JP2008170288A - Device for inspecting flaw, and method for inspecting flaw - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for inspecting flaw capable of accurately detecting flaws, and to provide a method for inspecting the flaws. <P>SOLUTION: A luminance acquisition means 31 acquires the luminance distribution of a screen 1A, on the basis of the photographed image due to a camera 2, and a compositing means 32 composites a plurality of the luminance distributions acquired by the luminance acquisition means 31. A flaw extraction means 33 extracts the flaw of the screen 1A, on the basis of the brightness distributions obtained by the compositing means 32. An operation control means 34 controls a panel drive device 11. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a defect inspection apparatus and a defect inspection method for extracting defects in an inspection area based on luminance information.

  As an inspection apparatus for detecting a pixel defect of a liquid crystal display device, an inspection apparatus that captures a display image by a camera and detects the defect by image processing is known. The inspection device captures the object to be inspected, captures an image from the camera, and numerically processes the luminance information of the entire inspection area of the object to be inspected. And an image processing device for extracting.

  In the image processing apparatus, after capturing an image, a differential enhancement filter is applied to the captured image as necessary. Next, the image is binarized to extract defect candidates, and each is numbered. Next, the feature amount is calculated for each defect candidate. The feature amount is a numerical value for determining a defect, and includes, for example, average luminance, maximum luminance, luminance volume, average contrast, maximum contrast, and the like. This feature amount is compared with a preset threshold value, identified, and output as a defect.

  In the inspection of the liquid crystal display device, all the red (R), green (G), and blue (B) pixels are individually lit to inspect defects that do not appear in the all-color lighting image. Are inspected. However, since the color portion other than the inspection target is a dark portion, there is a problem that a bright portion and a dark portion are mixed in the inspection region, and it is difficult to extract a defect. For example, a non-lighting portion of another color may be erroneously detected as a defect, or a dark defect adjacent to a non-lighting portion of another color may be missed. Further, since the inspection is performed for each emission color, there is a problem that the inspection takes a long time.

  An object of the present invention is to provide a defect detection apparatus and a defect detection method capable of detecting defects with high accuracy.

The defect inspection apparatus of the present invention is a defect inspection apparatus that extracts defects in an inspection area based on luminance information, and a luminance acquisition unit that acquires an actual luminance distribution of an inspection area where a predetermined luminance distribution is expected; A synthesizing unit for synthesizing the luminance distribution reversed with respect to the predetermined luminance distribution and the actual luminance distribution acquired by the luminance acquiring unit; and the inspection target based on the luminance distribution obtained by the synthesizing unit And a defect extracting means for extracting defects in the region.
According to this defect inspection apparatus, the luminance distribution obtained by combining the actual luminance distribution of the inspected area where a predetermined luminance distribution is expected and the luminance distribution reversed with respect to the predetermined luminance distribution. Therefore, the defect can be detected with high accuracy.

The defect inspection apparatus according to the present invention is a defect inspection apparatus that extracts defects in each of a plurality of divided areas constituting the inspection area by complementing each other, based on luminance information, and the luminance distribution of the plurality of divided areas A luminance acquisition unit that acquires the defect, a combination unit that combines the plurality of luminance distributions acquired by the luminance acquisition unit, and a defect that extracts a defect in the inspection area based on the luminance distribution obtained by the combination unit And an extracting means.
According to this defect inspection apparatus, since the luminance distributions of a plurality of divided regions are combined and the defects in the inspection region are extracted based on the luminance distribution obtained by the combination, the defects can be detected with high accuracy.

  The synthesizing unit may synthesize the luminance distribution after correcting the luminance of the plurality of divided regions to be uniform.

  Operation control means for operating the inspection object for each of the divided areas may be provided, and the luminance acquisition means may individually acquire the luminance distribution of the divided areas in the operating state by the operation control means.

  The luminance acquisition means acquires the luminance distribution of the entire inspection area at once using an imaging optical device, and acquires the luminance distribution of the plurality of divided areas by optically separating the luminance distribution. May be.

  Each of the divided regions may be configured as a set of pixels having a specific hue.

