KR20050033164A - Automatic alignment method for visual quality inspection of fpd - Google Patents

Abstract

An automatic alignment method for inspecting picture quality of a flat panel display is provided to obtain a boundary in a flat panel display image in consideration of dots and a black matrix of the flat panel display to calculate alignment errors, to thereby align the flat panel display with a camera. A basic posture of the alignment state of a flat panel display loaded on an inspection stage and the posture of an alignment error state corresponding to the basic posture are detected and compared to each other such that dots in an image obtained by photographing the flat panel display is discriminated from a black matrix in the image, to calculate alignment error data(s108). When the calculated alignment error data is larger than an allowable error, the inspection stage on which the flat panel display is loaded is automatically moved and aligned(s114).

Description

Automatic Alignment Method for Visual Quality Inspection of FPD}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image quality inspection system for a flat panel display (hereinafter, abbreviated as 'FPD'), and more particularly, to a liquid crystal display (LCD) or plasma display device (PDP) completed through a manufacturing process. The present invention relates to an automatic alignment method for inspecting the quality of an FPD such that the quality of the FPD is automatically inspected using a CCD (Charge Coupled Device) camera.

In general, the FPD is necessarily inspected image quality during the manufacturing process, the power of the manufactured FPD is turned on (on), after applying a predetermined pattern for inspection, the operator directly inspected the image quality with the naked eye.

That is, the operator visually inspects the image quality of the FPD that is automatically (or manually) loaded / unloaded, and then checks the quality of the FPD by checking the defective pixels or the image quality.

Small FPDs can be manually unloaded after being manually loaded by the operator to inspect pixels / image quality.

As such, the pixel / image quality inspection method by the operator is visually inspected by the operator, so it is not uniform according to the skill, concentration, education, and health status of the operator. There was an issue that affected.

Automated quality inspection apparatus according to the prior art is disclosed in the Republic of Korea Patent Registration No. 10-0187015, 10-0187016 (automatic quality inspection apparatus and method of the LCD module).

1 is a block diagram of an automated quality inspection apparatus of an LCD panel according to the prior art,

The image processing unit 3 for storing and processing the image data of the LCD panel 2 acquired by the CCD camera 1 and the position of the LCD panel 2 are horizontal (x), vertical (y), and twist angle ( LCD driver 4 for converting in the direction of [theta]), a position controller 5 for outputting a control signal to the LCD driver 4, and an LCD controller 6 for displaying a standard pattern for inspection on the LCD panel 2; It consists of).

In addition, a camera adjusting unit 7 which performs automatic focusing of the CCD camera 1, an illuminating unit 8 illuminating an LCD panel 2 seated on the LCD driving unit 4, and the image processing unit ( 3) The control unit outputs control signals to the LCD control unit 6 and the position control unit 5 so as to generate a standard pattern and to drive a measurement system. 10).

In the conventional LCD panel automatic image quality inspection device configured as described above, the LCD panel 2 is horizontally aligned by the position control unit 5 and the LCD driver 4 as shown in the flowchart shown in FIG. A first step and a second step in which the camera adjusting unit 7 automatically focuses the focus of the CCD camera 1, and a second step of eliminating aliasing which is an optical interference phenomenon between the CCD camera 1 and the LCD panel 2; There are three stages.

In addition, a fourth step in which the main computer 10 outputs a control signal to the LCD control unit 6, a fifth step in which the LCD control unit 6 outputs a reference pattern to the LCD panel 2, and the LCD panel A sixth step of acquiring the image output to (2) by the CCD camera (1) and outputting it to the main computer (10); the main computer (10) outputs a control signal to the position control unit (5) and the position The controller 5 controls the LCD driver 4 so that the LCD panel 2 is moved in position, and an image of the LCD panel 2 is input and stored from the CCD camera 1 to the image processor 3. And an eighth step of analyzing defects of the LCD panel 2 to determine whether there are good or bad, and a ninth step of displaying the result on the monitor 9.

