JPH0611679A - Method for inspecting and reading display element - Google Patents

Abstract

PURPOSE:To inspect light quantity emitted by a pixel with high accuracy by regarding an integrated value of detected light quantities from individual sensor pixels among sensor pixels group relating to a pixel image as the light quantity of pixels to be inspected. CONSTITUTION:Luminous intensity distribution 2 of the pixel image in which respective pixels are in lighting state is shown with a relation between a pixel image region 1 and sensor pixels 3 are shown with a relation between respective positions. The pixel image region of the noticed pixel on the sensor pixel surface is denoted by 1T, the luminous intensity distribution characteristic of the image region is denoted by 2T, respective pixel image regions of left and right sides pixels adjacent to the noticed pixel are denoted by 1NL, 1NR, the luminous intensity distribution characteristics of the image regions are respectively denoted by 2NL, 2NR and sensor pixels relating to the noticed pixel are denoted by 31 to 37. In order to detect all rays of light emitted from the pixel to be inspected, light quantities detected by all sensor pixels relating to the pixel image to be detected are summed. That is, the summed value of light quantities 42, 41, 43 detected by sensor pixels 33 to 35 is read-out as light quantity emitted from the pixel to be inspected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting a display device such as a liquid crystal display and a plasma display.
More specifically, the screen displayed by the display element is read,
It is related to the method of inspecting the quality.

[0002]

2. Description of the Related Art One of the inventors of the present invention disclosed in Japanese Patent Laid-Open No. 4-16895 "Display Element" as a method for efficiently reading an inspection target screen with a small number of sensor pixels, that is, with a small number of cameras. There is an invention filed by the same applicant as the present application proposed in “Inspection screen reading method” (hereinafter referred to as “first conventional example”) and Japanese Patent Application No. 3-50101 “Display element inspection method” (hereinafter referred to as “second conventional example”). In the former first conventional example, an image per pixel of the display element on the sensor light-receiving surface is formed by turning off the neighboring pixels around the pixel to be inspected and further forming a blurred image on the sensor light-receiving surface. It is characterized in that the readability is increased by enlarging the area and increasing the number of sensor pixels related to the image of one pixel of the display element. In the latter second conventional example, the positional relationship between the display element to be inspected and the sensor camera is always set to the reference position, and the luminous intensity information sensed by the sensor pixel is multiplied by the corresponding position correction coefficient extracted at the reference position. Is characterized in that the luminous intensity information of the display pixel is generated.
In both the first conventional example and the second conventional example, a part of the light emitted by the display pixel is detected by a single sensor pixel, and the light intensity of the entire pixel is evaluated.

[0003]

In the first conventional example,
The deeper the degree of blurring of the focus, the wider the light intensity distribution area, and the greater the number of sensor pixels involved in this distribution area, so the light intensity detection error can be reduced. Since the level also decreases, the detection limit occurs. For this reason, it is difficult to bring the light intensity distribution state into a practical uniform distribution state, and as a result that the detection level varies depending on the positional relationship between the LCD pixel and the sensor pixel, there is a drawback that the detection error increases. Further, as the degree of blurring of the focus is deepened, the number of neighboring pixels to be turned off increases, and the number of inspection screens for inspecting all the display pixels increases. As a result, the inspection time increases. Further, in the second conventional example, since the setting error when the positional relationship between the display element and the sensor camera is set to the reference position becomes the detection level error, it is necessary to strictly match the relative positions. Practically, there is a drawback that the detection accuracy is deteriorated as a result of a corresponding displacement. The present invention solves the above-mentioned drawbacks, can reduce the number of inspection screens, and provides a method capable of increasing the luminous intensity detection level accuracy in a state where the alignment accuracy of the display element to be inspected is relaxed, and more economical, It is to provide an efficient display element inspection method.

