JPH11257937A - Defect inspecting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the occurrence of pseudo defect signals at the most peripheral part of a display region and to enable the accurate detection of defects throughout the display region, by performing smoothing processing on an input image to obtain a smoothed input image, performing variable-density expanding processing on the smoothed input image to create a background image, and comparing the input image with the background image. SOLUTION: An inspection pattern displayed on a screen 10 is captured by a television camera 12, and the image data is stored in memory in an image processing device 13 as an image A. Next, smoothing processing is performed on the image A, and the luminance changes in micro areas such as spot defects are removed from the image data to obtain an image B. Next, variable-density expanding processing is performed on the smoothing-processed image B so as to stretch waveforms in the direction of arrows 22 to obtain an image C. Next, when a finite-difference image between the image A and the image C is obtained by image computation processing, pseudo defects 23 appear only outside a display region 20. This image is converted into a binarized image on the basis of a predetermined threshold value to extract only defects. Then the sizes, number, location information, etc., of the extracted defects are measured, and the GO/NO GO determination of a subject to be inspected is performed on the basis of these information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a defect inspection method for inspecting a display element such as a liquid crystal panel and a plasma display used in a computer monitor, a TV receiver, a liquid crystal projector and the like.

[0002]

2. Description of the Related Art When trying to detect minute defects in a display device, problems such as uneven brightness of illumination, a problem of a viewing angle of a liquid crystal, and a problem of non-uniform imaging light quantity in an optical system are caused. Depending on the position, the detection sensitivity of the defect may change, and it may be difficult to accurately detect a minute defect.
Japanese Patent Laying-Open No. 4-336384 discloses a method for solving such a problem. This method is a method known as a defect detection method based on so-called shading correction, in which a background image is generated by performing a smoothing filter process on an input image, and a difference between the background image and the input image is obtained.

[0003]

The inspection area is within the display area of a display element such as a liquid crystal panel. However, the area to be imaged by a TV camera is set so that a slight deviation in setting the work can be tolerated. It is set slightly wider than the display area. It is to be noted that a configuration as disclosed in Japanese Patent Application Laid-Open No. 8-254499 in which a plurality of TV cameras are used in order to increase the resolution is also conceivable. You. Therefore,
When a background image is created by performing a smoothing filter process on the input image input in this way, the difference in luminance between the display area and the area outside the display area (normally, the difference between the display area and the outside of the display area is larger than that of the display area. (In many cases, the luminance is low.) This causes a drop in luminance at the outermost peripheral portion of the display area, and if the shading correction is applied as it is, the outermost peripheral part of the display area is always detected as a defect. On the other hand, if the false defect signal detected in the outermost peripheral portion of the display area is neglected, the defect is missed when a defect actually exists in the outermost peripheral portion of the display area.

Accordingly, the present invention is to solve such a problem. An object of the present invention is to provide a method of detecting defects by shading correction so that a pseudo defect signal is not generated at the outermost peripheral portion of a display area. It is an object of the present invention to provide a method that enables accurate defect detection over the entire display area.

[0005]

According to a first aspect of the present invention, there is provided a defect inspection method for detecting a minute defect of a display element of a display element, wherein an input image is generated by imaging an inspection area of the display element. An input image generation step, a background image generation step of performing a density expansion process on a smoothed input image obtained by performing a smoothing filter process on the input image to generate a background image, and the input image and the background image. And a comparing step of comparing in this order.

[0006] Here, the above-mentioned shading expansion processing is the following processing. First, a certain pixel of interest P is determined from the smoothed input image obtained by performing the smoothing filter processing of the input image. Then, a predetermined region including the pixel P (for example, five pixels of a pixel P, a pixel on the pixel P, a pixel below the pixel P, a pixel on the left of the pixel P, and a pixel on the right of the pixel P) ), The signal of the pixel having the highest signal intensity in the
Replace with the original pixel P signal. This operation is performed for all pixels of the smoothed input image.

For this reason, when the signal in the display area of the display element is higher in intensity than the signal outside the display area, sagging occurs in the outermost peripheral portion of the display area. This sag portion is pushed out of the display area.

As a result, a pseudo defect signal is not generated at the outermost peripheral portion of the display area, so that accurate defect detection can be performed over the entire display area.

According to a second aspect of the present invention, there is provided a defect inspection method for detecting a minute defect of a display element, wherein an input image generating step of imaging an inspection area of the display element to generate an input image; A background image generating step of generating a background image by performing shading reduction processing of a smoothed input image obtained by performing the smoothing filter processing, and a comparing step of comparing the input image with the background image. It is characterized by having in order.

Here, the above-mentioned shading reduction processing is the following processing. First, a certain pixel of interest P is determined from the smoothed input image obtained by performing the smoothing filter processing of the input image. Then, a predetermined region including the pixel P (for example, five pixels of a pixel P, a pixel on the pixel P, a pixel below the pixel P, a pixel on the left of the pixel P, and a pixel on the right of the pixel P) ), The signal of the pixel with the lowest signal intensity in the
Replace with the original pixel P signal. This operation is performed for all pixels of the smoothed input image.

