JP2007017319A - Inspection device of coating irregularity - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an inspection device of coating irregularity capable of reducing the effect exerted on an inspection capacity with respect to the error factor present in both of the scanning direction (main scanning direction) of a line sensor camera and the feed direction (sub-scanning direction) of an inspection target. <P>SOLUTION: The inspection device of coating irregularity is equipped with a feed means for feeding a coating article in the direction right-angled to a coating direction for the purpose of sub-scanning in imaging, the line sensor camera 1 arranged in a feed route so as to incline the main scanning direction with respect to the coating direction for the purpose of main scanning in imaging and imaging the coating article to obtain the image of the coating article and a projection vector operating means for calculating the sum total of the pixel values of the pixels arranged in the coating direction of the obtained image to obtain the projection vector in the direction right-angled to the coating direction. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention belongs to the technical field of inspecting coating unevenness based on captured images. The present invention relates to a coating unevenness inspection apparatus that automatically inspects coating unevenness that occurs due to non-uniform discharge amounts at a plurality of nozzles in an inkjet head, non-uniform discharge slits in a die coat, and the like.

  The color filter has, for example, a configuration in which a large number of strip-like regions (stripes) that transmit each color of RGB (red, green, blue) are arranged in the order of RGB in the width direction (not RGB, but its complementary color, YMC ( yellow, magenta, cyan)). In order to give each stripe the property of transmitting each color of RGB, a coating film of each color of RGB is formed on the base material of the color filter. One of the methods for forming the coating film is an ink jet method. In the inkjet method, a coating film is formed on a substrate by ejecting and applying inks of RGB colors from an inkjet head. At this time, the base material of the color filter and the ink jet head are moved relative to each other in the extending direction of each stripe, whereby the entire stripe is coated.

In this ink jet method, coating unevenness occurs due to non-uniform discharge amounts at a plurality of nozzles existing in the ink jet head (see FIG. 9). Accordingly, there is a proposal for an inspection apparatus for inspecting such coating unevenness. For example, the direction of occurrence of coating unevenness in the color filter is set to be parallel to the scanning direction of the line sensor camera (see FIG. 10), and the image data obtained by imaging is orthogonal to the direction of occurrence of coating unevenness. There is a proposal of integrating in the direction and inspecting coating unevenness from the difference of the integrated values (Patent Document 1).
JP2003-42898

  In such a conventional inspection apparatus, the error factor existing in the scanning direction (main scanning direction) of the line sensor camera can be integrated to reduce the influence on the inspection performance. The error factors include individual differences among line sensor cameras, variations in sensitivity of light receiving elements, variations in luminance in the longitudinal direction of the light source, and the like. However, since the error factors existing in the conveyance direction (sub-scanning direction) to be inspected are reflected as they are without being integrated, the influence on the inspection performance cannot be reduced. The error factors are vibration during transportation, luminance fluctuations caused by flickering of the light source, and the like. Therefore, high inspection performance cannot be obtained with the conventional inspection apparatus.

  The present invention has been made to solve the above problems. The purpose of this is a coating unevenness inspection apparatus that can reduce the influence on the inspection performance of error factors that exist in both the scanning direction (main scanning direction) of the line sensor camera and the conveyance direction (sub-scanning direction) of the inspection object. Is to provide.

A coating unevenness inspection apparatus according to claim 1 of the present invention includes a transport unit that transports a coating material in a direction perpendicular to the coating direction for sub-scanning in imaging, and the coating direction for main scanning in imaging. A line sensor camera that is arranged on the conveyance path so that the main scanning direction is inclined and obtains a picked-up image by picking up the coating material, and obtains a sum of pixel values of pixels arranged in the coating direction in the picked-up image. Projection vector calculation means for obtaining a projection vector perpendicular to the coating direction.
A coating unevenness inspection apparatus according to claim 2 of the present invention is the coating unevenness inspection apparatus according to claim 1, further comprising geometric distortion correction means for correcting a geometric distortion of the captured image and obtaining a geometric distortion correction image. The application unevenness inspection apparatus, wherein the projection vector calculation means obtains a projection vector in a direction perpendicular to the application direction by obtaining a sum of pixel values of pixels arranged in the application direction in the geometrically corrected image.
A coating unevenness inspection apparatus according to a third aspect of the present invention is the coating unevenness inspection apparatus according to the first or second aspect, further comprising pass / fail determination means for performing pass / fail determination by comparing the projection vector with an inspection standard. It is what I did.

