JP4246319B2 - Illumination unevenness measuring method and measuring apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is based on image data captured by a line sensor camera, for example, whether the circuit pattern shape of a printed circuit board is good, the width of a transparent electrode formed on a plasma display panel or the like, the presence or absence of chipping, etc. The present invention relates to an illumination unevenness measuring method and an illumination unevenness measuring apparatus of an illumination device in an image recognition inspection system inspected by recognition processing.
[0002]
[Prior art]
In recent years, the density of circuit patterns has been increased more and more with the reduction in the size and size of printed circuit boards, and methods for automatically inspecting the shape of circuit patterns have been actively developed accordingly. In an image recognition inspection system that embodies one inspection method for that purpose, the circuit pattern is imaged by a recognition camera while irradiating illumination light to the circuit pattern of the printed circuit board from the illumination device, and an image of the captured image data Based on the recognition processing, the quality of the circuit pattern shape, through-hole formation position, hole diameter, and the like is inspected. In such an image recognition inspection system, it is necessary to irradiate the entire irradiation surface with uniform illuminance by illumination light from the illumination device.
[0003]
This is because, for example, in a through hole image recognition inspection, a circuit board is irradiated with light from one side, and light transmitted through the through hole is captured by a recognition camera installed on the other side. This is because if there is a variation in the illuminance due to the illumination light, the diameter of the through hole in the image data captured according to the brightness of the illumination light will be different, and a good through hole will be erroneously determined to be defective. On the other hand, in the circuit pattern image recognition inspection, the reflected light from the circuit pattern of the illumination light is captured as an input image into the recognition camera. This results in the inconvenience of erroneously determining it as a defective product even though it is a non-defective product.
[0004]
Therefore, conventionally, the illumination light of the illumination device is regularly measured to determine whether there is illumination unevenness. In the transmission illumination system that uses the transmitted light from the imaging target as an input image, the illumination device and the recognition camera The light of the illumination light is directly taken in as an input image without placing an imaging target between the two.
[0005]
On the other hand, in a reflective illumination system that captures the reflected light from the imaging target of the illumination light of the illuminator as an input image, the illumination light cannot be captured directly as an input image into the recognition camera when measuring illumination unevenness. Irradiation of the pattern is taken and the reflected light is taken as an input image, and the illumination unevenness is measured.
[0006]
[Problems to be solved by the invention]
However, in such a method for measuring illumination unevenness, it is a prerequisite that the entire test pattern has an accurate and uniform reflectance. However, it is extremely difficult to obtain such a test pattern. If there is a variation in reflectance, the variation in reflectance appears as a luminance variation in the pixels of the image data. Therefore, since it is impossible to determine whether the brightness of the pixel brightness is caused by the variation in the reflectance of the test pattern or the variation in the brightness of the illumination light, there is a problem that the illumination unevenness cannot be measured accurately. is there.
[0007]
In addition, conventionally, adjustment is performed so as to eliminate illumination unevenness while replacing the light source of the illumination device based on the measurement result of illumination unevenness, and there is a problem that a complicated adjustment work is required periodically.
[0008]
Accordingly, the present invention has been made to solve the above-described conventional problems, and illumination of an illumination device in an image recognition inspection system having a reflection illumination system that takes reflected light from an imaging target as an input image into a line sensor camera. An object of the present invention is to provide an illumination unevenness measuring method and an illumination unevenness measuring apparatus capable of accurately measuring unevenness.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, the present invention irradiates a measurement object with light, and has an image pickup means having a plurality of image pickup elements arranged in a line in the arrangement direction of the image pickup element with respect to the measurement target. imaging the measurement object while relatively moving, after a plurality save the images data obtained by the imaging, a plurality of image data each obtained by the storage, extracts image data of a particular same position of the measurement object The variation in illuminance is obtained from the illuminance distribution in the arrangement direction in the extracted image data .