For this reason, when the signal in the display area of the display element is lower in intensity than the signal outside the display area, sagging occurs at the outermost peripheral portion of the display area. This sag portion is pushed out of the display area.

As a result, a pseudo defect signal is not generated at the outermost peripheral portion of the display area, so that accurate defect detection can be performed over the entire display area.

According to a third aspect of the present invention, in the defect inspection method according to the first or second aspect, the comparison step is performed by performing an inter-image subtraction process.

Therefore, since the type of the image data may be an integer type, the data amount is relatively small, and the arithmetic processing can be performed at high speed. As a result, the defect inspection can be performed at high speed.

According to a fourth aspect of the present invention, in the defect inspection method of the first or second aspect, the comparison step is performed by performing an inter-image division process.

For this reason, it is necessary to change the type of the image data to a floating point type. However, even when the luminance unevenness is large in the display area, it is necessary to keep the SN ratio between the normal pixel and the defective pixel constant to some extent. This has the effect of enabling accurate defect detection.

[0017]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on embodiments.

Embodiment 1 FIG. 1 is a flowchart showing a defect inspection method according to Embodiment 1. FIG. 2 is a diagram illustrating a captured input image, and FIG. 3 is a configuration diagram of the defect inspection apparatus according to the first embodiment. The object to be inspected is an image projected by the liquid crystal projector 11. The inspection pattern 2 is projected on the screen 10 by the liquid crystal projector 11, and the projected inspection pattern 2 is imaged by the two-dimensional TV camera 12. The captured image is sent to the image processing device 13, where it is inspected whether or not there is a defect inside the projected inspection pattern 2, and the quality is determined. Here, the defects to be detected shall be only the defects of one pixel unit of the active matrix type display element called point defects, and the defects having a relatively large area such as color unevenness shall not be handled. I do. In the inspection pattern 2 to be projected, the entire display area is sequentially switched to white display, black display, and gradation display, and each of them is imaged and inspected.

In the image input step of step S1, a certain inspection pattern 2 displayed on the screen 10 is displayed on the TV.
The image data is captured by the camera 12 and stored in the memory of the image processing device 13 as the image A. In recent years, the resolution of display elements has been increasing from VGA to SVGA, XGA, and even higher, and accordingly, the resolution of TV cameras for inspecting such display elements is necessarily high. Conventionally, image processing of about 512 horizontal pixels × 480 vertical pixels has been generally performed, but a TV camera of 1000 pixels × 1000 pixels or more has begun to be used. Also, a method of dividing and inputting an image to be inspected to increase the resolution has been widely used. In any case, the input image A is an image in a range larger than the display area 2 so that all the display elements are captured, as shown in FIG. FIG. 4A shows the luminance change on the data extraction line 3 in the input image.
The luminance data is as follows. Although not shown in FIG. 2, it is assumed that there are a bright spot defect and a black spot defect on the data extraction line 3. The change on the graph corresponding to this defect is represented by the bright spot defect 30 and the black spot defect 3 in FIG.
It is one. The part corresponding to the display area 2 is 2 in this graph.
This is the range indicated by the arrow 0. It is assumed that the display area portion 2 has higher luminance than the portion outside the display area. When the data change in the display area is viewed macroscopically, a large mountain-like shape is relatively bright at the center, and the peripheral part is in a dark brightness unevenness state. This may be caused by uneven illuminance of the illumination light source or the characteristics of the liquid crystal panel or the lens optical system. However, it is relatively difficult to improve this situation due to technical and cost problems. Therefore, from the image state input in this way,
I want to detect the bright spot defect 30 and the black spot defect 31. The difference between the maximum value and the minimum value of the luminance unevenness is the defect 30, 31.
Therefore, it is not possible to simply binarize and detect only the defect. That is, the bright spot defect 30
If the maximum value of the luminance unevenness is larger than the peak of the luminance or the minimum value of the luminance unevenness is smaller than the minimum value of the black spot defect 31, a pseudo defect due to the luminance unevenness is detected simultaneously with the defect. Would be. Therefore, it is considered that only a minute defect such as a bright spot defect 30 or a black spot defect 31 is detected from such luminance unevenness by performing a shading correction process described later.

