JPH0792456A - Substrate alignment method of laminated optical pannel - Google Patents

Abstract

PURPOSE:To provide a substrate lignment method capable of reducing an aperture pattern provided in a alight shielding area. CONSTITUTION:One side substrate A forming an alignment mark consisting of the aperture pattern 7 with a window frame shape in the light shielding area 6 is set under a camera 5 in a first procedure. In a second procedure, the other substrate B forming the alignment mark consisting of a geometric pattern 8 with a center pointing shape is superposed and set on the one side substrate A. In a third procedure, the camera is operated, and both alignment marks superposed in each other are projected on an image recognition visual field. In a fourth procedure, the projected aperture pattern 7 is image recognized, and the position of the one side substrate A is detected. In a fifth procedure, the part of the geometric pattern 8 projected in the aperture pattern 7 is image- recognized, and arithmetic operation is performed to calculate the central point of the center pointing shape, thereby the position of the other substrate B is detected. finally, in a sixth procedure, both substrates A, B are aligned relatively based on both detected positions.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a substrate alignment method for a laminated optical panel represented by an active matrix liquid crystal panel or the like. More specifically, the present invention relates to a technique for automatically aligning both substrates by image recognition processing using alignment marks formed on a pair of substrates.

[0002]

2. Description of the Related Art First, the structure of a laminated optical panel will be briefly described with reference to FIG. 13 by taking an active matrix liquid crystal panel as an example to clarify the background of the present invention. The active matrix liquid crystal panel has a laminated structure in which a pair of substrates are bonded to each other through a predetermined gap and liquid crystal is held in the gap. As shown in (A), one substrate A has an effective screen area 101 and a peripheral area 102. In the effective screen area 101, switching elements including pixel electrodes and thin film transistors (TFTs) arranged in a matrix are integrally formed. Further, in the peripheral region 102, for example, a drive circuit for operating the switching element is integrally formed. Effective screen area 10
A light shielding film 103 is formed in the peripheral region 102 surrounding 1 to prevent light leakage and serves as a light shielding region. A pair of alignment marks 104 are formed in this light shielding region. In addition, a lead electrode 105 for external connection is formed at the lower end of the substrate A.

As shown in FIG. 3B, the other substrate B also has a central effective screen area 106 and a peripheral area 107 surrounding it. In the effective screen area 106, color filters and counter electrodes divided into RGB primary color segments are formed. A pair of alignment marks 108 are formed in the peripheral area 107. Both substrates A and B are bonded in alignment with each other using alignment marks 104 and 108, and liquid crystal is held between them to form an active matrix liquid crystal panel. As a result, the pixel electrode on the substrate A side and the color filter segment on the substrate B side are precisely aligned.

FIG. 14A shows the peripheral region 10 of the substrate A.
2 shows an example of the alignment mark 104 formed in 2. As shown in the figure, the alignment mark 104 is arranged in a window frame 109 provided by partially removing the light shielding film 103 by etching. Therefore, the alignment mark 104 can be imaged from the outside. On the other hand, FIG. 14B shows an example of the alignment mark 108 formed in the peripheral region 107 of the substrate B.

FIG. 15 shows a state where both substrates A and B are superposed. In the window frame 109 in which the light shielding film 103 is partially removed by etching, both alignment marks 104,
108 has appeared. Window frame 1 opened in the light shielding film 103
Both alignment marks 104 and 108 are imaged via 09 and are projected in the image recognition visual field. Image recognition is performed on the projected alignment marks 104 and 108, and the relative positions of the two substrates A and B are automatically adjusted so that the two align. This figure shows a state where the automatic alignment by the image recognition processing is completed, and the effective screen area 101 on the substrate A side and the effective screen area 106 on the substrate B side are aligned with each other within an allowable error range.

[0006]

FIG. 16 shows alignment marks 104, 1 projected in the image recognition visual field 110.
08 represents the state before automatic alignment. In the illustrated example, the light-shielding film is removed over a wide range, the opening area of the window frame is wide, and the light-shielding film is not present in the image recognition visual field 110. Therefore, the pair of alignment marks 104 and 108 can be clearly displayed in the image recognition visual field 110 regardless of the presence of the light shielding film. However, if the light-shielding film is removed over a wide range without limitation, light leakage may occur and the operating performance of the active matrix liquid crystal panel may be impaired.

On the other hand, FIG. 17 shows an example in which the removal area of the light shielding film 103 is extremely limited. That is, the window frame 109 opened in the light shielding film 103 also serves as the one alignment mark 104. In this case, since the amount of light leakage is extremely small, the operating performance of the active matrix liquid crystal panel is not adversely affected. However, since the other alignment mark 108 is hidden under the light shielding film 103, the image cannot be recognized. In order to enable image recognition with this configuration,
Prealignment must be performed mechanically in advance so that the alignment mark 108 appears in the window frame 109. However, it is practically extremely difficult to perform high-precision mechanical pre-alignment.

Therefore, in practice, the alignment mark structure as shown in FIG. 18 is adopted. That is, when the formation position of one alignment mark 104 is limited to the light shielding film 103 due to the structure of the panel, the light shielding film 103 is removed from the periphery of the alignment mark 104 in an appropriate range and the window frame 109 is provided. There is. In the pre-aligned state, the other alignment mark 1 is placed in the window frame 109.
08 is supposed to appear. Therefore, both alignment marks 104 and 108 are visible in the image recognition visual field 110 in a visible state.

Next, a method of setting the opening area of the window frame 109 will be described with reference to FIG. Center 10 of the alignment mark provided on the substrate side on which the light shielding film 103 is formed
4A, the center 108B of the alignment mark formed on the other substrate side varies in the range 111 shown by the dotted line. In other words, the accuracy of mechanical pre-alignment is about the range 111 shown by the dotted line. Further, the other substrate B
The appropriate opening area of the window frame 109 is determined in consideration of the size and margin of the alignment mark formed on the side. If an increase in the board size tolerance and a decrease in mechanical pre-alignment accuracy are recognized from the viewpoint of manufacturing cost, the opening area of the window frame 109 must be increased accordingly. