitudinal direction of the back sheet 2. It is arranged. A print area 33 (see FIG. 23) is arranged in the inspection area 31 of the back sheet 2. In such a case, the inspection area 31 is set so as to avoid the covering area because both sides in the width direction of the back sheet are covered with the side sheets 5. For example, it is preferable to avoid a range of 45 mm from both sides of the back sheet 2.
Moreover, in order to increase the detection sensitivity of the angular breakage of the semi-transparent nonwoven fabric 21, the inspection region 31 is spaced apart in the width direction, and two places are arranged on both ends in the longitudinal direction (also referred to as longitudinal front and rear ends). It is preferable. That is, it is preferable to dispose them at four locations apart.
For example, it is preferable to arrange the inspection region 31 so that the center in the width direction coincides with the position 20 mm outside the center in the width direction of the back sheet 2. In consideration of the arrangement fluctuation of the translucent nonwoven fabric 21 or the arrangement fluctuation of the back sheet 2, it is preferable to arrange the translucent nonwoven fabric 21 at a position 15 mm inward in the longitudinal direction from the longitudinal front and rear end portions.
The inspection area 31 is preferably set to ± 20 mm in the longitudinal direction from the center of the printing area 33, for example. In addition, the width of the inspection region 31 is set to 20 mm in order to provide the detection stability of the print region 33 and the freedom of design. That is, the inspection area 31 is 40 mm in the longitudinal direction and 20 mm in the width direction. The area of the print region 33 in the area of the inspection region 31 is preferably 50% or more in order to maintain detection sensitivity.
In addition, it is preferable that the print area 33 in the detection area 31 is colored in a solid color, and it is preferable that a print area 33 that is a color area of 50% or more with respect to the area of the solid color is secured. .

As shown in FIGS. 23A and 23B, the print area 33 can be composed of patterns having various shapes with respect to the inspection area 31. However, it is preferable that the size of one pattern is larger than, for example, 7 mm square considering the stability of inspection, that is, the resolution of the captured image.
Further, as shown in FIGS. 23 (f) and 23 (g), the print area 33 may include a front mark 34 indicating that it is “front” of the absorbent article with respect to the inspection area 31. . The front mark is preferably one in which characters, symbols, etc. are written in a solid rectangle.
The above numerical values are examples, and may be appropriately changed depending on the size of the absorbent article, detection sensitivity, and the like.

  In the said embodiment, the presence or absence of a translucent nonwoven fabric and an overlapping state can be test | inspected, without impairing the external appearance of a product (absorbent article). In addition, if a range that is regarded as a non-defective product is registered in advance, it is possible to inspect whether the design is changed or the color is changed. Alternatively, if a color registered in advance in the inspection area is used, it is possible to detect even a slight design change.

The translucent nonwoven fabric 21 is composed of fibers such as polypropylene (PP), polyethylene terephthalate (PET), polyethylene (PE), etc., for example, chemical bond nonwoven fabric, thermal bond nonwoven fabric, airlaid nonwoven fabric, spunlace nonwoven fabric, spunbond. Nonwoven fabrics, melt blown nonwoven fabrics, needle punched nonwoven fabrics, stitch bond nonwoven fabrics and the like can be mentioned.
The basis weight of the translucent nonwoven fabric 21 is 10 g / m ² or more and 50 g / m ² or less, preferably 15 g / m ² or more and 30 g / m ² or less. The translucent nonwoven fabric 21 has a white light transmittance of 20% or more, preferably 40% or more, and more preferably 50% or more.

The transmissivity of white light of the translucent nonwoven fabric 21 is measured by a transmissivity measuring apparatus 200 shown in FIG.
As shown in FIG. 24, the transmittance measuring apparatus 200 irradiates the illumination light L2 from the illumination unit 210 and transmits the transmitted light Lt of the translucent nonwoven fabric test piece 250 illuminated by the illumination light L2 irradiated from the illumination unit 210. Is received and imaged by the imaging unit 220.
The illumination unit 210 includes a power supply unit 212 that emits light from the light emitting element 211 that emits light that illuminates the inspection region. Although not shown, the light emitting unit 211 uses a white light emitting element such as an LED light emitting element, an organic electroluminescence (organic EL) element, a white fluorescent tube, or an incandescent bulb.

