JP5225064B2 - Inspection device - Google Patents

Description

  The present invention relates to an inspection apparatus that detects colored foreign matters and dirt attached to a sheet to be inspected.

2. Description of the Related Art Conventionally, in an inspection apparatus used for sheet defect inspection, a reflection lighting apparatus is installed in order to inspect dirt such as colored foreign matter and oil, and photographing with a camera is performed (for example, see Patent Document 1). . In the inspection apparatus disclosed in Patent Document 1, a filter having a color complementary to the colored foreign matter and dirt is used in order to accurately detect the colored foreign matter and dirt at the time of photographing. .
In addition, as an inspection device that can accurately inspect minute defects that exist at different positions such as the surface or inside of a light-transmitting object, it is equipped with a plurality of light sources that project light having different defect detection characteristics and wavelengths. An inspection apparatus has been proposed (for example, Patent Document 2).

JP 2001-318059 A JP 2000-9591 A

  However, in conventional inspection apparatuses, it is difficult to detect defects such as dirt on the surface of the sheet when it is attempted to increase the imaging accuracy of the foreign matter present in the sheet and the foreign matter present on the back surface of the sheet. When it is raised, there is a problem that it becomes difficult to detect foreign matter existing in the sheet, or foreign matter and dirt on the back side of the sheet.

  Therefore, the present invention can accurately detect both yellowish stains adhering to the surface of the sheet to be inspected (surface directed toward the first light source) and foreign matter in the sheet or on the front and back surfaces of the sheet. It relates to providing an inspection device.

  The present invention provides a first light source provided on one surface side of a sheet to be inspected, a second light source provided on the other surface side of the sheet to be inspected, and one surface side of the inspected sheet emitted from the first light source. An image sensor that receives the light reflected on the light and the light emitted from the second light source and transmitted through the sheet to be inspected to reach one surface; and image processing that processes an image signal output from the image sensor The light emitted from the first light source is blue light having a complementary color relationship with yellowish stains attached to the sheet to be inspected, and the second light source emits the light. The light provides an inspection device that is red light.

  According to the inspection apparatus of the present invention, it is possible to accurately detect both yellowish stains adhering to the surface of the sheet to be inspected (surface directed toward the first light source) and foreign matter in the sheet or on the front and back surfaces of the sheet. Can do. The yellow-colored dirt refers to dirt on which, for example, machine oil used in manufacturing equipment is adhered due to a malfunction.

Hereinafter, the present invention will be described based on preferred embodiments with reference to the drawings.
FIG. 1 shows an embodiment of an inspection apparatus according to the present invention. As shown in the drawing, the inspection apparatus according to the present embodiment includes a first light source 1 provided on one surface side of a sheet S to be inspected, which is a light transmissive sheet, and a wide width provided on the other surface side of the sheet S to be inspected. A second light source 2. In addition, the inspection apparatus of the present embodiment has the light emitted from the first light source 1 and reflected to one surface side by the inspection sheet S, and the light emitted from the second light source 2 and transmitted through the inspection sheet S to the one surface side. A CCD camera 3 having an image sensor for receiving the incident light and an image processing unit 4 for processing an image signal output from the image sensor are provided. In addition, the inspection apparatus of the present embodiment includes drive means (not shown) that travels the sheet S to be inspected, and amplifiers 5A and 5B that adjust the light amounts of the first light source 1 and the second light source 2. .
The image processing unit 4 includes an image processing device 41 connected to the CCD camera 3 and a monitor 42 connected to the image processing device 41. Examples of the image processing apparatus 41 include a personal computer main body in which image processing software is installed, a commercially available image processing machine (for example, Keyence Corporation CV-3500), and the like.

  In some cases, yellowish stains such as oil stains may adhere to the inspection sheet S. In order to detect this contamination, the light emitted from the first light source 1 is yellowish stains attached to the inspection sheet S. The blue light has a complementary color relationship. Here, the blue light means a case where the peak of the wavelength distribution of light is near the blue wavelength, and can be obtained by, for example, a blue light emitting diode. Alternatively, a blue filter may be attached to illumination that emits white light, such as a fluorescent lamp, and only blue light may be extracted. The light emitted from the second light source 2 is red light. Here, the red light means a case where the peak of the wavelength distribution of light is near the red wavelength, and can be obtained by, for example, a red light emitting diode. Or you may make it take out only red light by attaching a red filter to illumination which emits white light, such as a fluorescent lamp.