[0010]
Lighting unevenness measuring method this is to be measured, while relatively moving toward the arrangement direction of the imaging device, and stores the image data obtained by capturing a reflection surface of the measurement target, a specific reflection surface in the image data It is possible to extract only image data obtained by imaging the same part, and reflect from a specific part in the measurement object , that is, a part having the same reflectance, and change in accordance with the illuminance distribution of the illumination device with the relative movement. Since only the image data obtained by imaging the light can be extracted and the illuminance distribution in the relative movement direction of the lighting device can be obtained from the brightness of the image data, the adverse effect due to the variation in the reflectance over the entire surface to be measured is completely excluded. Thus, the illumination unevenness of the illumination device can be accurately measured.
[0011]
In the above invention, as a measurement object, using a test pattern having a measurement area part having a reflectance that is partially different from that of the other part, image data obtained by imaging the measurement area part among the image data accumulated in each image sensor. It is preferable to extract and to calculate the average value of the luminance of each pixel obtained by dividing the extracted image data for each image sensor.
[0012]
As described above, as a measurement object, a test pattern having a measurement area part having a reflectance different from that of another part in part, for example, a test pattern having a white stripe-shaped measurement area on a black background, is used to accumulate an image. If a certain number is extracted in order from the pixel having the highest luminance among the pixels corresponding to the respective imaging elements in the data, the image data of each extracted pixel is obtained by imaging the measurement region portion. Thereby, it is possible to easily and accurately extract image data obtained by capturing a specific identical portion of the test pattern from the accumulated image data.
[0013]
In the above invention, the extraction of the image data obtained by imaging the specific same portion of the reflecting surface among the image data stored in each image sensor, the relative movement speed with respect to the measurement target of the line sensor camera, the image capture period of the line sensor camera, It is preferable to perform the calculation based on the above.
[0014]
Thereby, the specific same location of a measuring object can be easily calculated | required by a calculation.
[0015]
In addition, by applying the present invention , reflected light from an inspection object of illumination light of an illuminating device that is integrated with the line sensor camera is taken as an input image into each image sensor of the line sensor camera, and the image data An image recognition inspection method for inspecting the quality of an object to be inspected based on recognition processing, the illuminance distribution data obtained by the above-described illumination unevenness measuring method being stored in a storage means, and the line sensor camera being The inspection object was imaged with each image sensor while being relatively moved in a direction orthogonal to the arrangement direction of the image sensors with respect to the inspection object, and image data obtained by the imaging was read from the storage unit It can correct | amend with illuminance distribution data, and can comprise so that to-be-inspected object may be test | inspected by the corrected image data .
[0016]
In this image recognition inspection method, the luminance of each pixel of image data obtained by imaging with a line sensor camera is corrected based on the illuminance distribution data, and the corrected image data has uneven illumination in the illumination device. Even if it is, the adverse effect is removed. Since the inspection object is inspected based on the image data, the inspection can be performed very accurately.
[0017]
On SL inspection object may be a transparent electrode formed of a transparent conductive film on the substrate.
[0018]
Thereby, since the brightness of the image data is corrected based on the illuminance distribution data of the measurement result obtained by measuring the illumination unevenness in advance, even if illumination unevenness exists in the illumination device, it is possible to perform an inspection that removes the adverse effects. For this reason, the present invention is applied to the inspection of the pattern width and chipping of the transparent electrode, which has conventionally been difficult to inspect by image recognition processing, and has a remarkable effect of obtaining an accurate inspection result by image recognition.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view of a main part showing an image recognition inspection system provided with an illumination unevenness measuring apparatus that embodies the illumination unevenness measuring method according to the first embodiment of the present invention. In the figure, an inspection table 1 holds a printed circuit board (not shown) to be inspected or a test pattern 2 for measuring uneven illumination at a predetermined mounting position. Above the inspection table 1, a ball screw 7 rotated by a servo motor 4 and a guide shaft 8 are horizontally disposed in parallel to each other in the X direction, and the carrier member 3 screwed into the ball screw 7 is Then, it is moved in the X direction at a predetermined constant speed while sliding on the guide shaft 8.