In step S2, a copy of the image A is created and a smoothing filter process is performed. The reason for making a copy of the image A is to use the image A when performing an inter-image operation later. This smoothing filter process is a process of removing a luminance change of a small area such as a point defect from image data.
Usually, it is realized by performing a convolution operation of a 3 × 3 or 5 × 5 matrix over the entire target image. An image created as a result of the smoothing filter processing is defined as an image B. The state of this processing will be described with reference to FIG. 4B after the smoothing filter processing. Compared to FIG. 4A, the luminance change of a fine cycle is removed, and the bright spot defect 30 and the black spot defect 31 are removed.
It can be seen that the change in luminance corresponding to is also removed on the graph. Further, the luminance change at the boundary between the display area on both sides of FIG. 4B and the outside of the display area is more drastic compared to the graph before the smoothing filter processing (FIG. 4A). And the change in luminance is gradual. 4C shows the input image A and the image B after the smoothing filter processing.
6 is a graph showing a result obtained by calculating a difference between. 4 (a) and 4 (b) and FIG. 4 (c), the scale of the vertical axis is changed in order to make the defect easy to understand. Conventionally, in order to remove the influence of luminance unevenness, such a difference image is obtained, binarized and a defect is detected. Here, before and after the dotted lines on both sides inserted for the purpose of indicating the position of the display area 20. Focusing on the data change in the vicinity, it can be seen that the pseudo defect 21 has been generated. In this state, if binarization is performed for the purpose of detecting a defect, the pseudo defect 21 will be detected simultaneously with the bright spot defect 32. Further, in order not to detect the pseudo defect 21, the inspection range must be set inside the display area 20 so as not to include the pseudo defect 21, and in such a case, the inspection range exists at the very end of the display area. You will overlook the defect.

In order to avoid this, in the defect inspection method according to the present embodiment, the shading expansion processing of step S3 is performed before obtaining the difference image. Regarding the shading expansion processing, refer to
Japanese Patent Application Laid-Open No. 4-336384 describes column 9 and FIG. 3, and the shading reduction process described later is described in Japanese Patent Application Laid-Open No. 4-336384, column 10 and FIG. FIG. 5 shows that the image B subjected to the smoothing filter processing is subjected to shading expansion processing.
As shown in (c), the effect of extending the waveform in both directions in the direction of arrow 22 is obtained. This image is called image C. As a result, the sagged portion of the shoulder near the boundary of the display area is driven out of the display area 20, and the step S4 is performed.
The difference image between the image A and the image C (this is referred to as an image D) is obtained by the image calculation processing of (1).

If the brightness of the display area is higher than that of the surroundings, the density expansion processing is performed in step S3. Conversely, if the brightness of the surrounding area is higher, the density reduction processing is performed. I do.

As a result, as can be confirmed by the waveform after the difference processing in FIG.
It appears only outside the dotted line of 0, and accurate defect detection can be performed over the entire display area. Step S
In the inter-image operation 4, the difference between the two images has been described as described above. However, as the image data, an integer data type of 8 to 16 bits per pixel is generally used. Therefore, the difference calculation can be performed with the data type as it is, and the processing is simple, so that the speed is easily increased. However, even for defects of the same type and similar, there may be a large difference in the contrast of the defect between the case where the luminance unevenness is bright and the case where the defect is dark, but in such a case, step S
Using a division process by 4 gives better results. On the other hand, it is necessary to convert the image data to a floating-point type, so that the data amount tends to increase and the processing time tends to increase.

In step S5, a binarized image is formed based on a predetermined threshold value, and only defects are extracted. Here, the size, number, position information, and the like of the defects are measured, and in step S6, the quality of the inspection target is determined based on the information.

The threshold value for binarization may be determined in advance by checking the threshold level of a defect to be detected with a limit sample or the like, or based on the standard deviation of data in the case of a non-defective product. In the case where the directions in which the defect data appears are opposite, such as the difference between the bright spot defect 34 and the black spot defect 35 in FIG. 5D, a plurality of thresholds for extracting the defect are prepared. It is also necessary.

The above description is an example of a case where a defect inspection is performed from an inspection pattern projected on a screen by a liquid crystal projector, but the liquid crystal panel itself faces the liquid crystal panel with a flat backlight placed behind it. Even when a defect is detected by a TV camera, the inspection can be performed based on the same principle.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an example of a defect inspection method according to a first embodiment.

FIG. 2 is a diagram showing an input screen imaged;

FIG. 3 is a configuration diagram of an inspection apparatus using the defect inspection method according to the first embodiment.

FIG. 4 is a graph illustrating a procedure of a conventional shading correction.

FIG. 5 is a graph for explaining a shading correction procedure according to the first embodiment.

[Explanation of symbols]

 1 ... input area 2 ... display area

Claims (4)

[Claims] 1. A defect inspection method for detecting minute defects of a display element, comprising: an input image generation step of imaging an inspection area of the display element to generate an input image; and performing a smoothing filter process on the input image. A background image generating step of performing a density expansion process on the smoothed input image obtained by the above to generate a background image, and a comparing step of comparing the input image and the background image,
Defect inspection method in this order. 2. A defect inspection method for detecting minute defects of a display element, comprising: an input image generation step of imaging an inspection area of the display element to generate an input image; and performing a smoothing filter process on the input image. A background image generating step of performing a density reduction process on the smoothed input image obtained by the above to generate a background image, and a comparing step of comparing the input image and the background image,
Defect inspection method in this order. 3. The defect inspection method according to claim 1, wherein the comparing step is performed by performing an inter-image subtraction process. 4. The defect inspection method according to claim 1, wherein the comparing step is performed by performing an inter-image division process.