According to the coating unevenness inspection apparatus of the first aspect of the present invention, the coating material is transported in the direction perpendicular to the coating direction for the sub-scanning in the imaging by the transport unit, and the coating direction for the main scanning in the imaging. The applied object is imaged by a line sensor camera arranged in the conveyance path so that the main scanning direction is inclined to obtain an imaged image, and the pixel values of the pixels arranged in the application direction in the imaged image by the projection vector calculation means The sum is obtained and a projection vector perpendicular to the coating direction is obtained. That is, the main scanning direction of the line sensor camera is inclined with respect to the application direction of the application, and the integration is performed, so that the influence on the inspection performance can be reduced. Accordingly, there is provided a coating unevenness inspection apparatus capable of reducing the influence on the inspection performance with respect to error factors existing in both the scanning direction (main scanning direction) of the line sensor camera and the conveyance direction (sub-scanning direction) of the inspection target. Is done.
According to the coating nonuniformity inspection apparatus according to claim 2 of the present invention, the geometric distortion of the captured image is corrected by the geometric distortion correcting means to obtain a geometric distortion corrected image. The total sum of the pixel values of the pixels arranged in the application direction is obtained, and a projection vector perpendicular to the application direction is obtained. A projection vector can be easily obtained from the geometric distortion corrected image. Further, since a geometric distortion corrected image is obtained, it can be suitably used for displaying on a monitor screen.
A coating unevenness inspection apparatus according to a third aspect of the present invention is the coating unevenness inspection apparatus according to the first or second aspect, further comprising pass / fail determination means for performing pass / fail determination by comparing the projection vector with an inspection standard. It is what I did. According to the present invention, it is possible to automatically determine pass / fail.

Next, embodiments of the present invention will be described with reference to the drawings. 1 to 4 are diagrams showing the configuration of the coating unevenness inspection apparatus of the present invention. 1 shows the overall configuration as a pictorial diagram, FIG. 2 shows the overall configuration as a block diagram, and FIG. 3 shows the conveyance direction (sub-scanning direction), the application direction, the extension direction of the line sensor camera and the light source (main scanning direction). 4 shows a configuration in which the number of installed line sensor cameras in FIG. 3 is increased.
1 to 4, 1 is a line sensor camera, 2 is a white fluorescent lamp, 3 is a roller conveyor, 4 is a processing unit, 5 is a server, and 100 is a color filter.

  The color filter 100 is an example of an application and is an inspection object. The color filter 100 has a configuration in which a plurality of strip-like regions (stripes) that transmit RGB (red, green, blue) colors are arranged in the order of RGB in the width direction. Each color stripe in the color filter is formed by a coating film obtained by applying and drying each color ink discharged from the inkjet head. Whether or not the coating film is uniformly formed is important in the quality of the color filter. In an inkjet head, there may be a case where the uniformity of the discharge amount at a plurality of nozzles cannot be obtained. As shown in FIG. 9, there is a coating unevenness caused by this, and the coating unevenness is continuous and conspicuous in the stripe direction, which can be a serious defect in the color filter. The coating unevenness inspection apparatus of the present invention inspects whether or not the degree of such coating unevenness is acceptable.

  The line sensor camera 1 is an image pickup means for picking up an image of the color filter 100 that is an inspection object to obtain a picked-up image. As shown in FIGS. 3 and 4, in the coating unevenness inspection apparatus of the present invention, the direction of scanning (main scanning) by the line sensor camera 1 for imaging is inclined with respect to the coating direction in the color filter 100. . The color filter 100 is conveyed by the roller conveyor 4. At this time, the color filter 100 is conveyed so that the application direction in the color filter 100 is a direction perpendicular to the conveyance direction (sub-scanning direction). Therefore, if the angle formed by the coating direction and the main scanning direction is θ, the angle formed by the main scanning direction and the sub-scanning direction is (90 ° −θ). In other words, the line sensor camera 1 is installed such that its main scanning direction is (90 ° −θ) with respect to the sub-scanning direction.

  The reason why the main scanning direction is inclined with respect to the coating direction in the coating unevenness inspection apparatus of the present invention exists in both the scanning direction (main scanning direction) of the line sensor camera 1 and the conveyance direction (sub-scanning direction) of the inspection object. This is to reduce the influence of the error factor on the inspection performance. Regarding the main scanning direction, there are error factors such as individual differences of the line sensor camera 1, variations in sensitivity of the light receiving elements, variations in luminance in the longitudinal direction of the light source, and the like. Further, there are error factors in the sub-scanning direction such as vibration during conveyance, luminance fluctuation due to flickering of the light source, and the like. In the coating unevenness inspection apparatus of the present invention, a process (to be described later) for obtaining a projection vector perpendicular to the coating direction by accumulating the pixel values of the image pickup pixels in the coating direction is performed. The accuracy of the value of can be increased.