The prior art made as described above has a problem in correcting an error generated when the LCD panel and the CCD camera are aligned by controlling and aligning positions without considering image specificity of the LCD panel.

The present invention is to solve the above problems of the prior art, the purpose of the FPD image in consideration of the dot (light emitting part) and the partition (BM: Black Matrix) of the FPD to match the position of the FPD and the camera It is to provide an automatic alignment method for FPD image quality inspection that calculates and aligns alignment errors (horizontal error, vertical error, and error angle) that can be corrected by obtaining a boundary line.

In order to achieve the object of the present invention, an automatic alignment process for checking the image quality of the FPD according to the present invention generates a pattern after loading the inspection table of the image quality inspection apparatus to inspect the image quality of the manufactured FPD, In the FPD image quality inspection method of matching the photographed image with the camera position, and checking the image quality, the reference position of the alignment state of the FPD and corresponding to it so that the light emitting body and the partition wall area are distinguished from the image of the FPD loaded on the inspection table Detecting and comparing the posture of the alignment error state to calculate alignment error data, and automatically adjusting and aligning the FPD-loaded inspection table when the calculated horizontal error, vertical error, and error angle are more than the allowable error. Characterized in that it comprises a second process.

Here, the first process is a step of calculating three reference points when the picked-up FPD is in alignment, a second step of detecting the picked-up FPD image horizontal / vertical border, the image at the horizontal / vertical border, Three steps of calculating an approximation equation of each straight line from the four-direction boundary lines of the lower, left and right, four steps of calculating the intersection points of the straight lines corresponding to the three reference points calculated in the first step, and the three reference points And five steps of calculating an alignment error by comparing three intersection points selected to correspond thereto.

The present invention made as described above will be described in detail with reference to the accompanying drawings.

3 is a block diagram of an image quality inspection apparatus for implementing an automatic alignment method of an FPD according to an embodiment of the present invention. The inspection table 122 and the inspection table 121 for loading / unloading an inspection target FPD are shown in FIG. CCD camera 101 for photographing the FPD 121, an autofocus unit 102 added to the CCD camera 101 to adjust auto focus for accurate imaging of the FPD 121, and the CCD camera 101. The position correction data is analyzed by using the image captured by the camera to calculate the position correction data 'horizontal error (x), vertical error (y), error angle (θ)' and the CCD camera 101 by the calculated correction data. A control unit 110 for generating a control signal so as to match the position of the FPD 121 with the image pickup area, transmitting the image and inspection results to the display monitor, and transmitting image quality inspection result data to the production management server; The pattern of the FPD 121 is controlled according to the control of the controller 110. A test table controller 123 for adjusting the position of the test table 122 in the 'x, y, θ' direction by the pattern generator 111 and the position correction data calculated by the controller 110. It is composed of

Here, the inspection stand controller 123 adds a function of adjusting the upper and lower positions z between the CCD camera 101 and the inspection counter FPD 121 according to its size when loading the FPD 121.

The operation according to the embodiment of the present invention configured as described above will be described in more detail with reference to FIGS. 3 to 8 as follows.

First, according to the present invention, as the target FPD 121 for image quality inspection is loaded on the inspection table 122, the distance between the loaded FPD 121 and the CCD camera 101 is determined by the inspection table controller 123. Keep it.

The CCD camera 101 captures the FPD 121 loaded on the examination table 122 after auto focusing by the auto focus unit 102.

4 is a flowchart of an automatic alignment process for checking the image quality of the FPD according to an embodiment of the present invention. As shown in the drawing, as the FPD 121, which is the image quality inspection target, is loaded on the test table 122, the controller 110 Generates a predetermined pattern by controlling the pattern generator 111 (S102) (104).

The controller 110 receives an image of the FPD 121 from the CCD camera 101 (S106).

In the received FPD 121 image, an alignment error between the FPD 121 and the CCD camera 101 is calculated to distinguish the light emitting area and the partition wall area from each other (S108).