[0004]

In the conventional example, a part of light emitted from a display pixel is detected by a single sensor pixel, and the light intensity of the whole pixel is inspected and evaluated by the detected value.
The detection capability is increased by increasing the detection amount of the light amount emitted from the display pixel with a plurality of sensor pixels. That is, in an inspection method for determining the quality of a display screen of a display element as an inspection screen, (1) a display pixel to be inspected is brought into a bright state, and a pixel near the display pixel to be inspected is darkened. The display element inspection screen reading method according to claim 1, wherein the light amount of the pixel to be inspected is the integrated value of the detected light amount of each sensor pixel of the sensor pixel group related to the pixel image, and (2) each pixel image Is set so that the sensor pixel group related to the pixel image area fits in the area including the pixel image area and the adjacent pixel image area, and the integrated value of the detection light amount of each sensor pixel of the sensor pixel group related to the pixel image The present invention constitutes the second aspect of the invention, which is a display element inspection screen reading method in which the amount of light is used.

[0005]

Since the present invention is configured as described above, a wide range from the peak to the skirt of the luminous intensity distribution characteristic of the pixel image region of the pixel to be inspected can be detected as the sense light amount by using a plurality of sensor pixels. The inspection and evaluation can be performed with a smaller error with respect to the variation of the relative position between the display element and the sensor element. Further, it is possible to detect a defect in a partial area of the display pixel as an insufficient amount of light. Furthermore, by setting the inspection target pixel in the bright state and setting the pixels in the vicinity of the inspection target pixel in the dark state, the amount of light from the neighboring pixels is reduced,
Since the ratio of the amount of light from the inspection target pixel to be sensed can be increased, the inspection can be performed accurately. Also,
By limiting the size of the sensor pixel integration area, if three or less consecutive pixels in the one-dimensional direction are detected as abnormal, it can be considered as a single defective pixel. A pixel exhibiting an abnormal peak value can be regarded as a defective pixel.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the drawings.
A liquid crystal display (LCD) is used as the display element, and C is used as the sensor.
The present invention will be described in detail below with reference to the drawings by taking the case of using a CD two-dimensional sensor as an example. The LCD pixel and
CCD pixels have a two-dimensional spread in both x and y directions that are orthogonal to each other. However, if it is explained in a two-dimensional spread, it becomes unnecessarily complicated and hinders understanding. Furthermore, a two-dimensional spread is easily explained by focusing on only one of the x and y directions. Since the case can be expanded and understood, the present description will be given with a one-dimensional spread unless otherwise specified. Note that in this specification, the intensity of light emitted from a display pixel is expressed by a term called luminous intensity. The luminous intensity is an arbitrary unit, and the intensity of light emitted from the display pixel per unit area and the intensity of light of the pixel image formed on the light receiving surface of the sensor per unit area are all expressed in terms of luminous intensity. When it is desired to emphasize the amount obtained by integrating the light intensity over a certain region, the light amount may be used in some cases. In this specification, a pixel of a display element to be inspected such as an LCD pixel is simply called a pixel, and a sensor pixel such as a CCD pixel is called a sensor pixel.