The illumination unit 210 preferably uses an illumination device that emits normal white illumination light (color temperature is, for example, 4200K to 6500K). Further, although not shown, a plurality of white LED light emitting elements are arranged in a plurality of rows so as not to cause unevenness in the illumination light L2 irradiated to the test piece 250 (100 mm × 100 mm) of the translucent nonwoven fabric held by the holding unit 240. Arranged in columns. Further, it is preferable that a fly-eye lens (not shown) or the like is provided in order to obtain illumination light with uniform illuminance without unevenness. The illuminance may be illuminance at which a transmission image can be obtained, and is 1000 lx or more, preferably 2000 lx or more, and more preferably 5000 lx or more. Further, depending on the sensitivity of the image sensor, the illuminance or less that does not cause overexposure is 50000 lx or less, preferably 30000 lx or less.
In order to suppress uneven illumination, it is preferable that the illumination is reflected in the center of the image, and the range of the illumination central part (for example, 50 mm × 50 mm) is set as the measurement region.

  The imaging unit 220 preferably uses an imaging device having a camera body unit 221 and an imaging lens unit 222. By using a solid-state image sensor as the image sensor of the camera body 221, digital image processing is facilitated. As the solid-state image sensor, either a color image sensor or a monochrome image sensor can be used. Any imaging device capable of expressing gradations of 256 gradations or more may be used, and it may be a charge coupled device (CCD) or a CMOS sensor. The number of pixels of the image sensor is preferably 150,000 pixels or more, more preferably 300,000 pixels or more in order to obtain a sufficient resolution. Since the image processing speed is slow, it is not necessary to have higher image quality (higher pixels) than necessary. Further, the gradation expression may be enhanced by using a plurality of pixels as one pixel in two or four pixels or more of the image sensor.

An image processing unit 230 that processes the captured image signal is connected to the imaging unit 220. For example, a dedicated image sensor controller may be used as the image processing unit 230, or a personal computer may be used.
The transmittance measuring device 200 is used as a method for measuring the transmittance of the translucent nonwoven fabric test piece 250. First, an image is taken without a test piece, and the brightness of the illumination at that time is measured from the imaged image. That is, the average gradation value in the measurement area is calculated.
Next, the semi-transparent nonwoven fabric test piece 250 is imaged, and the average gradation value in the measurement region at that time is calculated from the captured image. The gradation value of each pixel is an average value of RGB gradation values.
Then, if the gradation is A when there is no nonwoven fabric and the gradation is B when there is a nonwoven fabric, the transmittance can be obtained by (B / A) × 100 (%).

  The present invention further discloses the following embodiments with respect to the above-described embodiments.

<1>
A method for inspecting an absorbent article for inspecting an arrangement state of a translucent sheet placed on a component constituting an absorbent article on which printing has been performed,
Irradiating illumination light to a component on which the translucent sheet is arranged, and imaging an inspection area including the translucent sheet on the printed area;
Obtaining color information from the captured image;
A step of extracting the existing area of the translucent sheet by comparing the color range information of the translucent sheet acquired in advance with the acquired color information;
And a step of determining whether the arrangement state of the semi-transparent sheet is good or not based on the extracted region of the semi-transparent sheet.
<2>
The method for inspecting an absorbent article according to <1>, wherein the translucent sheet is made of a nonwoven fabric.
<3>
The said color range and the said color information are the inspection methods of the absorbent article as described in <1> or <2> which uses HSV color system or HSL color system.
<4>
The method for inspecting an absorbent article according to <3>, wherein the presence region of the translucent sheet is extracted based on the color range and the saturation of the color information.
<5>
The method for inspecting an absorbent article according to any one of <1> to <4>, wherein the illumination for irradiating the illumination light uses reflected illumination.
<6>
The said reflective illumination is a test | inspection method of the absorbent article as described in <5> which uses white light.
<7>
The inspection method for absorbent articles according to any one of <1> to <6>, wherein the inspection area is set on a back sheet disposed on the non-skin side of the absorbent body in the absorbent article.
<8>
The said inspection area | region is an inspection method of the absorbent article any one of <1> to <7> set so that 1/2 of the area of this inspection area | region may cover the said semi-transparent sheet | seat.
<9>
A semi-transparent non-woven fabric, which is the semi-transparent sheet, is used as the non-absorbent nonwoven fabric disposed on the outermost surface (non-skin side) of the back sheet of the absorbent article, and the inspection is performed near each corner of the semi-transparent non-woven fabric. The inspection method for absorbent articles according to any one of <1> to <8>, wherein the region is set.
<10>
Inspection of the absorbent article according to <9>, wherein the inspection area is set on the printing area that is printed on the non-skin contact surface side of the back sheet and includes a design of colored figures, characters, and symbols. Method.
<11>
The inspection region is an inspection method for absorbent articles according to <9> or <10>, wherein the inspection regions are set at a total of four locations in the width direction ends of both sides facing each other in the longitudinal direction of the translucent nonwoven fabric.
<12>
The said inspection area | region covers the said translucent nonwoven fabric, The remainder remove | deviates from the said translucent nonwoven fabric, and exists on the said back sheet, The inspection method of any one of <9> to <11>.
<13>
The inspection method for absorbent articles according to any one of <9> to <12>, wherein the inspection region is imaged from the translucent nonwoven fabric side and a captured image of the inspection region is acquired.
<14>
<13> The method for inspecting an absorbent article according to <13>, wherein the captured image in the inspection region is converted from RGB to an HSV color system, and pixel binarization processing is performed.
<15>
The absorptivity according to <14>, in which an image in which pixels in the inspection region and in the non-defective range are converted to white (255, 255, 255) and pixels outside the non-defective range are converted to black (0, 0, 0) is created. Inspection method for goods.
<16>
The white pixels in the inspection area obtained by the binarization processing of the pixels are counted, the preset upper limit pixel number TH and lower limit pixel number TL are compared with the white pixel number n, and n is TL <n <TH. The inspection method for absorbent articles according to <14> or <15>, in which a non-defective product is determined as long as it falls within the range, and the other products are determined as defective products.
<17>
<1> to <16> An absorbent article inspection apparatus in which the method for inspecting an absorbent article according to any one of <1> is implemented,
An illumination unit for illuminating the inspection area;
An imaging unit that images the inspection region illuminated by the illumination unit;
Color information is acquired from the image of the inspection area captured by the imaging unit, and the color range of the translucent sheet acquired in advance is compared with the acquired color information to determine the presence area of the translucent sheet. An inspection apparatus for absorbent articles, comprising: a processing unit that extracts and determines whether the arrangement state of the translucent sheet is good or not based on the extracted region of the translucent sheet.
<18>
The inspection device for absorbent articles according to <17>, wherein the imaging unit uses an imaging device having a solid-state imaging device and an imaging lens.
<19>
The inspection apparatus for absorbent articles according to <18>, wherein the solid-state imaging device uses a color imaging device.
<20>
The inspection device for absorbent articles according to <18> or <19>, wherein the solid-state imaging device is an imaging device capable of expressing gradations of 128 gradations or more.
<21>
The inspection device for absorbent articles according to any one of <18> to <20>, wherein the imaging lens is a long focal lens.