The defect inspection method of the present invention is a defect inspection method for extracting defects in an inspected area based on luminance information, obtaining an actual luminance distribution of the inspected area where a predetermined luminance distribution is expected, and A luminance distribution inverted with respect to the luminance distribution of the image and the actual luminance distribution acquired by the step of acquiring the luminance distribution, and the luminance distribution obtained by the combining step. And extracting a defect in the inspection area.
According to this defect inspection method, the luminance distribution obtained by combining the actual luminance distribution of the inspected area where the predetermined luminance distribution is expected and the luminance distribution reversed with respect to the predetermined luminance distribution. Therefore, the defect can be detected with high accuracy.

The defect inspection method of the present invention is a defect inspection method for extracting defects in each of a plurality of divided areas constituting an inspection area by complementing each other, based on luminance information, and the luminance distribution of the plurality of divided areas , A step of synthesizing the plurality of luminance distributions acquired by the step of acquiring the luminance distribution, and a defect in the inspection area is extracted based on the luminance distribution obtained by the synthesizing step. And a step.
According to this defect inspection method, the luminance distributions of a plurality of divided regions are synthesized, and the defects in the region to be inspected are extracted based on the luminance distribution obtained by the synthesis, so that the defects can be detected with high accuracy.

  According to the defect inspection apparatus of the present invention, the actual luminance distribution of the inspected area where the predetermined luminance distribution is expected and the luminance distribution inverted with respect to the predetermined luminance distribution are synthesized and obtained by synthesis. Since the defect in the inspection area is extracted based on the luminance distribution, the defect can be detected with high accuracy.

  According to the defect inspection apparatus of the present invention, since the luminance distributions of a plurality of divided areas are synthesized and the defects in the inspected area are extracted based on the luminance distribution obtained by the synthesis, the defects can be detected with high accuracy.

  According to the defect inspection method of the present invention, the actual luminance distribution of the inspected area where a predetermined luminance distribution is expected and the luminance distribution inverted with respect to the predetermined luminance distribution are synthesized and obtained by synthesis. Since the defect in the inspection area is extracted based on the luminance distribution, the defect can be detected with high accuracy.

  According to the defect inspection method of the present invention, the luminance distributions of a plurality of divided regions are synthesized, and the defects in the region to be inspected are extracted based on the luminance distribution obtained by the synthesis, so that the defects can be detected with high accuracy.

  Hereinafter, embodiments of a defect inspection apparatus according to the present invention will be described.

  FIG. 1 is a block diagram showing the configuration of the defect inspection apparatus according to the present embodiment.

  As shown in FIG. 1, the defect inspection apparatus of the present embodiment includes a monochrome camera 2 that images a screen 1A of a matrix display type liquid crystal display panel 1 as an object to be inspected, and a liquid crystal display based on an imaging signal of the camera 2. And an image processing device 3 for extracting defective pixels of the panel 1. The image processing apparatus 3 can be configured by a personal computer or the like.

  The liquid crystal display panel 1 is a color panel including red (R), green (G), and blue (B) pixels. As shown in FIG. 1, a panel driving device 11 is connected during inspection. The

  The image processing apparatus 3 includes a luminance acquisition unit 31 that acquires the luminance distribution of the screen 1A based on an image captured by the camera 2, a combination unit 32 that combines the plurality of luminance distributions acquired by the luminance acquisition unit 31, and a combination. Defect extraction means 33 for extracting defects on the screen 1A based on the luminance distribution obtained by the means 32, and operation control means 34 for controlling the panel drive device 11 are configured.

  FIG. 2 is a flowchart showing the operation of the image processing apparatus 3.

  In step S1 to step S3 in FIG. 2, the process of acquiring the luminance distribution of the screen 1A by causing the pixels of the liquid crystal display panel 1 to emit light one color at a time.

  In step S <b> 1, one light emission color is selected in the operation control device 34, and only the pixels of the selected light emission color among the pixels of the liquid crystal display device 1 are turned on by the panel driving device 11. Next, in step S <b> 2, the luminance distribution of the screen 1 </ b> A is captured by the image luminance acquisition unit 31 based on the imaging signal of the screen 1 </ b> A by the camera 2.

  Next, in step S3, it is determined whether or not the luminance distribution has been acquired for all the emission colors. If the determination in step S3 is affirmed, the process proceeds to step S4. If the determination is negative, the process returns to step S1, and the processes of steps S1 and S2 are repeated for the next emission color.