In order to calculate the alignment error, first, three reference points of the aligned state of the FPD 121 are preset (FIG. 5A), and three corresponding points are selected from the imaging area of the FPD 121 actually photographed. After the calculation, the alignment error is calculated as compared with the three reference points. In this case, if the alignment error is within the tolerance, the alignment process is not required, and thus the quality inspection is performed.

5A is a flowchart of a process of calculating three reference points after an operator manually aligns a loaded FPD.

As shown in the figure, after the operator loads the FPD, the dots of the FPD and the partitions are identified by a display monitor (not shown) and the positions of the FPD and the CCD camera are aligned.

The aligned FPDs are output to the imaging and control unit 110 through the CCD camera 101 to calculate three reference points for calculating the alignment error of the inspection target FPD 121.

In other words, each boundary line of the FPD in the manually aligned state is determined based on the captured image of the FPD received from the CCD camera 101 (S132).

An approximation equation for four straight lines is calculated from the boundary lines in the four directions of up, down, left, and right directions (S134).

The four intersection points '(up, left), (up, right), (down, left), (bottom, right)' are calculated from the approximation equations for the four straight lines (S136).

Three of the calculated four intersections are selected and set as reference intersections (S138).

The three reference points set through the above process are provided as alignment reference data.

FIG. 5B is a flowchart illustrating a process of determining a misalignment after calculating and comparing three reference points calculated in FIG. 5A and three corresponding points, respectively, in order to detect an alignment state of the loaded inspection target FPD.

As shown in the figure, the upper and lower horizontal (H) and left and right vertical (V) boundary lines of the captured image of the inspection target FPD 121 received from the CCD camera 101 are determined.

An approximation equation for four straight lines is calculated from the boundary lines in the four directions of up, down, left, and right. The approximate expression is calculated as y = ax + b, x '= ay' + b (S144).

Each intersection (up, left), (up, right), (down, left), (down, right) is calculated from the approximation equations for the four straight lines. (S146)

Three intersections corresponding to the preset three reference nodes are selected from the calculated four intersections (S148).

Comparing the three intersection points of the reference three intersection points (S138) calculated and set in FIG. 5A and the inspection target FPD 121 calculated in the above process, the degree of distortion is equal to the horizontal error (x), the vertical error (y) and the error angle. It calculates by (theta). (S150)

Here, the method of calculating the alignment error data (x, y, θ) is to be calculated by the Affine Transformation using three pairs of corresponding intersections.

It is determined whether the FPD 121 and the CCD camera 101 are distorted according to the calculated alignment error (x, y, θ) value (S110).

If the alignment errors (x, y, θ) are all within the tolerance, it is determined that the FPD 121 and the CCD camera 101 are in the alignment state, and the image quality inspection process is performed. S116)

When the image quality test is completed, the FPD 121 is unloaded. (S120)

However, when the alignment error (x, y, θ) occurs, the control unit 123 controls the inspection stand to adjust the error, and then returns to the FPD 121 imaging step (S106) of the CCD camera 101. In order to proceed with the alignment error calculation process again (S114).

FIG. 6 is a view showing an FPD image captured according to an exemplary embodiment of the present invention. Each partition wall is formed between pixels R, G, and B that emit light, and each block includes a CCD cell 140 of a camera. Is displayed.

7 is a diagram illustrating camera pixels, RGB dots, and barrier ribs according to an exemplary embodiment of the present invention.

As shown here, the CCD of the camera is composed of 2000 X 2000 pixels, and the FPD is composed of RGB dots and barrier ribs.

The postures of the FPD 121 and the CCD camera 101 configured as described above are judged, and the inspection table 122 is controlled and aligned as much as the calculated alignment error.

8 is a graph showing the degree of display and brightness of the FPD 121 image according to the degree of distortion in accordance with an embodiment of the present invention.

Table 1 below is an embodiment showing the degree of distortion of the FPD 121 and the CCD camera 101.

Mis-alignment error θ L 0.01 ° ㎛10㎛

In addition, as illustrated in the graph of FIG. 8, when misalignment occurs, defect detection of a light emitting pixel becomes difficult.

