JP2004239733A - Defect detection method and apparatus of screen - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect both of a light defect and a dark defect and reduce a process time for detecting the defects. <P>SOLUTION: A defect detection method has a process for implementing a defect enhancement process for enhancing and detecting the various defects in an image of a captured screen as an object to be inspected, a process for adding a median of gray levels of a displayed image as an offset to the defect enhanced image, a process for calculating statistical data based on pixel intensity of an offset added image and a process for setting threshold values for extracting the light defect and the dark defect based on the statistical data and extracting defect candidates of the light defect and the dark defect based on the statistical data and the threshold values. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention provides a screen defect detection method for automatically and accurately detecting defects such as point defects and spot / mura defects in an inspection process in the production of a display device such as a liquid crystal light valve or a projector which is an application product thereof, and Equipment related.
[0002]
[Prior art]
Conventionally, a point defect detection device that detects a point defect such as a bright spot or a dark spot on a liquid crystal panel, which is a light modulation device, is included in image data captured by an image capturing device due to a noise removal function of a processing device. , The noise of the image capturing device is removed, and the spatial filter for bright point defect enhancement and the spatial filter for dark point defect enhancement are used for the image data from which noise has been removed by the point defect enhancement function of the processing device. The convolution operation is performed to emphasize the point defect portion of the light modulation device. Based on the image data in which the point defect portion of the processing device is emphasized, the point defect on the light modulation device is detected, and the point defect information is obtained. This is output (for example, see Patent Document 1).
[0003]
[Patent Document 1]
JP 2001-228050 A (page 9, FIG. 1)
[0004]
[Problems to be solved by the invention]
In the conventional point defect detection device as described above, the bright spot defect included in the image data captured by the image capturing device is based on the image data emphasized by the spatial filter for bright spot defect enhancement. The dark spot defect included in the image data captured by the image capturing device is detected based on the image data emphasized by the spatial filter for enhancing the dark spot defect. Since the image data needs to be filtered twice, such as to detect a defect, there is a problem that it takes a long processing time to detect a point defect.
[0005]
The present invention has been made in order to solve such problems, and can detect not only white but also black defects such as point defects and spot / uneven defects. It is an object of the present invention to provide a method and an apparatus for detecting a defect on a screen which can shorten the time.
[0006]
[Means for Solving the Problems]
A screen defect detection method according to the present invention includes a step of performing a defect enhancement process for enhancing and detecting various defects in an image obtained by capturing an image of a screen of an inspection target; Adding the approximate center value of the gray scale displaying the offset value as an offset value, calculating statistical data based on the luminance value of the pixel of the image to which the offset value has been added, and determining a bright defect based on the statistical data. Setting a threshold for extracting a dark defect, and extracting defect candidates of a bright defect and a dark defect from the statistical data and the threshold.
[0007]
With this configuration, a defect enhancement process is performed to enhance and detect various defects in an image obtained by capturing an image of the screen of the inspection target, and an image is displayed on the image subjected to the defect enhancement process. The approximate center value of the gradation is added as an offset value, statistical data based on the luminance value of the pixel of the image to which the offset value is added is calculated, and a threshold for extracting a bright defect and a dark defect based on the statistical data is calculated. By setting each of them and extracting the defect candidates of the bright defect and the dark defect from the statistical data and these threshold values, it is possible to detect the bright defect and the dark defect of the screen to be inspected by one defect emphasis process, The processing time for detecting a defect can be shortened.
[0008]
In the defect detection method for a screen according to the present invention, in the defect emphasizing process in the step of performing the defect emphasizing process, if the defect is a point defect, a bright spot defect emphasis filter is applied. Can be emphasized by simple filtering.
[0009]
Further, in the screen defect detection method according to the present invention, the defect emphasizing process in the step of performing the defect emphasizing process includes reducing the image obtained by capturing the screen of the inspection target in a plurality of stages when the defect is a spot defect. Since a plurality of reduced images are created and a stain defect emphasis filter is applied to each of the reduced images, stain defects of various sizes can be emphasized by simple filter processing.
[0010]
In the defect detection method for a screen according to the present invention, the defect enhancement processing in the step of performing the defect enhancement processing may include, if the defect is a non-uniform defect, a reduced image of a predetermined size from an image obtained by capturing the screen of the inspection target. And a first difference image between the reduced image and an image obtained by flattening a copy of the reduced image, and a first difference image of the flattened image and an image obtained by flattening the flattened image again. Since the second difference image and the second difference image are created, it is possible to emphasize uneven defects having a predetermined size.
[0011]
Further, in the screen defect detection method according to the present invention, the statistical data in the step of calculating the statistical data, the luminance value of each pixel in a predetermined area of the image in which the defect is enhanced processing is obtained, from the obtained luminance value The average value, the standard deviation, the maximum value, and the minimum value of the obtained entire predetermined area, and the threshold for extracting the light defect and the dark defect set based on the statistical data in the step of extracting the defect candidate, Since it is calculated by a predetermined formula based on the average value and the standard deviation obtained from the luminance value obtained from the image subjected to the enhancement processing, the image having the defect is extracted by the statistical processing in the step of extracting the defect candidate. Can be.
[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a block diagram showing a configuration of a screen defect detection apparatus according to an embodiment of the present invention.
In FIG. 1, reference numeral 1 denotes a liquid crystal light valve to be inspected, and reference numeral 2 denotes a projector, which is an image projection device. The liquid crystal light valve 1 can be set from outside. Reference numeral 3 denotes a pattern generator, which is a pattern generator for outputting various patterns to the liquid crystal light valve 1. Reference numeral 4 denotes a screen. Reference numeral 5 denotes a CCD camera, which is an image capturing device that captures an image projected on the screen 4. A CCD having a resolution of 1 or more is mounted.
Reference numeral 6 denotes a computer device which controls the pattern generator 3 and the CCD camera 5 to detect a point defect of the liquid crystal light valve 1, and 7 denotes a display device connected to the computer device 6.
[0013]
The computer device 6 includes an image input unit 61, a display area extraction unit 62, a background image difference processing unit 63, a defect emphasis processing unit 64, an offset value addition unit 65, a bright / dark defect extraction unit 66, And a display processing means 67.
Image data of a captured image captured by the CCD camera 5 is input to the image input means 61 of the computer device 6. The captured image is stored in storage means (not shown).
[0014]
The display area extracting means 62 extracts only the image portion which is the display area of the inspected portion from the captured image.
The background image difference processing unit 63 calculates a difference between the image in the display area and a previously created background image to create a background difference image from which defects caused by something other than the inspection target are removed.
The defect emphasis processing means 64 performs a convolution operation process on the background difference image using a spatial filter for bright spot defect emphasis, thereby emphasizing a point defect portion of the liquid crystal light valve 1.
[0015]
The offset value adding means 65 adds, as an offset value, the median value of the gray scale displaying the screen to the luminance value of each pixel of the image in which the point defect is emphasized.
The light / dark defect extraction means 66 sets a threshold for extracting a bright area for extracting a bright defect and a threshold for extracting a dark area for extracting a dark defect for the image to which the offset value is added. Each is set, cut out at these two thresholds, and processing is performed to obtain an image from which bright and dark defects have been extracted.
The result display processing means 67 performs a process for displaying the result of the image from which the light / dark defect has been extracted on the display device 7.
[0016]
Next, the operation of the screen defect detection device according to the embodiment of the present invention will be described.
FIG. 2 is a flowchart for explaining the operation of the screen defect detection apparatus according to this embodiment. The operation shown in FIG. 2 is realized by a program executed on the computer device 6.
First, the liquid crystal light valve 1 to be inspected is set on the projector 2, the pattern generator 3 is controlled by the computer device 6, a specific pattern is displayed on the liquid crystal light valve 1, and the pattern is projected on the screen 4 by the projector 2. . Then, the image projected on the screen 4 is photographed by the CCD camera 5, the image of the photographed data is output to the computer 6, and the computer 6 performs defect detection processing. Is displayed on the display device 7.
[0017]
Here, the operation of the point defect detection by the computer device 6 will be described based on the flowchart of FIG.
First, an image projected on the screen 4 is photographed by the CCD camera 5, and an image of the photographed data is taken into the image input means 60 of the computer device 6. At this time, the photographed data is converted by an A / D converter (not shown). , And 4096-gradation digital data are taken into the computer device 6 (step S1).
Next, the display area extracting means 62 extracts a display area for extracting only a screen portion which is a display area of the inspection target from the captured image data (step S2).
This extraction screen is a pattern matching process for the coordinates of the four corners of the image of the part to be inspected (for four small areas of several tens of pixels × several tens of pixels near the four corners of the image data, four corner reference images prepared in advance and The pattern can be extracted by performing pattern matching processing and specifying the coordinates of the four corners).
[0018]
Subsequently, the background image difference processing means 63 performs a background image difference process for removing a defect caused by something other than the liquid crystal light valve 1, such as illumination or a lens (step S3).
In this background image difference processing, the background image shown in FIG. 3B is subtracted from the image of the inspection target image data shown in FIG. A value of 1/2 of 4098 gradations is added as an offset value so as not to cause the background difference image shown in FIG. 3C, and the background difference image is a pixel corresponding to two pixels. An image of each difference is obtained. The background image is obtained by imaging a plurality of samples having as few defects as possible, creating an averaged image thereof, and extracting only the screen portion of the inspection target from the image.
The above steps S1 to S3 are called pre-processing.
[0019]
In the background difference image, it is difficult to detect a minute level of bright spot defect or the like, so that a defect enhancement process is performed using a spatial filter (step S4).
In this embodiment, a 7 × 7 Tophat filter is used to emphasize a bright point defect as a spatial filter, and the point defect is emphasized.
In order to detect whether the filter is isolated with respect to the surroundings, the 7 × 7 Tophat filter performs a convolution operation by weighting the value of the pixel of interest so that the difference from the value of the surrounding pixels is more emphasized. For example, a 7 × 7 Tophat filter which is a spatial filter as shown in FIG. 4 is shown.
[0020]
Here, enhancement of point defects will be described.
FIG. 5 schematically shows an image having a bright defect and a dark defect, which are point defects of a bright point and a dark point. As a result of enhancing the image of FIG. 5 by the 7 × 7 Tophat filter, a bright spot defect having a small luminance difference from the background as shown in FIG. Has become. On the other hand, a dark defect having a small luminance difference from the background is hardly emphasized, and the dark defect is hardly detected. FIG. 6A shows an image subjected to the Tophat filter processing, and FIG. 6B shows an image obtained by cutting out the image shown in FIG. 6A at a predetermined threshold value for extracting a bright region. It is an image detected by highlighting a bright defect of a point.
As described above, in the image subjected to the Tophat filter, a bright point indicating a bright point defect appears as a gray level having a positive value, whereas a dark point indicating a dark point defect appears as a gray level having a negative value. In the image processing format, image data usually takes only a positive value, so that the component of the dark point becomes 0 as it is, and it cannot be detected because there is no dark point data in the processed image data.
[0021]
Thus, when the screen is represented by 12 bits of 4096 gradations, a value of 2048 which is half of the value is added as an offset value to the result of the filter processing so that a dark point can be detected from the same image (step S5). When the image format is an 8-bit gray scale, the image has 256 gradations, and a half value of 128 is added to the result of the filter processing as an offset value.
As a result, the data of the dark point also appears as a tone having a positive value, so that point defects (bright and dark defects) of the bright point and the dark point can be detected by one filtering process. The image of FIG. 7A is an image obtained by adding an offset value to the image resulting from the filtering process.
[0022]
Then, the light / dark defect extraction processing means 66 cuts out the image to which the offset value has been added using two types of thresholds, a threshold for extracting a bright region and a threshold for extracting a dark region. Then, the extracted images are combined to obtain an image in which a light defect and a dark defect are detected as shown in FIG.
That is, in the point defect extraction image, a pixel having a luminance value larger than the bright defect extraction threshold can be determined to be a bright defect, and a pixel having a luminance value smaller than the dark defect extraction threshold can be determined as a dark defect. Can be determined.
Here, the threshold value for extracting the bright region and the threshold value for extracting the dark region are the average of the luminance values obtained from the entire screen of the image in which the point defect is emphasized by the point defect emphasis processing means 64. It is calculated and calculated by the following equation based on the value and the standard deviation.
Bright area extraction threshold = average (average value) + 4 * σ (standard deviation)
Threshold for dark area extraction = average (average value)-4 * σ (standard deviation)
Note that the magnification by which the standard deviation is multiplied in the above equation is not limited to 4, but is appropriately determined according to the degree of the point defect.
[0023]
According to this embodiment, the computer device 6 first compares only the screen portion of the part to be inspected from among the images captured by the CCD camera 5 including the entire screen of the liquid crystal light valve 1 to be inspected. A display area extraction process for extraction is performed, a background image difference process for removing a defect caused by something other than the liquid crystal light valve 1 is performed, and then a background image difference process is performed on the image subjected to the background image difference process by a Tophat filter. A point defect emphasis process is performed in which a bright point defect having a small difference in luminance is emphasized.
In the 7 × 7 Tophat filter, a filter that is isolated with respect to the surroundings is emphasized. Therefore, a filter that is isolated with respect to the surroundings such as a point defect of the liquid crystal light valve 1 is emphasized, and a moire fringe is formed. In the image in which the moire fringe is generated, only the point defect portion of the liquid crystal light valve 1 is emphasized even if the image is not isolated from the surroundings.
[0024]
Further, a threshold value for extracting a bright region from the image obtained by adding the median value of the gradation of the image as an offset value to the image in which the point defect is emphasized and a dark region from the image to which the offset value is added is added. In order to obtain an image in which point defects such as a light defect and a dark defect can be detected by binarizing with two types of threshold values, that is, a threshold value for extracting a light defect and a dark defect of the liquid crystal light valve 1 by one filtering process. The defect can be detected, and the processing time for detecting the defect can be shortened.
[0025]
In this embodiment, a method and an apparatus for detecting a point defect on a screen are described. However, in the case of detecting a stain on a screen, similarly to the above-described embodiment, the computer device 6 serving as an arithmetic unit is configured to perform preprocessing. A plurality of reduced images are created for the reduced image, the defect enhancement processing means 64 of the computer device 6 applies a Tophat filter to each reduced image, and the offset value adding means 65 applies a filtering process to each of the reduced images. The median value of the gradation is added as an offset value, the brightness / dark defect extraction means 66 obtains the brightness value of each pixel from the image to which the offset value has been added, calculates statistical data based on the brightness value of the pixel, Is set based on the threshold value, and a defect candidate having a light defect and a dark defect, which are white and black spots, is extracted from the statistical data and the threshold value.
[0026]
Also, in the case of detecting the unevenness of the screen, similarly to the above-described embodiment, a reduced image of a predetermined size is created for the image pre-processed by the computer 6 as the calculating means, and the defect enhancement of the computer 6 is performed. The processing means 64 re-flattens the first difference image in which the defect between the reduced image and the image obtained by flattening the copy of the reduced image is emphasized, and the flattened image and the flattened image are again processed. A second difference image in which a defect from the image is emphasized is created, and an offset value adding unit 65 adds, for each of the first and second difference images, a median value of the gradation of the image as an offset value; The brightness / dark defect extraction means 66 obtains the luminance value of each pixel for the image to which the offset value has been added, calculates statistical data based on the luminance value of the pixel, and sets a threshold value of the luminance value based on the statistical data. And so as to extract the defect candidate from the statistical data and the threshold value having a bright defect and dark defect is unevenness in black and white.
[0027]
The statistical data based on the luminance values of the pixels in the case of detecting the spots and unevenness of the screen is obtained by acquiring the luminance values of the respective pixels in a predetermined area of the image in which the defect is emphasized, and obtaining the predetermined luminance values from the acquired luminance values. The average value, standard deviation, maximum value and minimum value in the entire area. In addition, in the case of spot detection and unevenness detection, the threshold value of the luminance value for extracting a defect candidate having a bright defect and a dark defect, which are white and black stains or unevenness, is emphasized as in the case of the point defect. Based on the average value and the standard deviation of the luminance values obtained from the entire screen of the obtained image, it is calculated by the following equation.
Threshold = average (average value) + 4 * σ (standard deviation)
Threshold = average (average value) -4 * σ (standard deviation)
Note that the magnification by which the standard deviation is multiplied in the above formula is not limited to 4, but is appropriately determined according to the degree of the spot or unevenness defect.
In this way, not only when detecting a point defect on the screen, but also when detecting a spot or unevenness, the median of the gradation of the image is used as the offset value in the image in which the defect is emphasized. With respect to the image to which the addition and the offset value are added, not only white spots and white spots but also black spots and black spots can be detected.
[0028]
In this embodiment, the inspection object is described as a liquid crystal light valve. However, display parts such as a liquid crystal panel, a plasma display, an organic EL display, a DMD (digital mirror device), and a display device and a product inspection using the same are also applicable. Can be applied.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of a screen defect detection device according to an embodiment.
FIG. 2 is a flowchart showing the operation of the defect detection device on the same screen.
FIG. 3 is a diagram showing a background image difference process of the defect detection device on the same screen.
FIG. 4 is a diagram showing an example of a 7 × 7 Tophat filter used in the defect detection device on the same screen.
FIG. 5 is an image having a bright defect and a dark defect of the defect detection device on the same screen.
FIG. 6 is an image subjected to Tophat filtering by the defect detection device on the same screen.
FIG. 7 is an image subjected to bright / dark defect extraction processing by the defect detection device on the same screen.
[Explanation of symbols]
1. Liquid crystal light valve to be inspected, 2 projector, 3 pattern generator, 4 screen, 5 CCD camera, 6 computer device, 7 display device, 61 image input device, 62 display area extraction device, 63 background image difference processing device, 64 defect Emphasis processing means, 65 offset value adding means, 66 light / dark defect extracting means, 67 defect display processing means.

Claims (7)

A step of performing a defect enhancement process to enhance and detect various defects in an image obtained by capturing a screen of the inspection target;
A step of adding, as an offset value, a substantially median value of the gradation displaying the image to the image subjected to the defect enhancement processing;
Calculating statistical data based on the luminance value of the pixel of the image to which the offset value has been added;
Setting a threshold for extracting a light defect and a dark defect based on the statistical data, respectively, and extracting a defect candidate for a light defect and a dark defect from the statistical data and these thresholds;
A screen defect detection method, comprising: 2. The screen defect detection method according to claim 1, wherein the defect emphasizing process in the step of performing the defect emphasizing process is to apply a bright spot defect emphasizing filter when the defect is a point defect. The defect emphasizing process in the step of performing the defect emphasizing process, when the defect is a stain defect, creates a plurality of reduced images by reducing the image obtained by capturing the screen of the inspection target screen in a plurality of stages, and performs 2. The screen defect detection method according to claim 1, wherein a stain defect emphasis filter is applied to each of the images. The defect emphasizing process in the step of performing the defect emphasizing process includes, when the defect is a non-uniform defect, creating a reduced image of a predetermined size from an image obtained by capturing the screen of the inspection target, and copying the reduced image and the reduced image. A first difference image from the flattened image; and a second difference image between the flattened image and the flattened image. The method according to claim 1, wherein the screen defect is detected. The statistical data in the step of calculating the statistical data, the luminance value of each pixel in a predetermined area of the image in which the defect has been emphasized processing is obtained, the average value, the standard deviation, the maximum in the entire predetermined area obtained from the obtained luminance value 2. The method according to claim 1, wherein the value is a minimum value. The threshold for extracting the bright defect and the dark defect set based on the statistical data in the step of extracting the defect candidate is based on the average value and the standard deviation obtained from the luminance value obtained from the image in which the defect is emphasized. 6. The screen defect detection method according to claim 1, wherein the screen defect is calculated by a predetermined formula. 7. A screen defect detection apparatus, comprising: an arithmetic unit for performing the screen defect detection method according to claim 1.

Images