[0020]
A line sensor camera 9 and an illumination device 10 are attached to the carrier member 3 in combination at predetermined positions. The line sensor camera 9 is a one-dimensional image pickup device in which CCDs (not shown) are arranged in a line, and includes an angle changing motor 18 and is configured to be able to rotate 90 ° around a vertical line. Yes. At the time of uneven illumination measurement, the line sensor camera 9 is arranged in a direction where the CCD arrangement direction coincides with the X direction and at a position where the measurement target set on the inspection table 1, that is, the test pattern 2 can be imaged. On the other hand, at the time of image recognition inspection, the line sensor camera 9 is rotated by 90 °, and the inspection object set on the inspection table 1, that is, the printed circuit board, is imaged with the CCD arrangement direction coinciding with the Y direction. It is arranged at a position where it can. The illuminating device 10 is positioned so that the arrangement direction of the light sources 11 such as light emitting diodes coincides with the X direction, and the illumination light from the light sources 11 is easily incident on the line sensor camera 9 after being reflected by the inspection object. ing.
[0021]
FIG. 2 is a block diagram showing the image recognition inspection system. In the figure, the same components as those in FIG. 1 are denoted by the same reference numerals. A printed circuit board 12, which is an object to be inspected, is set on the inspection table 1, and this printed circuit board 12 is irradiated with illumination light L1 from the illumination device 10, and reflected light L2 from the printed circuit board 12 is input image. Is taken into the CCD 13 of the line sensor camera 9.
[0022]
The control unit 14 controls the entire system in accordance with a preset control program. The control unit 14 controls the rotation of the servo motor 4 shown in FIG. Then, the angle of the line sensor camera 9 is set to a predetermined angle. In addition, when the test pattern 2 is set on the inspection table 1, the control unit 14 outputs the image data captured from the line sensor camera 9 to the illumination unevenness measurement system 19 and also prints the printed circuit board on the inspection table 1. When an inspection object such as 12 is set, the image data taken from the line sensor camera 9 is controlled to be output to the image recognition inspection system 20.
[0023]
The illumination unevenness measurement system 19 includes an image data storage unit 21 that temporarily stores the image data of the test pattern 2 input from the control unit 14, and each pixel corresponding to each CCD in the image data of the image data storage unit 21. A pixel extraction unit 22 for extracting a predetermined number of pixels having a high luminance, a luminance average value calculation unit 23 for calculating an average luminance value of each of the extracted pixels for each CCD 13a to 13h, An illuminance distribution determination unit 24 that determines the illuminance distribution of the lighting device 10 based on the calculated average value of luminance, and an illuminance distribution data storage unit 27 that stores illuminance distribution data that is a determination result of the illuminance distribution determination unit 24. It is configured with.
[0024]
On the other hand, the image recognition inspection system 20 temporarily stores the image data of the printed circuit board 12 of the inspection object input from the control unit 14, and the image stored in the image data storage unit 28. An image correction unit 29 that corrects the luminance of each pixel of the data based on the illuminance distribution data stored in the illuminance distribution data storage unit 27, and image data of a specified fixed region of the image correction unit 29 is extracted. The image processing means 30 for measuring the shape of the circuit pattern to be inspected and the measurement result of the image measurement processing of the image processing means 30 are compared with a predetermined criterion value to determine whether the shape of the inspection object is good or bad. And a determination processing means 31 for determination.
[0025]
Next, measurement of illumination unevenness of the illumination device 10 in the image recognition inspection system will be described with reference to the explanatory diagram of FIG. FIG. 3A shows one row of the test pattern 2, the horizontal parallel lines indicate the black background portion 2a, and the black background portion 2a has the measurement region portion 2b formed in a stripe shape on a white background with high reflectivity. Yes. As the test pattern 2 used in the illumination unevenness measuring method of the present invention, it is not necessary for the black background portion 2a and the measurement region portion 2b to have uniform reflectance. However, the black background portion 2a needs to have at least one measurement region portion 2b having a predetermined width or more, and the measurement region portion 2b of the line sensor camera 9 is set when the inspection table 1 is set. It is preferable to provide a direction orthogonal to the direction in which the CCDs 13 are arranged.