That is, due to the structural factors (sequential arrangement of dots and BM) of the FPD 121 itself, the inspection degree of all the quality inspection items is changed according to the degree of alignment error of the sample.

Although a preferred embodiment according to the present invention has been described above, the present invention may be modified in various ways to align FPDs and CCD cameras for image quality inspection, and those skilled in the art may claim It is understood that various modifications and variations can be made without departing from the scope.

As described above, the automatic alignment method for checking the image quality of the FPD according to the present invention, after the image alignment with the CCD camera considering the dots and partitions of the FPD in order to automatically check the image quality of the FPD to the captured image of the FPD By allowing the image quality to be inspected, there is an effect of automatically inspecting all the items for the image quality inspection such as dots, foreign matter defects and homogeneity defects.

1 is a block diagram of an automated quality inspection apparatus of an LCD panel according to the prior art,

Figure 2 is a flow chart of the automated quality inspection process of the LCD panel according to the prior art,

Figure 3 is a block diagram of an automated quality inspection device of a flat panel display for implementing the present invention,

4 is a flowchart of an automatic alignment process for inspecting image quality of a flat panel display according to an embodiment of the present invention;

FIG. 5A is a flowchart of a process of calculating reference intersection data in an aligned state for calculating the degree of distortion of the flat panel display and the CCD camera in FIG. 4;

FIG. 5B is a flowchart of a process of calculating the degree of distortion of the flat panel display and the CCD camera in FIG. 4;

6 is a diagram showing a flat panel display image captured according to an embodiment of the present invention,

7 is a view showing camera pixels, RGB dots, and barrier ribs according to an embodiment of the present invention;

8 is a graph showing a flat display image display degree and brightness degree according to the degree of distortion according to an embodiment of the present invention.

<Description of Symbols for Major Parts of Drawings>

101: CCD camera 102: auto focus

110: control unit 111: pattern generating unit

121: flat panel display 122: inspection table

123: control unit 140: CCD cell

Claims (7)

In the flat panel display image quality inspection method, which is loaded on the inspection table of the image quality inspection device to inspect the image quality of the manufactured flat panel display, the pattern is generated, and the image quality of the flat panel display is matched with the camera position using the camera. In A first process of calculating alignment error data by detecting and comparing a reference posture of an alignment state of the flat panel display and a posture of an alignment error state corresponding to the light emitting body and the partition wall area in the image of the flat panel display loaded on the inspection table. ; And And a second process of automatically adjusting and aligning the inspection table loaded with the inspection target flat panel display when the calculated horizontal error, vertical error, and error angle are more than the allowable error. How to sort. The method of claim 1, The first step may include: calculating three reference points when the photographed flat panel display is in an aligned state; Detecting a four-direction boundary line of upper, lower, left, and right sides of the photographed inspection target flat panel display image; Calculating an approximation equation of each straight line from the four-direction boundary lines of the upper, lower, left and right; Calculating four intersections from the four approximations calculated in the third step and selecting three intersections to correspond to the calculated three reference points in the first step; And And a five step of calculating an alignment error by comparing the three reference points with the three points selected in step 4 so as to correspond to the three reference points. The method of claim 2, The step of calculating the three reference intersections in the first step is an automatic alignment method for the image quality inspection of the flat panel display, characterized in that automatically calculates three intersections after manually operating the alignment state. The method of claim 2, In the step 3, each boundary line is calculated by an approximation equation of a straight line. The method of claim 2, The intersection point calculated in step 4 is '(up, left), (up, right), (bottom, left), (bottom, right)' automatic sorting method for image quality inspection of a flat panel display. The method according to claim 1 or 2, The horizontal error (x), the vertical error (y), the error angle (θ) is calculated by the affine transformation (Affine Transformation) method using three pairs of matching points, automatic alignment for the quality test of the flat panel display Way. The method of claim 1, In addition to the first process, the inspection target flat panel display loaded on the inspection table further includes adjusting a vertical distance to maintain a proper distance from the CCD camera according to the dot size. Automatic sorting method for.