  In the present embodiment, the blue light is used for inspecting the surface of the sheet S to be inspected, and is effective for inspecting yellow-type soil (for example, oil stain) having a complementary color relationship with blue. By using the complementary color relationship, the brightness difference between the yellow stain and the sheet S to be inspected is wider than when the complementary color relationship is not used, so that the yellow stain can be easily detected in the image processing apparatus 41. It is. Moreover, when an operator confirms oil stains on the monitor 42, it is easy to visually recognize. The red light is used to inspect foreign matters in the inspection sheet S and dirt on the back surface of the inspection sheet S. The reason why red light is used is that the transmittance of the resin sheet is higher than that of other visible light. That is, visible light other than red light has a higher reflectance when entering the resin sheet than red light.

As the first light source 1 and the second light source 2, light emitting diodes (LEDs) are used. Since a specific range of wavelengths is emitted from the light emitting diode, the blue light from the first light source 1 and the red light from the second light source 2 do not mix.
A single-plate color CCD is preferably used as the image sensor in order to receive these light and take an image. Of course, the light from the first light source 1 and the second light source 2 is received and dispersed into components of each color (RGB) using a spectroscopic prism, and an image sensor for blue light and red light is provided for imaging. May be.
In the present embodiment, since blue light and red light are irradiated over a wide range in the inspection sheet S, the inspection sheet S is moved stepwise at predetermined intervals to inspect foreign matter and dirt. . The movement interval of the sheet S to be inspected is set to be narrower than the photographing range of the CCD camera 3 so that a non-inspected portion does not occur.

  FIG. 2 is a plan view of the inspection apparatus of the present embodiment. The inspection sheet S moves in the direction of the arrow in the figure. The second light source 2 has a rectangular shape, and a sheet S to be inspected is provided so as to cross almost most of the width. In the vicinity of the second light source 2, a pair of opposed rod-shaped first light sources 1 are provided. A CCD camera 3 is provided above the center of the rectangular second light source 2. FIG. 2 shows the main body portion 31 and the lens portion 32 of the CCD camera 3.

FIG. 3 shows the flow of operation processing of the inspection apparatus of this embodiment. First, in step S1, the CCD camera 3 receives red light and blue light, and transmits an image signal output from the image sensor to the image processing unit 4. Next, in step S2, the red component image and the blue component image are analyzed by the calculation unit of the image processing apparatus 41 to inspect for foreign matter and dirt.
In the inspection of the presence or absence of foreign matter and dirt, first, in step S3, the judgment of the presence or absence of foreign matter in the inspected sheet S or foreign matter on the front or back surface of the inspected sheet S is made based on the analysis result of the red component image. The If no foreign object exists, the process proceeds to step S4. If there is a foreign object, the presence of dust is displayed on the monitor 42 in step S7. In step S4, the surface of the inspection sheet S (the surface on the first light source side) is inspected for dirt based on the analysis result of the blue component image. If no dirt is found, it is determined in step S5 that the product is non-defective. If dirt is found, it is determined that the product is defective, and the presence of dirt is displayed on the monitor 42 in step S6.

An example of a preferable process for inspecting the presence or absence of foreign matter and dirt will be described.
(1) A red component image and a blue component image are separated from the captured image.
(2) Binarization processing is performed on the red and blue component images using threshold values T1 and T2, respectively. In the binarization process, the threshold values T1 and T2 are set low enough to prevent dark portions when a good image is used, and pixels below the threshold value are converted to black, and pixels equal to or higher than the threshold value are converted to white. Hereinafter, the binarized red component image is also referred to as an R image, and the binarized blue component image is also referred to as a B image.
(3) The dark area is searched from the R image to obtain the area. When the area of the dark part is equal to or greater than the threshold value T3 ^*, it is determined that there is a foreign object, and the coordinates are transferred to the memory M1.
(4) The dark area is searched from the B image to obtain the area. When the area of the dark part is equal to or greater than the threshold value T4 ^*, it is determined that there is a foreign substance or yellowish stain, and the coordinates are transferred to the memory M2.
As for the threshold T3 ^* and the threshold T4 ^* , for example, a minimum sample of foreign matter to be detected is photographed on a sheet, and the threshold is set from the image.
(5) The distinction between the dirt and the foreign matter and the front / back discrimination of the foreign matter are performed as follows based on the coordinate information of M1 and M2.
If the dark parts of the R image and the B image are on the same coordinates, it is determined that the foreign matter is on the sheet surface.
If there is a dark part only in the R image, it is determined as a foreign object on the back side of the sheet.
When only the B image has a dark portion, it is determined that the color is yellowish.
According to this method, yellow stains such as oil stains adhering to the sheet surface (surface directed toward the first light source) as well as foreign matters on the sheet or on the front and back surfaces of the sheet can be accurately detected. .