DESCRIPTION OF SYMBOLS 2 Back sheet 5 Side sheet 21 Translucent nonwoven fabric 31 Inspection area 33 Printing area 100 Inspection apparatus 110 Illumination part 111 Light emission part 112 Power supply part 120 Imaging part 121 Camera body part 122 Imaging lens part 130 Image processing part 200 Transmittance measurement apparatus 210 Illumination Unit 211 light emitting unit 212 power supply unit 220 imaging unit 221 camera body unit 222 imaging lens unit 230 image processing unit 240 gripping unit L, L2 illumination light Lr reflected light Lt transmitted illumination light

Claims (7)

A method for inspecting an absorbent article for inspecting an arrangement state of a translucent sheet placed on a component constituting an absorbent article on which printing has been performed,
Irradiating illumination light to a component on which the translucent sheet is arranged, and imaging an inspection area including the translucent sheet on the printed area;
Obtaining color information from the captured image;
Comparing the color range of the translucent sheet acquired in advance with the acquired color information and extracting the existing area of the translucent sheet;
And a step of determining whether the arrangement state of the semi-transparent sheet is good or not based on the extracted region of the semi-transparent sheet.   The said translucent sheet | seat consists of a nonwoven fabric, The test | inspection method of the absorbent article of Claim 1   The said color range and the said color information are the inspection methods of the absorbent article of Claim 1 or 2 which uses a HSV color system or a HSL color system.   The method for inspecting an absorbent article according to claim 3, wherein the presence area of the translucent sheet is extracted based on the color range and the saturation of the color information.   The method for inspecting an absorbent article according to any one of claims 1 to 4, wherein reflection illumination is used as illumination for irradiating the illumination light.   The method for inspecting an absorbent article according to claim 5, wherein the reflected illumination uses white light. An absorbent article inspection apparatus in which the absorbent article inspection method according to any one of claims 1 to 6 is implemented,
An illumination unit for illuminating the inspection area;
An imaging unit that images the inspection region illuminated by the illumination unit;
Color information is acquired from the image of the inspection area captured by the imaging unit, and the color range of the translucent sheet acquired in advance is compared with the acquired color information to determine the presence area of the translucent sheet. An inspection apparatus for absorbent articles, comprising: a processing unit that extracts and determines whether the arrangement state of the translucent sheet is good or not based on the extracted region of the translucent sheet.