JP2004289571A - Pixel defect mask method for imaging device - Google Patents

Abstract

A method for masking a CCD defect from an inspection object, comprising means for automatically identifying a CCD defective pixel in a panel image quality inspection apparatus for inspecting a point defect corresponding to one pixel such as a plasma display.
A plurality of images are taken while moving the position of an imaging device with respect to a measurement target, and one integrated image is created by integrating each image in pixel units. Next, this integrated image is binarized at an arbitrary threshold value.
The position detected by binarization is specified as a defect position of the image pickup device of the image pickup device, the position coordinates are registered, and the registered position is masked and excluded from the inspection target.
[Selection diagram] FIG.

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an inspection apparatus for inspecting a point defect such as a missing or scratched luminous body used in a plasma display or the like.
[0002]
[Prior art]
A conventional inspection apparatus will be described with reference to FIGS. FIG. 1 is a block diagram showing a schematic configuration of a conventional inspection apparatus. FIG. 2 is a flowchart for explaining a conventional inspection processing operation.
[0003]
Reference numeral 11 denotes an object to be measured, 12 denotes an image pickup device including at least an image pickup device such as a CCD and an enlargement unit such as an optical microscope, 13 denotes an image processing unit for performing an inspection process, and 14 denotes a support for fixing the object to be measured 11.
[0004]
The measurement target 11 to be inspected is fixed to the support base 14 in parallel with the surface to be measured and the x-axis direction and the y-axis direction of the support base 14.
[0005]
The image processing unit 13 includes a drive unit (not shown) attached to the imaging device 12 (for example, an x-axis driving unit, a y-axis driving unit, and a focal axis) in order to cause the imaging device 12 to acquire an image of a desired area of the measurement target 11. Drive unit) to drive the imaging device 12. The imaging device 12 is set perpendicular to the measurement target surface of the measurement target 11 (referred to as the z-axis direction, the optical axis direction, or the focal axis direction), captures an image of a desired region of the measurement target, and obtains information of an acquired image. Is output to the image processing unit 13.
[0006]
Although not shown, the image processing unit 13 includes at least a CPU and controls the inspection device according to a predetermined program. The image processing unit 13 performs image processing of the image acquired by the imaging device 12 by the CPU according to a predetermined inspection algorithm.
[0007]
Further, although not shown, the image processing unit 13 controls the imaging device 12 so as to focus the enlargement unit. In addition, the image processing unit 13 includes a display unit (not shown) that displays an image of the inspection target 11 captured by the imaging device 12, a result image analyzed by the image processing unit 13, and inspection result information. In addition, the image processing unit 13 includes an operation device (not shown) for controlling the inspection device in FIG.
[0008]
The measurement target 11 is, for example, a substrate made of glass or the like, on which a luminous body such as a desired phosphor is applied, and is a plasma display, an LCD, or the like.
[0009]
In FIG. 1, first, the measurement target 11 is caused to emit light by well-known means (not shown), and an optical image incident from the measurement target 11 is captured by the imaging device 12. For example, when the measurement target 1 is a plasma display, ultraviolet light is irradiated on the substrate surface to emit a desired phosphor on the plasma display substrate surface. Further, for example, when the measurement target 1 is an LCD, the power is turned on and the backlight is turned on.
[0010]
The imaging device 12 outputs the obtained image to the image processing unit 13, and the image processing unit 13 processes the input image according to the processing flow illustrated in FIG. 2 to detect a point defect. Hereinafter, the processing operation of the prior art of FIG. 2 will be described.
[0011]
Hereinafter, the processing flow of FIG. 2 will be described.
FIG. 2A is a flowchart of the processing operation, and FIG. 2B is a diagram for explaining the processing operation.
[0012]
In step 201, the measurement target 11 is imaged, and an image (original image) is obtained (see (1) in FIG. 2 (b), a blackened portion 211). x-axis direction.
[0013]