  As described above, in the processing from step S1 to step S3, the luminance distribution of the screen 1A is acquired for all the emission colors (R color, G color, B color), and the process proceeds to the next processing.

  In addition, when the liquid crystal display panel 1 and the camera 2 are fixed and the respective emission colors (R color, G color, and B color) are imaged, it is not necessary to perform image alignment in the composition processing described later.

  In step S4 of FIG. 2, the synthesizing unit 32 executes luminance distribution synthesis processing.

  FIG. 3 is a flowchart illustrating the specific contents of the luminance distribution composition processing in step S4.

  In step S41 of FIG. 3, based on the luminance distribution acquired in the processing of steps S1 to S3, the average luminance value of light emission for each divided region of the screen 1A is calculated for each of R, G, and B colors.

  FIG. 4 is a diagram illustrating an example of a divided area of the screen 1A. In the example of FIG. 4, the screen 1A is divided into 100 divided areas (divided areas R00 to 99) by dividing the screen 1A into 10 parts in the vertical direction and the horizontal direction, respectively. In step S41, the luminance average value of the light emission is calculated for the R color, the G color, and the B color for each of the divided regions.

  Next, in step S42, using the calculation result of step S41, the ratio of the luminance average value of the light emission of B color and R color is calculated for each divided region.

  Next, in step S43, based on the calculation result of step S42, the luminance of R light emission is corrected equally for all the divided regions. Here, for example, a standard ratio is selected from the ratios of the luminance average values for all the divided regions calculated in step S42. Next, the luminance of the R light emission is corrected for all the divided regions in accordance with the selected ratio. Thus, the luminance distribution of the R light emission can be corrected for the entire screen 1A so that the average luminance of the R color matches the average luminance of the B color in a standard region in the screen 1A.

  Next, in step S44, using the calculation result of step S41, the ratio of the luminance average value of the light emission of B color and G color is calculated for each divided region.

  Next, in step S45, based on the calculation result of step S42, the luminance of G light emission is uniformly corrected for all the divided regions. Here, for example, a standard ratio is selected from the ratios of the luminance average values for all the divided regions calculated in step S42. Next, the luminance of G light emission is corrected for all the divided regions according to the selected ratio. Thereby, the luminance distribution of the G light emission can be corrected for the entire screen 1A so that the average luminance of the G color matches the average luminance of the B color in a standard region in the screen 1A.

  Next, the B light emission luminance distribution acquired in steps S1 to S3, the R light emission luminance distribution corrected in step S43, and the G light emission luminance distribution corrected in step S45. Are combined, and the process returns to step S5 (FIG. 2).

  Next, in step S5 of FIG. 2, optical characteristics such as the optical system 21 of the camera 2 are complemented by shading correction, and the synthesized image is corrected.

  Next, in step S <b> 6 to step S <b> 8, defect part candidates are extracted by the defect detection means 33.

  First, in step S6, a combined enhancement filter is applied to the synthesized luminance distribution image. Next, in step S7, the image is binarized to extract defect candidates. Here, the image is binarized with “1” being a level higher than a preset threshold and “0” being a lower level, and a high level portion is extracted as a defect candidate. Next, in step S8, each defect candidate is numbered.

  Next, in step S9 to step S10, the defect detection unit 33 determines whether the defect candidate is a defect.

  In step S9, a feature amount is calculated for each defect candidate. The feature amount is a numerical value for determining a defect, and includes, for example, average luminance, maximum luminance, luminance volume, average contrast, maximum contrast, and the like. In step S10, this feature amount is compared with a preset threshold value and identified. A defect candidate whose feature quantity exceeds the threshold is determined as a defect. The feature amount is appropriately defined according to the defect to be detected.

  Next, in step S11, a defect is output based on the determination result (step S10), and the process ends.

  As described above, in this embodiment, for each emission color (R color, G color, B color), the luminance distribution is individually acquired, and the luminance distribution of R color and G color is corrected and then the luminance distribution is synthesized. is doing.

  FIG. 5 is a diagram conceptually illustrating the synthesis of the luminance distribution.