[0026]
The test pattern 2 is set on the inspection table 1 in an arrangement in which the direction of the measurement area 2b is orthogonal to the direction in which the CCDs 13 are arranged in the line sensor camera 9, that is, the moving direction (X direction in FIG. 1). In this embodiment, a case where the measurement area 2b having a large width is used will be described. The line sensor camera 9 captures an image of the test pattern 2 while irradiating the test pattern 2 with the illumination light L1 from the illumination device 10, and the servomotor 4 in FIG. 9 is moved in the arrangement direction of the CCD 13. Thereby, the illuminance distribution data in the X direction at a certain predetermined position in the Y direction can be obtained. Note that two-dimensional illuminance distribution data can be obtained by sequentially changing the position in the Y direction.
[0027]
Now, it is assumed that the CCD 13 of the line sensor camera 9 captures an image while moving in the orthogonal direction across an intermediate portion of the measurement region 2b in the test pattern 2 as indicated by an arrow in FIG. In the line sensor camera 9, eight CCDs 13a to 13h are arranged in a line in this embodiment, and the first CCD 13a is opposed to the measurement region 2b as an initial position. It shall be moved toward the direction.
[0028]
The line sensor camera 9 takes in the electric charges by the imaging accumulated in the CCDs 13a to 13h at a predetermined cycle. S1 to S8 shown in FIG. 3A indicate the movement positions of the line sensor camera 9 at the timing of taking in the accumulated charges of the CCDs 13a to 13h, respectively. When the line sensor camera 9 moves from the S1 position to the S8 position, image data 32 as shown in FIG. 3B is obtained. As described above, the line sensor camera 9 is moved in the arrangement direction of the CCDs 13a to 13h as the imaging elements, that is, in the X direction, so that the two-dimensional as shown in FIG. Image data 32 is obtained.
[0029]
The control unit 14 outputs the image data 32 shown in FIG. 3B captured from the line sensor camera 9 to the image data storage unit 21 and temporarily stores it. The pixel extraction means 22 extracts and reads out a predetermined number of pixels from the image data storage means 21 in order from the highest brightness among the pixels for each CCD in each image data. Portions of the image data 32 divided by the vertical broken lines in FIG. 3B are pixels obtained by imaging the CCDs 13a to 13h, and a predetermined number of pixels having high luminance among the pixels for the CCDs 13a to 13h. Are pixels that image the measurement region portion 2b in this embodiment. Therefore, each of the predetermined number of pixels extracted for each of the CCDs 13a to 13h is obtained by imaging a specific identical portion in the measurement region 2b.
[0030]
Subsequently, the luminance average value calculation means 23 calculates the average value of the luminance for each of the extracted pixels for each of the CCDs 13a to 13h. That is, the average value of each luminance in the pixels within each range indicated by the arrow in FIG. Furthermore, the illuminance distribution determination means 24 determines the illuminance distribution in the array direction of the CCDs 13a to 13h in the illumination device 10 based on the average value of the luminance calculated for each of the CCDs 13a to 13h, and the illuminance that is the determination result The distribution data is stored in the illuminance distribution data storage means 27.
[0031]
In this illumination unevenness measuring method, the test pattern 2 is imaged while moving the line sensor camera 9 in the direction in which the CCDs 13a to 13h are arranged, and for each CCD 13a to 13h of the pixels in the imaged image data. Since a predetermined number of pixels with high luminance are extracted, the pixels extracted for each of the CCDs 13a to 13h are obtained by imaging a specific identical part in the test pattern 2, that is, a part having the same reflectance. Therefore, even if there is a variation in the reflectance of the test pattern 2, the adverse effects due to the variation in the reflectance of the test pattern 2 are completely excluded by imaging the specific same portion of the test pattern 2 with the CCDs 13 a to 13 h. It is possible to accurately measure illumination unevenness caused by variations in luminous intensity of each light source 11 of the illumination device 10.