In step 202, the brightness level of the acquired image (see FIG. 2 (b) -2: -1: brightness level waveform 213) is, for each pixel, for example, as shown in FIG. The linear portion 220 is differentiated in the x-axis direction, the amount of change in the luminance level of each pixel is obtained, and binarized based on a predetermined threshold value I (FIG. 2B (2) -1: Differentiation). , (2) -2: binarized), and a binarized image (see (2) -2: binarized waveform 214 in FIG. 2B) is obtained. That is, the position of the defect is detected based on the set threshold value I.
[0014]
In step 203, the obtained binarized image is subjected to a well-known labeling process to detect a pixel position having a luminance level change, and the vicinity of the detected pixel position (FIG. 2 (b) (1) A region indicated by a broken line 212 including the region 211 is extracted.
[0015]
Next, in step 204, the luminance level of the region x1 near the extracted pixel position (see the luminance level waveform 215 in the figure enlarged to (3) in FIG. 2B) is compared, and a pixel different from the surroundings is detected as a defect. I do. The lateral width in the figure of (3) is the width of the broken line 212 in the figure of (1) in the x-axis direction. The comparison is performed with reference to the threshold value II, and a pixel area equal to or smaller than the threshold value II is detected as a pixel defect (for example, see Patent Document 1).
[0016]
[Patent Document 1]
JP-A-8-201228 (prior art, FIG. 4)
[0017]
[Problems to be solved by the invention]
In the above-described prior art, for example, when a pixel defect is present on the image pickup device side of an image pickup device such as a CCD, there is a defect that even if the inspected measurement target has no defect, it is erroneously detected as having a defect. That is, as shown in FIG. 3, in an image obtained by imaging a target, an actual defect (hereinafter, referred to as an actual defect) 31 and a pixel defect (hereinafter, referred to as a CCD defect) 32 of an image sensor (CCD) are the same. Therefore, the actual defect 31 and the CCD defect 32 cannot be distinguished. For this reason, the CCD defect 32 was erroneously detected as a real defect of the measurement target 11. In FIG. 3, the size of a pixel defect (actual defect, CCD defect) is emphasized with respect to the size of one screen imaged by the imaging device.
[0018]
An object of the present invention is to provide a defect inspection method that eliminates the above-described defects and that does not make a defect a defect to be measured even if there is a pixel defect on the imaging element side of the imaging device.
[0019]
[Means for Solving the Problems]
In order to achieve the above object, a pixel defect masking method for an image pickup device according to the present invention specifies a pixel position of a CCD defect of an image pickup device such as a CCD in advance, and masks the specified pixel position for erroneous detection. Is to be eliminated.
[0020]
That is, in the pixel defect mask method for an image sensor according to the present invention, a defective pixel of an image sensor is identified in a defect inspection method in which an image of an illuminant is captured by an image sensor, the captured image is processed, and a defect of the illuminant is detected. Then, the position information is registered, and the position of the registered position information is masked.
[0021]
Further, in the pixel defect mask method for an image sensor according to the present invention, the image sensor captures a plurality of images by shifting the luminous body by a predetermined number of pixels, and integrates the luminance levels of the plurality of acquired images. Then, a total sum image is created, and a defective pixel of the image pickup device is specified for the total sum image based on a predetermined threshold value.
[0022]
BEST MODE FOR CARRYING OUT THE INVENTION
First, a means for specifying a pixel position of a CCD defect of an image sensor in advance according to the present invention will be described with reference to FIGS. FIG. 4 is a block diagram showing a schematic configuration of an embodiment of the inspection apparatus of the present invention, and FIG. 5 is a view for explaining an embodiment of a method for specifying a pixel position of a CCD defect of an image pickup device.
[0023]
As shown in FIG. 4, the imaging device 12 is moved, for example, horizontally (x-axis direction) to capture an image of the measurement target 11, and creates a plurality of images separated by a predetermined distance. The predetermined distance for horizontal movement is, for example, one pixel to several pixels.
[0024]
Next, each image is integrated (accumulated) for each pixel to create one image (sum image).
[0025]
In this total image, as shown in FIG. 5, the position of the CCD defect 32 always exists at the same position (32-1, 32-2, 32-3,...), And the actual defect 31 Each time is moved, the position changes (31-1, 31-2, 31-3,...). Here, 310 is a real defect area of the total image, and 320 is a CCD defect area of the total image.
[0026]
Therefore, the brightness level of the sum pixel (the sum brightness level waveform in FIG. 5) is such that the brightness level of the area 310 due to the pixel of the real defect 31 in the sum image does not increase much, and the pixel position of the CCD defect 32 does not increase. Only the luminance level of the region 320 is added, and the luminance value increases as the number of images to be integrated increases, and only the CCD defect 32 is emphasized.
[0027]
Therefore, by binarizing this integrated image with an arbitrary threshold value, the position coordinates of the pixel of the CCD defect can be specified.
[0028]
In FIG. 5, the measurement target is horizontally imaged one pixel at a time from right to left, and the size of a pixel defect (actual defect, CCD defect) is larger than the size of one screen imaged by the imaging device. I emphasize it and draw it large. In addition, the brightness values of the first to third images in the figure are added to the total image in FIG.
[0029]
Further, for the sake of explanation, the total luminance level waveform of FIG. 5 is a right half waveform and a left half waveform divided by a vertical line 501 corresponding to the right half linear part 502 and the left half linear part 503 of the total image, respectively. Waveforms are drawn together.
[0030]
By registering the specified position coordinates and masking the CCD defect during the inspection, it is possible to prevent the CCD defect from being erroneously detected as a defect to be measured during the inspection. Therefore, according to this method, the presence / absence of a defect in the pixel to be measured is not determined at the pixel defect position on the imaging device side, so that erroneous detection due to a pixel defect of the imaging device can be eliminated.
[0031]
The method of the mask processing is generally well known, and is described in, for example, JP-A-2001-175959.
[0032]
Note that the image acquired by the image processing unit 13 can be observed on a display device such as a monitor by outputting the image to the monitor 15, and information on an image processing result and an inspection result can be displayed.
[0033]
In the above embodiment, the imaging device is moved by a predetermined distance to acquire each image for creating an integrated image. However, the measurement target 11 side, for example, the support base 14 may be moved, or a combination thereof may be used. Is self-evident.
[0034]
【The invention's effect】
As described above, conventionally, the defect of the image sensor has also been erroneously detected as a defect of the measurement target. However, according to the present invention, it is possible to remove the defect of the image sensor and inspect the measurement target. It has become possible.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a conventional inspection apparatus.
FIG. 2 is a flowchart for explaining a conventional inspection processing operation.
FIG. 3 is a diagram illustrating a pixel defect of an image sensor and a pixel defect to be measured.
FIG. 4 is a block diagram showing a schematic configuration of an embodiment of the inspection apparatus of the present invention.
FIG. 5 is a view for explaining an embodiment of a method for specifying a pixel position of a CCD defect of the image sensor according to the present invention.
[Explanation of symbols]
11: measurement target, 12: imaging device, 13: image processing unit, 14: support base, 15: monitor, 31, 31-1, 31-2, 31-3, Δ: actual defect, 32, 32- 1, 32-2, 32-3, Δ: CCD defect, 211: black portion, 212: broken line portion, 213: brightness level, 214: binarized waveform, 215: brightness level waveform, 220: straight portion 310, 320: area.

Claims (2)

In the defect inspection method of imaging the luminous body by the imaging element, performing image processing on the captured image, and detecting the defect of the luminous body,
Identify the defective pixel of the image sensor and register its position information,
A pixel defect masking method for an image sensor, wherein a position of registered position information is masked. 2. The pixel defect mask method for an image sensor according to claim 1,
The imaging device acquires a plurality of images by imaging the luminous body by shifting by a predetermined number of pixels,
A sum total image is created by integrating the brightness levels of the obtained plurality of images,
A pixel defect mask method for an image sensor, wherein a defective pixel of the image sensor is specified for the total image based on a predetermined threshold value.

Images