  In the example shown in FIG. 5, the case where the dark defect 51 exists in the pixel of G color is shown. However, even if it is attempted to extract the dark defect 51 based on the image 5G of the G luminance distribution, the defect extraction may be hindered by the mixture of the bright part and the dark part. On the other hand, in the image 50 obtained by synthesizing the image 5R of the luminance distribution of R color, the image 5G of the luminance distribution of G color, and the image 5B of the luminance distribution of B color, the dark portion of the image 5G is the image 5R and the image Interpolated by 5B, the image becomes flat as a whole. For this reason, the dark defect 51 can be easily extracted, and the inspection accuracy can be improved.

  Further, in this embodiment, defects can be extracted at a time for each emission color (R color, G color, and B color) by image processing on the combined image. For this reason, inspection time can be shortened.

  In the above-described embodiment, the example using the camera 2 which is an area sensor in which photoelectric conversion elements are two-dimensionally arranged is shown. However, the inspection sensor is not limited to the area sensor, and the photoelectric conversion element is one-dimensional (straight line). The present invention can also be applied to an inspection apparatus using a line sensor arranged in the above.

  In the above embodiment, the luminance distribution is acquired with the lighting pattern for each light emission color (R color, G color, B color), but any lighting pattern that complements each other can be used.

  It is also possible to use a 3CCD camera that optically separates images and captures them with separate CCDs. That is, in the above embodiment, the liquid crystal display panel 1 is turned on for each emission color (R color, G color, B color), and the camera 2 takes an image three times. The panel 1 may be imaged with a 3CCD camera, and three images obtained by each CCD may be combined. In addition, the liquid crystal display panel 1 that was fully lit (white lighted) was photographed with a camera capable of switching R, G, and B color filters, and optically color-separated using each filter. Images may be combined.

  In this embodiment, the defect inspection apparatus of the present invention is applied to defect inspection of an image sensor.

  FIG. 6 is a block diagram showing the configuration of the defect inspection apparatus of this embodiment.

  As illustrated in FIG. 6, the image processing apparatus 4 includes a luminance acquisition unit 41 that acquires the luminance distribution of the captured image of the image sensor 61, and a synthesis unit 42 that combines the plurality of luminance distributions acquired by the luminance acquisition unit 41. A defect extraction unit 43 that extracts pixel defects of the image sensor 61 based on the luminance distribution obtained by the synthesis unit 42 and an operation control unit 44 that controls the light source 62 are configured.

  The light source 62 includes transmission filters for R, G, and B colors, and functions as a single color light source for R, G, and B colors by switching the filters.

  FIG. 7 is a flowchart showing the operation of the image processing apparatus 4.

  In step S21 to step S23 in FIG. 7, R light, G light, and B light are sequentially emitted from the light source 62, and processing for acquiring the luminance distribution of the captured image of the image sensor 61 is executed.

  In step S <b> 21, the operation control device 44 selects one of R color, G color, and B color, and irradiates light of the color selected from the light source 62. Next, in step S <b> 22, the luminance distribution of the captured image is captured by the image luminance acquisition unit 41 based on the imaging signal of the image sensor 61.

  Next, in step S23, it is determined whether or not the luminance distribution has been acquired for all colors. If the determination in step S23 is affirmative, the process proceeds to step S24. If the determination is negative, the process returns to step S21, and the processes in steps S21 to S22 are repeated for the next emission color.

  As described above, in the processing from step S21 to step S23, the luminance distribution of the captured image of the image sensor 61 is acquired for all colors (R color, G color, B color), and the process proceeds to the next processing.

  Next, in step S24, the synthesizing unit 42 executes luminance distribution synthesis processing. Here, an image in which the luminance distribution of each color is added after the signal intensity of each color is aligned is created.

  Next, in step S <b> 26 to step S <b> 28, defect candidate candidates are extracted by the defect extraction means 43.

  First, in step S26, a differential enhancement filter is applied to the synthesized luminance distribution image. Next, in step S27, the image is binarized to extract defect candidates, and in step S28, each defect candidate is numbered.

  Next, in step S29 to step S30, the defect extraction unit 43 determines whether the defect candidate is a defect.

  In step S29, a feature amount is calculated for each defect candidate. The feature amount is a numerical value for determining a defect, and includes, for example, average luminance, maximum luminance, luminance volume, average contrast, maximum contrast, and the like. In step S30, the feature amount is compared with a preset threshold value and identified. A defect candidate whose feature quantity exceeds the threshold is determined as a defect.