[0032]
In the above description, the case where the control unit 14 controls the rotation of the servo motor 4 to move the line sensor camera 9 and the illumination device 10 has been described. However, the line sensor camera 9 and the illumination device 10 are fixed and the inspection table is fixed. Even if 1 is moved in the X direction, the illumination unevenness can be measured in the same manner as described above, and the adverse effect on the image data due to vibration during the movement of the line sensor camera 9 can be excluded, which is preferable.
[0033]
In addition, since the image recognition inspection system of the above-described embodiment is integrally provided with the illumination unevenness measurement system 19, for example, the illumination unevenness is measured at a rate of once before daily operation, and the measurement result is obtained. When the illuminance distribution data is different from the previous day, the illuminance distribution data is updated and stored in the illuminance distribution data storage means 27. Thereby, the illuminance distribution data always accurately indicates the state of the lighting device 10 at that time. After the measurement of the illumination unevenness, an image recognition inspection of the inspection object is performed. Next, an image recognition inspection method for the inspection object will be described.
[0034]
The operator sets the illumination unevenness measurement by operating the input device (not shown) and performs the above-described illumination unevenness measurement. Then, the operator sets the image recognition inspection using the input device and takes out the test pattern 2 from the inspection table 1. After that, setting is made such that, for example, the printed circuit board 12 of the inspection object is sequentially set on the inspection table 1. The controller 14 drives the angle changing motor 18 to set the direction of the line sensor camera 9 so that the CCD arrangement direction coincides with the Y direction, and then drives the servo motor 4 to drive the line sensor camera. 9 and the illumination device 10 are moved at a constant speed in the X direction. As a result, the line sensor camera 9 outputs image data obtained by imaging the printed circuit board 12 with the CCDs 13 a to 13 h to the control unit 14. The control unit 14 temporarily stores the two-dimensional image data obtained by accumulating the image data in the image data storage unit 28 of the image recognition inspection system 20.
[0035]
Next, the control unit 14 instructs the illuminance distribution data storage unit 27 to send the illuminance distribution data to the image correction unit 29. The image correction unit 29 corrects each luminance of the image data read from the image data storage unit 28 based on the illuminance distribution data. Subsequently, the image processing unit 30 extracts image data of a specified fixed area from the corrected image data, reads it from the image correction unit 29, and measures the width and length of the circuit pattern in the image data. Perform image processing. Further, the determination processing unit 31 compares the image processing result obtained by the image measurement of the image processing unit 30 with a predetermined determination reference value to determine whether the circuit pattern is good or bad. The image recognition result is obtained by converting the luminance of each pixel of the image data captured by the line sensor camera 9 into image data corrected based on the illuminance distribution data of the illumination device 10 and accurately corresponding to the shape of the circuit pattern. Since recognition processing is performed based on the data, a very accurate inspection can be performed.
[0036]
Next, an illumination unevenness measuring method according to the second embodiment of the present invention will be described. Although this embodiment is different from the first embodiment in that it is not shown in the figure, it is only a configuration in which an image conversion means is provided instead of the pixel extraction means 22 in FIG. The image conversion means stores image data based on the moving speed v of the line sensor camera 9 calculated from the input data from the encoder 33 that detects the rotation of the servo motor 4 and the preset image capture period T and resolution a. The coordinates D1 (x, y) of the image data 32 temporarily stored in the means 21 are converted into image data 34 as shown in FIG. 4 by performing an operation based on the following equation (1). That is, if the coordinate of the converted image data 34 is D2 (x, y),
D2 (x, y) ≈D1 [x, y + (a × x) / (v × T)] (1)
The approximate expression of is calculated. As shown in FIG. 4, the image data 34 thus converted is arranged in a line with pixels obtained by imaging the specific identical portions of the test pattern 2 with the CCDs 13 a to 13 h respectively corresponding to the arrangement directions of the CCDs 13 a to 13 h. It will be converted as follows. Therefore, since the pixels that image the same portion of the measurement region portion 2b in the test pattern 2 are arranged in a horizontal row, the calculation for calculating the average value of the luminance of the pixels performed thereafter can be simplified.