  In the present invention, the entire sheet having optical transparency can be used as a sheet to be inspected (a sheet to be inspected). Examples of the light transmissive sheet include plastic sheets, non-woven fabrics, fiber sheets such as paper, and laminates thereof. Among these, a nonwoven fabric and a semi-transparent uniform thing are good.

  As mentioned above, although preferable embodiment of this invention was described, this invention is not restrict | limited to those embodiment, It can change suitably.

  In the embodiment described above, an example in which an image is acquired using a single-plate color CCD has been described, but the present invention is not limited to this. For example, as shown in FIG. 4, the spectral prism 3 </ b> A is used to divide each wavelength (RGB) into wavelengths, which are separately received by the image sensor and analyzed by the image processing unit 4.

  In addition, when the inspection device of the present invention is introduced into the production line for processed sheet products as a process after determining that the inspected sheet has foreign matter or dirt, a signal indicating a defective sheet appearance is transmitted from the image processing unit to the subsequent process. Then, it can be discharged as a defective product.

〔Example〕
As shown in FIG. 5A, a black adhesive tape as a foreign material 6 is attached to one side of the sheet S to be inspected, and oil stains (yellowish stains) 7 are attached to the other side by machine oil. As the foreign material 8, a non-woven fabric to which a black adhesive tape was attached was used. The sizes of the foreign matter 6, the oil stain 7 and the foreign matter 8 are 1 × 1 mm, 4 mm × 6 mm and 1 × 1 mm, respectively, and the foreign matter 6 and the oil stain 7 are separated by 12 mm in the longitudinal direction of the nonwoven fabric. Non-woven fabric (20 g / m ² ) was used as the sheet S to be inspected, and LL-310 No. made by Sanhayato was used as the adhesive tape.
From the blue LED bar illumination (Keyence CA-DBB13) as the first light source on the surface (hereinafter also referred to as surface) having the foreign matter 6 and oil stain 7 of the nonwoven fabric using the inspection apparatus shown in FIGS. The blue light was irradiated, and the surface (hereinafter also referred to as the back surface) having the foreign material 8 of the nonwoven fabric was irradiated with red light from a red LED backlight illumination (CA-DSR9 manufactured by Keyence) as a second light source. The distance between the first light source and the nonwoven fabric was 50 mm, and the distance between the second light source and the nonwoven fabric was 20 mm. The distance between the CCD camera and the nonwoven fabric was 160 mm.

The blue light reflected on the surface of the nonwoven fabric and the red light transmitted through the nonwoven fabric were imaged with a CCD camera (Keyence CV-H100C), and the captured images were captured in a personal computer installed with image processing software. Using the image processing software, the red component and the green component of the captured image were removed, and only the blue component was binarized with a threshold value 50 to generate a B image. Similarly, the blue component and the green component were removed, and the threshold value was binarized with 65 for only the red component to generate an R image. FIG. 5B shows a blue component image (B image) after binarization, and FIG. 5C shows a red component image (R image) after binarization.
About the foreign material 6 on the surface of the nonwoven fabric, a dark portion 6 ′ corresponding to the foreign material 6 appeared on the same coordinates of the R image and the B image, and the areas of the foreign material 6 and the dark portion 6 ′ were the same. As for the oil stain 7, a dark portion 7 ′ corresponding to the oil stain 7 appears only in the B image, and the area of the oil stain and the area of the dark portion 7 ′ are almost the same. Regarding the foreign matter 8 on the back surface of the nonwoven fabric, the dark portion 8 ′ corresponding to the foreign matter 8 appeared only in the R image, and the areas of the foreign matter 8 and the dark portion 8 ′ were the same.
Therefore, when the dark portion 7 ′ appears only in the B image and no dark portion appears on the same coordinates of the R image, it can be determined that there is an oil stain 7 (yellowish stain). In addition, it is possible to determine whether the foreign matter exists on the nonwoven fabric surface or on the back surface when the dark portion is only on the R image, it can be determined that it is on the front surface, and when the dark portion is on the same coordinates in the R image and the B image, It can be determined that there is.

[Comparative Example]
As the sheet S to be inspected, the same non-woven fabric as that used in the example (with the foreign matters 6 and 8 and the oil stain 7 adhered) was used.
The surface of the nonwoven fabric (surface with oil stain) is irradiated with red light from diagonally upward by red LED bar lighting (CA-DBR13 manufactured by Keyence), and green LED bar lighting (manufactured by Moritex) from diagonally upward on the opposite side. MBRL-CG13015) was irradiated with green light. Moreover, the blue light was irradiated to the back surface of the nonwoven fabric by blue LED backlight illumination (CA-DSR9 made by Keyence). The distance between the red illumination and the nonwoven fabric was 50 mm, the distance between the green illumination and the nonwoven fabric was 50 mm, and the distance between the blue illumination and the nonwoven fabric was 20 mm.
The CCD camera (same as in the example) disposed 160 mm above the surface of the nonwoven fabric was used to image red light and green light reflected from the surface of the nonwoven fabric and blue light transmitted through the nonwoven fabric with the CCD camera. A blue component image and a red component image were generated from the captured images in the same manner as in the example, and the threshold values were set to 55 for the blue component and 65 for the red component, respectively. FIG. 5D shows a blue component image (B image) after binarization, and FIG. 5E shows a red component image (R image) after binarization.
As a result, using the binarized blue component image (B image) and the binarized red component image (R image), the foreign matter 6 existing on the surface of the nonwoven fabric and the foreign matter 8 existing on the back surface of the nonwoven fabric. However, the oil stain 7 could not be extracted for both the R image and the B image even if the threshold value was changed. When the brightness of the green illumination is adjusted, the brightness of the red component image and the blue component image changes, and it is necessary to readjust the binarization threshold.

FIG. 1 is an explanatory view showing an embodiment of the inspection apparatus of the present invention. FIG. 2 is a plan view of the inspection apparatus of FIG. FIG. 3 is a flowchart showing a flow of operation processing of the inspection apparatus of FIG. FIG. 4 is an explanatory view showing a modification of the inspection apparatus of FIG. FIG. 5 is an explanatory diagram for explaining a method for detecting dirt and foreign matter, and FIG. 5 (a) is a diagram showing a nonwoven fabric as a sheet to be inspected used in Examples and Comparative Examples, and FIG. b) is a binarized blue component image (B image) after being separated into the blue component image and the red component image of the embodiment, and FIG. 5C is separated into the blue component image and the red component image of the embodiment. It is a figure which shows the binarized red component image (R image) after performing. 5D is a binarized blue component image (B image) after being separated into a blue component image and a red component image of the comparative example, and FIG. 5E is a blue component image of the comparative example. It is a figure which shows the binarized red component image (R image) after isolate | separating into a red component image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 1st light source 2 2nd light source 3 CCD camera 3A Spectroscopic prism 4 Image processing part 41 Image processing apparatus 42 Monitor 5A, 5B Lighting amplifier S Test sheet

Claims (1)

A first light source provided on one side of the inspection sheet;
A second light source provided on the other side of the inspection sheet;
An image sensor that receives light emitted from the first light source and reflected to one surface by the inspection sheet; and light emitted from the second light source and transmitted through the inspection sheet to the one surface side; ,
An image processing unit for processing an image signal output from the image sensor;
Driving means for running the inspected sheet;
An inspection device comprising:
The light emitted from the first light source is blue light having a complementary color relationship with yellowish stains attached to the inspection sheet,
The light emitted from the second light source is red light ,
The first light source and the second light source are light emitting diodes,
The yellowish stain is oil stain,
The image processing unit is an inspection apparatus that inspects for the presence of foreign matter and dirt by the following method.
(1) A red component image and a blue component image are separated from the captured image.
(2) Binarization processing is performed on the red and blue component images using threshold values T1 and T2, respectively.
(3) The dark area is searched from the R image which is the binarized red component image, and the area is obtained. When the area of the dark part is equal to or greater than the threshold value T3 ^*, it is determined that there is a foreign object, and the coordinates are transferred to the memory M1.
(4) The dark area is searched from the B image which is the binarized blue component image, and the area is obtained. When the area of the dark part is equal to or greater than the threshold value T4 ^*, it is determined that there is a foreign substance or yellowish stain, and the coordinates are transferred to the memory M2.
(5) The distinction between the dirt and the foreign matter and the front / back discrimination of the foreign matter are performed as follows based on the coordinate information of the memory M1 and the memory M2.
When the dark part of the R image and the dark part of the B image are on the same coordinate, it is determined that the foreign object is on the one surface of the sheet to be inspected.
When a dark part exists only in the R image, it is determined as a foreign object on the other surface of the inspection sheet.
When only the B image has a dark part, it is determined as yellowish stains .

Images