  The inclination angle θ may be, for example, about 20 pixels, that is, about 20/4096 radians (several degrees) when the number of pixels (number of light receiving elements) of the line sensor camera 1 is 4096 pixels. If the angle θ is too large, the width of the imaging area by the line sensor camera 1 becomes small, which is not preferable. Therefore, it is preferable to set the angle θ to a small value under a condition where the effect can be obtained.

  The line sensor camera 1 includes CCD (charge coupled device), line sensor elements such as MOS (metal oxide semiconductor), amplifiers, drive circuits, A / D (analog to digital) converters, memories, input / output circuits, and imaging. A well-known line sensor camera including a lens (imaging lens), a housing, and the like can be used. The output of the line sensor camera 1 may be either an analog signal or a digital signal.

  The color filter 100 has a configuration in which a plurality of strip-like regions (stripes) that transmit RGB (red, green, blue) colors are arranged in the order of RGB in the width direction. The line sensor camera 1 distinguishes and detects each color stripe. Therefore, as the line sensor camera 1 in the coating unevenness detection apparatus of the present invention, an R color line sensor camera having sensitivity only for the R color, a G color line sensor camera having sensitivity only for the G color, and only the B color. There are three types of line sensor cameras for B color that have sensitivity. The line sensor camera for each color has sensitivity to only each color, for example, by mounting a color filter that transmits each color.

  Each of the three types of line sensor cameras is independent, and a plurality of line sensor cameras are arranged in parallel in a direction substantially perpendicular to the transport direction. In the example illustrated in FIG. 3, the line sensor camera 1 includes three line sensor cameras for each of RGB colors. The configuration shown in FIG. 3 is suitable for inspecting macroscopic (wide area) coating unevenness because a wide area is imaged with a minimum number of units. In the example shown in FIG. 4, the line sensor camera 1 has a configuration in which a plurality of sets of line sensor cameras each including three RGB color sensors are arranged in parallel. The configuration shown in FIG. 4 is suitable for inspecting microscopic (narrow area) coating unevenness because it captures images with high resolution.

  The white fluorescent lamp 2 is an illumination means for the line sensor camera 1 to image the color filter 100 with transmitted light. The white fluorescent lamp 2 is arranged at a position where the imaging optical axis of the line sensor camera 1 extends through the imaging area of the color filter 100 (a position facing the line sensor camera 1). The white fluorescent lamp 2 is suitable as an illuminating means for the line sensor camera 1 in which the imaging region is similarly linear because the light emitting region is linear. The light emitted by the white fluorescent lamp 2 includes light of each RGB color so that the line sensor camera 1 can capture each color of RGB.

  Electric power is supplied to the white fluorescent lamp 2 by a high frequency lighting power source. When the exposure time of the line sensor camera 1 is taken as a scale by the combination of the phosphor afterglow characteristics and the high-frequency lighting in the white fluorescent lamp 2, there is substantially no variation in the amount of radiant light in the illumination means 2. As shown in FIG. 2, the high-frequency lighting power supply receives a signal for controlling its operation (power on / off, power supply amount, etc.) from the processing unit 4.

  The roller conveyor 3 is a transport unit that transports the color filter 100 as a coating material in a direction perpendicular to the coating direction for sub-scanning in the imaging of the line sensor camera 1. As shown in FIG. 2, the roller conveyor 3 includes a mechanical system (the roller conveyor itself) and a control system thereof. The mechanical system includes a plurality of rollers that transmit a force for conveying the color filter 100, a motor that rotationally drives the plurality of rollers, and the like. The control system is a PLC (programmable sequence controller). The PLC controls the mechanical system and outputs data related to the conveyance state (conveyance speed, conveyance distance, etc.) to the processing unit 4 connected to the PLC. Further, the PLC 20 inputs data related to the conveyance control (operation stop, etc.) of the web 100 from the processing unit 4. The processing unit 4 inputs data from the PLC and reads a captured image synchronized with the conveyance based on the data related to the conveyance state.

  The processing unit 4 includes an image storage unit that inputs an analog signal or a digital signal output from the line sensor camera 1 and stores it as a captured image (input image), that is, an image memory. The processing unit 4 performs data processing on the image stored in the image memory. In addition, data processing related to settings, operations, and the like related to the image input device, data processing related to the user interface, and the like are performed.