  Next, in step S31, a defect is output based on the determination result (step S30), and the process ends.

  As described above, in this embodiment, for each emission color (R color, G color, and B color), the luminance distribution is individually acquired, the luminance is corrected, and the luminance distribution is synthesized. Therefore, as described with reference to FIG. 5, pixel defects can be easily extracted, and inspection accuracy can be improved. In addition, it is possible to extract defects for pixels of each color (R color, G color, and B color) at once by image processing on the combined image. For this reason, inspection time can be shortened.

  The defect inspection apparatus according to the present embodiment shows an example in which defects are extracted after an actually captured image and an image prepared in advance are combined.

  FIG. 8 is a block diagram illustrating the configuration of the defect inspection apparatus according to the present embodiment. The same elements as those in the first embodiment (FIG. 1) are denoted by the same reference numerals and the description thereof is omitted.

  The image processing apparatus 7 includes a luminance acquisition unit 71 that acquires a luminance distribution of the screen 1A that is lit in a specific pattern based on a photographed image by the camera 2, a luminance distribution acquired by the luminance acquisition unit 71, and the specific pattern. Are combined with a luminance distribution of the reverse pattern and a defect extracting unit 73 for extracting defects on the screen 1A based on the luminance distribution obtained by the combining unit 72.

  FIG. 9 is a flowchart showing the operation of the image processing apparatus 7.

  In step S51 to step S52 in FIG. 9, the process of obtaining the luminance distribution of the screen 1A by causing the pixels of the liquid crystal display panel 1 to emit light in a specific pattern is executed.

  In step S51, the liquid crystal display panel 1 is driven by the panel driving device 11, and the pixels of the liquid crystal display device 1 are caused to emit light in a specific pattern. For example, only the G color pixel is caused to emit light.

  In step S <b> 52, the luminance distribution of the screen 1 </ b> A is captured by the image luminance acquisition unit 71 based on the imaging signal of the screen 1 </ b> A from the camera 2. Next, in step S53, the optical distribution such as the optical system 21 of the camera 2 is complemented by shading correction, and the luminance distribution is corrected.

  Next, in step S54, the synthesizing unit 72 executes a luminance distribution synthesizing process. Here, an image is created by combining the luminance distribution corrected in step S53 and the virtual luminance distribution prepared in advance. The virtual luminance distribution is a luminance distribution inverted with respect to an ideal luminance distribution when light is emitted with the specific pattern. When the ideal luminance distribution and the virtual luminance distribution are combined, uniform luminance can be obtained over the entire screen 1A. Therefore, only the defect appears in the synthesized image.

  Next, in step S55 to step S57, the defect extraction means 73 extracts defect site candidates.

  First, in step S55, a differential enhancement filter is applied to the synthesized luminance distribution image. Next, in step S56, the image is binarized to extract defect candidates. Here, the image is binarized with “1” being a level higher than a preset threshold and “0” being a lower level, and a high level portion is extracted as a defect candidate. Next, in step S57, each defect candidate is numbered.

  Next, in step S58 to step S59, the defect extraction unit 73 determines whether the defect candidate is a defect.

  In step S58, a feature amount is calculated for each defect candidate. The feature amount is a numerical value for determining a defect, and includes, for example, average luminance, maximum luminance, luminance volume, average contrast, maximum contrast, and the like. In step S59, this feature amount is compared with a preset threshold value and identified. A defect candidate whose feature quantity exceeds the threshold is determined as a defect. The feature amount is appropriately defined according to the defect to be detected.

  Next, in step S60, a defect is output based on the determination result (step S59), and the process ends.

  As described above, in this embodiment, the luminance distribution obtained by imaging and the virtual luminance distribution prepared in advance are combined to make the image uniform and to easily extract defects.

  The present invention is widely applied to inspection of an inspection object using luminance information. For example, the present invention can be applied to inspection of a liquid crystal panel substrate on which TFT elements are arranged, various color filters, a PDP (plasma display panel) and its substrate, and a silicon wafer on which an IC pattern is formed.