FIG. 1 is a diagram for explaining the principle of the present invention. In the figure, the luminous intensity distribution 2 of the pixel image in the state where each pixel is lit is shown in relation to the pixel image region 1, and further the sensor pixel 3 is shown in positional relation with these. Here, in the luminous intensity distribution, when the luminous intensity in the lighting state is relatively large, the luminous intensity at the central portion is strong, and the luminous intensity becomes lower as it approaches the end portion, and this characteristic is shown. The present invention provides a method for inspecting accurately even when the number of sensor pixels corresponding to one pixel to be inspected is small. In this example, one pixel image area corresponds to approximately two sensor pixels. It is arranged at a certain ratio. It is difficult to make the relative positional relationship between the pixel image and the sensor pixel the same for all the pixels over the entire surface of the display element. In some cases, however, two sensor pixels may fit exactly in a single pixel image area. In most cases, the sensor pixels at both ends overlap the adjacent pixel image areas as seen in this example. The pixel image area of the pixel of interest on the sensor pixel surface is shown by 1 _T , and the luminous intensity distribution characteristic of the image area is shown by 2 _T. Also, the pixel image areas of the left and right pixels adjacent to the pixel of _interest are indicated by 1 _NL and 1 _NR , and the luminous intensity distribution characteristics of the image area are indicated by 2 _NL and 2 _NR , respectively. Also it shows a sensor pixel according to the target pixel in 3 ₁ to 3 _7. In order to detect all the light emitted from the pixel to be inspected, the light amounts detected by all the sensor pixels involved in the pixel image to be inspected are added. That is, read as the amount of light emitted in the pixel to be inspected with the amount obtained by adding the amount of light 4 _2, 4 _1, 4 ₃ detected in the figure the sensor pixel 3 _3, 3 _4, 3 _5. In this case, if the adjacent pixels are lit, the sensor pixels 3 ₃ and 3 ₅ also take in the light amounts 5 _L and 5 _R from the unnecessary adjacent pixels, resulting in an error. In the above explanation, 3 ₃ ,
Although the example in which the detection light amounts of the three sensor pixels 3 ₄ and 3 ₅ are integrated has been described, in general, if the light intensity distribution characteristics are widened, it is desirable to increase the number of sensor pixels forming the sensor pixel group for detection. However, when the luminous intensity distributions spread over the adjacent pixel regions, the unnecessary light amounts from the adjacent pixels are integrated, which also causes the detection accuracy to deteriorate. Therefore, in order to eliminate the light amount from the neighboring pixels and accurately inspect the luminous intensity in the lighting state, it is sufficient to turn off the neighboring pixels and reduce the light emitted from the neighboring pixels. That is, when systematically inspecting all the pixels, it is systematic and convenient to read and inspect the display screen while displaying the distributed lighting inspection pattern q times for every q pixels. In the above description, the pixel is, for example, X.
In the description in the case of one-dimensional arrangement in the direction, generally in the Y direction orthogonal to this direction, the lighting state is set every r pixels.
Explaining this two-dimensional array, q
・ Displaying the inspection pattern r times to inspect all pixels. In this inspection method, if the neighboring pixels cannot control the switching between the extinguished state and the lit state, and if there is a defect that is always lit in the adjacent pixels, all pixels can be controlled to the extinguished state and controlled. Normal pixels are in the off state, and defective pixels that cannot be controlled are in the on state.Display information is read and stored as background information in advance, and the background information is subtracted from the display information in each distributed lighting inspection pattern. The defective pixel that is always lit can be turned off equivalently, and even if there is a defect in a neighboring pixel, the luminous intensity and the light amount of the inspection target pixel can be accurately inspected.

Next, a method of inspecting the light quantity in the off state will be described. What is a defective pixel in the off state?
A pixel that emits a larger amount of light than the original amount of light in the off state. FIG. 2 shows the correspondence relationship between the pixel image area 1 and the sensor pixel 3 in the case where one pixel A is isolated and becomes defective and emits an abnormal light amount, and the neighboring pixels B and C are normal. In this example, a case where approximately two pixels correspond to one pixel image area is shown. In addition, the luminous intensity distribution at a low luminous intensity level in the off state is generally close to a uniform distribution, unlike the luminous state distribution in the lit state. Therefore, here, the pixel is described as a uniform distribution. 3 ₄ The sensor group of pixels integrated detects the amount of the defective pixel A centering 3 ₃ -
If there are three 3 ₅ sensor pixels, the sensor pixel group for detecting the light amount of the B pixel adjacent thereto is 3 _{1 to} 3 ₃ , and 3 for the C pixel.
₄ will be set to three to 3 _7. The basic condition of this configuration is that the integrated sensor pixel group is set within the area formed by the pixel image area to be inspected and the adjacent pixel image area. As far as in this example, the light intensity of the defective pixel can be correctly detected by the integrated sensor pixel group consists of 3 _{3-3 5,} abnormal amount of defective pixels in sensor pixel 3 ₃ a light quantity detection without pixel B defect 4 _B is mistakenly detected. Further, even in the pixel C having no defect, the sensor pixel 3 ₅ erroneously detects the abnormal light amount 4 _C of the defective pixel. This erroneously detected abnormal light amount changes depending on the corresponding positional relationship between the pixel image area and the sensor pixel. In general, the correspondence between the pixel image area and the sensor pixel is not unique, and changes spatially due to lens distortion or the like. It also changes depending on the number of corresponding sensor pixels per pixel image. In particular, when the number of corresponding sensor pixels per pixel image is not an integer, the relative position of the sensor pixel slightly shifts every time the pixel position shifts by one pixel even if there is no distortion in the lens. Therefore, the abnormal light amount that is erroneously detected differs depending on the pixel position. That is, depending on the corresponding positional relationship, the amount of light that is erroneously detected differs between pixels on both sides, and one may be large and the other may be small. In some cases, the amounts are almost equal. Further, both detection amounts may be extremely small. That is, depending on how to select a threshold level for detection, a single defect may be detected as 3 to 2 including adjacent pixels, and only one true defective pixel may be detected as a defect. As is clear from the above description, if the size of the integrated sensor pixel group is set so that the integrated sensor pixel group related to each pixel image fits within the area composed of the pixel image area and the adjacent pixel image area, three consecutive pixels are obtained. When a defect equal to or less than the pixel is detected, it can be regarded as a single defect, and the pixel having the maximum abnormal light amount can be determined as the defective pixel.