  The data processing performed by the processing unit 4 includes a projection vector calculation for obtaining a projection vector in a direction perpendicular to the application direction by obtaining a sum of pixel values of pixels arranged in the application direction of the color filter 100 in the captured image stored in the image memory. Processing is included. The captured image is an image having skew distortion as shown in the left side of FIG. 5 or FIG. 7, and the application direction is a predetermined angle with respect to the coordinate axis of the image. The application direction in the captured image is determined by the application direction of the color filter 100 conveyed by the roller conveyor 3 and the main scanning direction of the line sensor camera 1. That is, the angle θ formed by the coating direction and the main scanning direction described above.

  In order to obtain a projection vector directly from a captured image, first, the coordinates of pixels existing in the direction of the angle θ are assumed by the number of pixels of main scanning belonging to the color filter 100 in the captured image. For example, when the coordinates of the pixel of the main scanning (y axis) to which it belongs are (x, 1), (x, 2)... (X, m), the coordinates of the pixel existing in the direction of the angle θ are ( x + tan θ, 1), (x + 2 tan θ, 2)... (x + m tan θ, m). Only pixels belonging to the color filter 100 in the coordinates are targeted. The boundary of the color filter 100 in the captured image is determined by the magnitude of the pixel value. Outside of the boundary, the pixel value increases because the light emitted from the white fluorescent lamp 2 directly reaches the line sensor camera 1 without being interrupted.

  Next, the pixel value of the pixel at the coordinates is calculated by applying a known interpolation method (described later) using the pixel values of neighboring pixels that actually exist in the captured image. Pixel values are calculated for all assumed pixels and the sum is obtained. Thereby, the value of one element in the projection vector is obtained. By performing this calculation for the number of sub-scanning pixels belonging to the color filter 100, that is, x = 1 to n, the values of all the elements are obtained, and the projection vectors (S1, S2,... Sn) are obtained. can get.

The processing unit 4 performs pass / fail judgment processing based on the obtained projection vector. That is, pass / fail judgment is performed by comparing the projection vector with the inspection standard. For example, it is assumed that the inspection standard includes two upper limit vectors and lower limit vectors. Then, even if one of the elements of the projection vector is out of the range of the upper limit value and the lower limit value given by the corresponding elements of the upper limit vector and the lower limit vector, it is determined as defective.
The server 5 inputs the result of the pass / fail judgment process in the processing unit 4 and stores it as quality data of the color filter 100. In addition, when the defect is determined, processing such as outputting a command to stop the operation is performed including the related manufacturing apparatus.

  The processing in the processing unit 4 can be configured not to obtain directly as described above but to obtain a projection vector after correcting the skew distortion, which is the geometric distortion of the captured image. A geometric distortion corrected image can be obtained by this correction, so that it can be suitably used for displaying on a monitor screen. A process for obtaining a projection vector after correcting the skew distortion will be described with reference to FIGS. FIGS. 5 and 6 show, as pictorial diagrams, processing in a captured image by the imaging system shown in FIG. 3 for inspecting macroscopic (wide area) coating unevenness. FIGS. 7 and 8 show, as pictorial diagrams, processing in a captured image by the imaging system shown in FIG. 4 for inspecting microscopic (narrow region) coating unevenness.

  The captured image (input image) has skew distortion as shown in FIG. The actual color filter 100 is rectangular, but in the captured image, it is a parallelogram. The processing unit 4 performs affine transformation to correct skew distortion in the captured image. A geometric distortion corrected image (post-conversion image) can be obtained by the affine transformation. The affine transformation formula in this case is u = x + ytan θ, v = y, as shown in FIG. Here, the coordinates (x, y) before conversion, the coordinates (u, v) after conversion, and the angle θ formed by the application direction and the main scanning direction are used.

  Since the coordinates (x, y) of the pixels before conversion are the positions of the pixels in the captured image, they are coordinates (lattice points) arranged in a matrix at equal intervals. On the other hand, the coordinates (u, v) of the pixel after conversion are not grid points. Therefore, the processing unit 4 performs an interpolation process for calculating the pixel value of the pixel at the lattice point based on the pixel value of the pixel at the converted coordinates (u, v). Of course, the pixel value of the pixel at the coordinate (u, v) after conversion is equal to the pixel value of the pixel at the coordinate (x, y) before conversion. A known interpolation method can be applied to the interpolation process. For example, nearest neighbor interpolation, bi-linear interpolation, and bi-cubic interpolation can be applied. In particular, the bicubic interpolation method with less information loss is preferable.