[0037]
FIG. 5A shows a transparent electrode 37 formed in a precise pattern of a certain width by a transparent conductive film (ITO film). Such a transparent electrode 37 is used in a plasma display, a liquid crystal display, an electroluminescence display, or the like. It is used. The transparent electrode 37 is inspected for width and chipping after being formed by a photoetching method, but image recognition is not employed for this inspection. The reason is that, since the reflectance of the transparent electrode 37 is extremely low, a high contrast of brightness and darkness cannot be obtained when the illumination light is irradiated, so that all the portions of the image data in the portions where the amount of light is low are recognized as defects. It is.
[0038]
On the other hand, in the image recognition inspection system of the above embodiment, the illumination unevenness measurement system 19 measures the illuminance distribution of the illumination device 10 in advance, and the image data 39 of the transparent electrode 37 shown in FIG. Since the image data is converted into image data having luminance corresponding to the resulting illuminance distribution data and the image data is recognized, the image data from which the adverse effects are removed even if the illumination device 10 has uneven illumination. Obtainable. For this reason, the image recognition inspection system is applied to the inspection of the pattern width and chipping of the transparent electrode 37, and has a remarkable effect that an accurate inspection result can be obtained by image recognition.
[0039]
【The invention's effect】
As described above, according to the present invention, even when there is a variation in reflectance over the entire reflection surface to be measured, only the image data obtained by capturing a specific same portion on the reflection surface is extracted, and the luminance in the image data is extracted. it is possible to obtain the illuminance distribution of the illumination device from and exclude the adverse effect of variations in the reflectance of the reflecting surface, it is possible to accurately measure the illumination unevenness of the illumination device.
[Brief description of the drawings]
FIG. 1 is a partial perspective view showing an image recognition inspection system including an illumination unevenness measuring apparatus that embodies an illumination unevenness measuring method according to a first embodiment of the present invention.
FIG. 2 is a block diagram showing the image recognition inspection system.
FIG. 3A is a plan view showing a test pattern in the illumination unevenness measuring method, and FIG. 3B is a diagram showing image data obtained by the imaging means using the test method.
FIG. 4 is a view showing image data obtained by an image conversion step in the illumination unevenness measuring method according to the first embodiment of the present invention.
5A is a cross-sectional view showing a transparent electrode that can be an inspection object of the image recognition inspection system of the present invention, and FIG. 5B is a diagram showing image data obtained by imaging the transparent electrode.
[Explanation of symbols]
2 Test pattern (measurement target)
2b Measurement area 9 Line sensor camera 10 Illumination device 11 Light source 12 Printed circuit board (inspection object)
13, 13a-13h CCD (imaging device)

Claims (4)

And irradiates a light to the measurement object, imaging the measurement object while the imaging means having a plurality of imaging elements arranged in a line are relatively moved in the arrangement direction of the image sensor relative to the measurement object,
After the images data obtained by the imaging plurality saved,
Illumination variation is extracted from each of the plurality of stored image data , and image data of a specific same part of the measurement object is extracted, and illuminance variation is obtained from the illuminance distribution in the arrangement direction in the extracted image data . Unevenness measurement method. The method for measuring illumination unevenness according to claim 1, wherein the measurement object has a measured portion smaller than the size of the imaging means in the arrangement direction of the imaging elements. Image the measurement object for each distance between the centers of adjacent image sensors arranged in a line at an equal pitch,
The method according to claim 1 or 2, wherein at least the number of the image pickup devices stores data obtained by the image pickup. A lighting device;
An imaging unit having imaging elements arranged in a line and fixed to the illumination device;
Moving means for moving the imaging means in the arrangement direction of the imaging elements;
A control unit that irradiates the measurement object with light from the illumination device, and causes the imaging unit to capture a plurality of images of the measurement object while relatively moving the measurement object and the imaging unit;
A process of extracting image data of a specific same portion of the measurement object from each of a plurality of image data obtained by imaging by the imaging unit, and obtaining variation in illuminance from the illuminance distribution in the arrangement direction in the extracted image data An illumination nonuniformity measuring apparatus comprising: means.

Images