In the above embodiments, the description has been given of the 2-gradation display element which has two states of the lighting state and the extinguishing state. However, even in the case of multi-gradation, the bright state which is relatively brightened is changed to the lighting state. The same inspection can be performed by making the dark state, which is made relatively dark, correspond to the unlit state. As is clear from the above description, it is important to associate sensor pixels with pixel images. The sensor pixels corresponding to each LCD pixel image generate correspondence relation information in advance and hold it in the storage device, and perform integration processing using this correspondence relation information. The correspondence information is generated, for example, by a method of using a non-defective display element and regarding a sensor pixel that has detected the light amount of each pixel as a peak value as a sensor pixel corresponding to the pixel. In the case of this method, the setting position of the work, which is the display element to be inspected thereafter, is almost the same as the position of the display element initially set for generating the correspondence information.

Correspondence relationship information is information for associating the address B of the sensor pixel corresponding to the pixel of the address A. For example, the address A of the memory device is made to correspond to the address A of the pixel, and the content of the address A of the memory device is set to B. Is stored as the correspondence information between the pixel and the sensor pixel. Therefore, in this case, in order to detect the light amount of the pixel of the address A, the content B of the address A of the storage device is read out,
The LCD of the address A is generated by integrating the detected light amount of the sensor pixel while generating the addresses of a predetermined number of sensor pixels before and after the sensor pixel of the address B.
It is the light amount of the pixel.

[0011]

According to the present invention as set forth in claim 1, a display pixel to be inspected is brought into a bright state, and a pixel in the vicinity of the display pixel to be inspected is brought into a dark state, and a sensor relating to the pixel image concerned. The display element inspection screen reading method, which uses the integrated value of the detected light amount of each sensor pixel of the pixel group as the light amount of the inspection target pixel, can detect most of the light amount from the inspection target pixel and can reduce the light amount from the adjacent pixel. There is an effect that can be accurately inspected. According to a second aspect of the present invention, the sensor pixel group related to each pixel image is set so as to fit within the area including the pixel image area and the adjacent pixel image area, and each of the sensor pixel groups related to the pixel image is set. The display element inspection screen reading method, which uses the integrated value of the detected light amount of the sensor pixel as the light amount of the inspection target pixel, can distinguish the isolated defect from the continuous multiple defects and detect the defect position and its light amount. It has the effect of increasing ability.

[Brief description of drawings]

FIG. 1 is a principle explanatory view for explaining a light amount detection method by an integration method using a plurality of sensor pixels according to the basics of the present invention. The correspondence between the pixel image area and the sensor pixel and the relationship between these and the luminous intensity distribution characteristic in the lighting state are shown.

FIG. 2 is a diagram for explaining an inspection method in an off state, showing a correspondence relationship between a pixel image region, a sensor pixel, and a luminous intensity distribution characteristic.

[Explanation of symbols]

 1 Pixel image area 2 Luminous intensity distribution characteristics 3 Sensor pixels 4, 5 Light intensity detected by sensor pixels

Claims (2)

[Claims] 1. A display pixel to be inspected is set to a bright state, and pixels in the vicinity of the display pixel to be inspected are set to a dark state to detect the amount of light detected by each sensor pixel of a sensor pixel group related to the pixel image. Display element inspection screen reading method characterized in that the integrated value is used as the light amount of the inspection target pixel 2. The detection light amount of each sensor pixel of the sensor pixel group related to the pixel image is set so that the sensor pixel group related to each pixel image fits into a region consisting of the pixel image area and its adjacent pixel image area. Display element inspection screen reading method characterized in that the light amount of the pixel to be inspected is determined by the integrated value of