  In this way, the processing unit 4 performs a process of obtaining a projection vector for an image (geometric distortion corrected image) having a pixel value of a lattice point by correcting the skew distortion and applying the interpolation method. The geometric distortion corrected image is composed of pixels whose coordinates are lattice points, and the y-axis direction (see FIG. 6) in the geometric distortion corrected image is the application direction of the color filter 100. Therefore, the process of obtaining the projection vector is simple. That is, the sum of pixel values (luminance values) of pixels belonging to the color filter 100 and arranged in the coating direction in the geometric distortion corrected image is obtained. Thereby, the value of one element in the projection vector is obtained. By performing this calculation for the number of sub-scanning pixels, the values of all elements are obtained, and the projection vectors (S1, S2,... Sn) are obtained.

The processing in the processing unit 4 for inspecting macroscopic (wide area) coating unevenness illustrated in FIGS. 5 and 6 has been described above. Since the processing in the processing unit 4 for inspecting microscopic (narrow region) coating unevenness shown in FIGS. 7 and 8 is basically the same, the description thereof is omitted.
Since there are three types of line sensor camera 1, an R color line sensor camera, a G color line sensor camera, and a B color line sensor camera, an R color captured image, a G color captured image, There are three types of B color captured images. Accordingly, there are three types of projection vectors: R color projection vector, G color projection vector, and B color projection vector.
Further, in the inspection of microscopic (narrow area) coating unevenness, as shown in FIGS. 7 and 8, since the color filter 100 is generally arranged in the same color every three lines, as shown in FIGS. A stripe of the same color as the color filter mounted on the sensor camera 1 is a bright portion, and a stripe of other colors is a dark portion.

It is a figure which shows the whole structure in the coating nonuniformity inspection apparatus of this invention as a pictorial diagram. It is a figure which shows the whole structure in the coating nonuniformity inspection apparatus of this invention as a block diagram. It is a figure which shows the relationship between the conveyance direction (subscanning direction) of a macroscopic application nonuniformity detection imaging system in the application nonuniformity inspection apparatus of this invention, a coating direction, and the extension direction (main scanning direction) of a line sensor camera and a light source. It is a figure which shows the relationship between the conveyance direction (sub-scanning direction) of the microscopic coating nonuniformity detection imaging system in the coating nonuniformity inspection apparatus of this invention, a coating direction, and the extension direction (main scanning direction) of a line sensor camera and a light source. The process (the 1) in the captured image for test | inspecting macro application | coating nonuniformity is shown as a pictorial diagram. The process (the 2) in the picked-up image for test | inspecting a macro application | coating nonuniformity is shown as a pictorial diagram. The process (the 1) in the picked-up image for test | inspecting micro application | coating nonuniformity is shown as a pictorial diagram. The process (the 2) in the picked-up image for test | inspecting micro application | coating nonuniformity is shown as a pictorial diagram. It is explanatory drawing of the coating nonuniformity which arises due to the nonuniformity of the discharge amount in the nozzle which exists in multiple numbers in an inkjet head. It is explanatory drawing of the conventional imaging system which made the generation | occurrence | production direction of the coating nonuniformity in a color filter parallel to the scanning direction of a line sensor camera.

Explanation of symbols

1 Line Sensor Camera 2 White Fluorescent Light 3 Roller Conveyor 4 Processing Unit 5 Server 100 Color Filter

Claims (3)

Transport means for transporting the coating material in a direction perpendicular to the coating direction for sub-scanning in imaging;
A line sensor camera that is arranged in the path of conveyance so that the main scanning direction is inclined with respect to the application direction for main scanning in the imaging, and obtains a captured image by imaging the application;
A projection vector calculation means for obtaining a projection vector in a direction perpendicular to the application direction by obtaining a sum of pixel values of pixels arranged in the application direction in the captured image;
A coating unevenness inspection apparatus comprising: The coating unevenness inspection apparatus according to claim 1, further comprising geometric distortion correction means for correcting a geometric distortion of the captured image to obtain a geometric distortion correction image, wherein the projection vector calculation means includes the application direction in the geometric correction image. A coating unevenness inspection apparatus characterized in that a sum of pixel values of pixels arranged in a line is obtained to obtain a projection vector perpendicular to the coating direction. The coating unevenness inspection apparatus according to claim 1, further comprising: a quality determination unit that performs quality determination by comparing the projection vector with an inspection standard.

Images