  As described above, according to the defect inspection apparatus of the present invention, since the luminance distributions of a plurality of divided areas are synthesized and the defects in the inspected area are extracted based on the luminance distribution obtained by the synthesis, the defects are accurately detected. Can be detected well. Further, according to the defect inspection apparatus of the present invention, the actual luminance distribution of the inspected area where the predetermined luminance distribution is expected and the luminance distribution reversed with respect to the predetermined luminance distribution are synthesized and obtained by synthesis. Since the defect in the inspection area is extracted based on the obtained luminance distribution, the defect can be detected with high accuracy.

  The scope of application of the present invention is not limited to the above embodiment. The present invention can be widely applied to a defect inspection apparatus and a defect inspection method for extracting defects in an inspection area based on luminance information.

1 is a block diagram illustrating a configuration of a defect inspection apparatus according to Embodiment 1. FIG. 6 is a flowchart showing the operation of the image processing apparatus. The flowchart which illustrates the specific content of the synthetic | combination process of luminance distribution. The figure which shows the example of the division area of a screen. The figure which shows notionally composition of luminance distribution. FIG. 3 is a block diagram illustrating a configuration of a defect inspection apparatus according to a second embodiment. 6 is a flowchart showing the operation of the image processing apparatus. FIG. 9 is a block diagram illustrating a configuration of a defect inspection apparatus according to a third embodiment. 6 is a flowchart showing the operation of the image processing apparatus.

1A screen (inspected area)
31 Luminance Acquisition Unit 32 Synthesis Unit 33 Defect Extraction Unit 34 Operation Control Unit 41 Luminance Acquisition Unit 42 Synthesis Unit 43 Defect Extraction Unit 44 Operation Control Unit 61 Image Sensor (Inspected Area)
71 luminance acquisition means 72 synthesis means 73 defect extraction means

Claims (8)

In a defect inspection apparatus that extracts defects in an inspection area based on luminance information,
A luminance acquisition means for acquiring an actual luminance distribution of an inspection area where a predetermined luminance distribution is expected;
Synthesis means for synthesizing the luminance distribution reversed with respect to the predetermined luminance distribution and the actual luminance distribution acquired by the luminance acquisition means;
Defect extraction means for extracting defects in the inspected area based on the luminance distribution obtained by the synthesis means;
A defect inspection apparatus comprising: In a defect inspection apparatus that extracts defects in each of a plurality of divided areas constituting an inspection area by complementing each other based on luminance information,
A luminance acquisition means for acquiring a luminance distribution of the plurality of divided regions;
Combining means for combining the plurality of luminance distributions acquired by the luminance acquisition means;
Defect extraction means for extracting defects in the inspected area based on the luminance distribution obtained by the synthesis means;
A defect inspection apparatus comprising: The defect inspecting apparatus according to claim 2, wherein the synthesizing unit synthesizes the luminance distribution after correcting the luminance of the plurality of divided regions to be uniform. Comprising an operation control means for operating the inspection object for each of the divided areas;
4. The defect inspection apparatus according to claim 2, wherein the luminance acquisition unit individually acquires the luminance distribution of the divided regions in an operation state by the operation control unit. The luminance acquisition unit acquires the luminance distribution of the entire inspection area at once using an imaging optical device, and acquires the luminance distribution of the plurality of divided areas by optically separating the luminance distribution. The defect inspection apparatus according to any one of claims 2 to 4, wherein The defect inspection apparatus according to claim 2, wherein each of the divided areas is configured as a set of pixels having a specific hue. In a defect inspection method for extracting defects in an inspection area based on luminance information,
Obtaining an actual luminance distribution of an inspected area where a predetermined luminance distribution is expected;
Synthesizing the luminance distribution reversed with respect to the predetermined luminance distribution and the actual luminance distribution acquired by the step of acquiring the luminance distribution;
Extracting a defect in the inspection area based on the luminance distribution obtained by the combining step;
A defect inspection method comprising: In a defect inspection method for extracting defects in each of a plurality of divided areas constituting an inspection area by complementing each other based on luminance information,
Obtaining a luminance distribution of the plurality of divided regions;
Synthesizing the plurality of luminance distributions acquired by the step of acquiring the luminance distribution;
Extracting a defect in the inspection area based on the luminance distribution obtained by the combining step;
A defect inspection method comprising:

Images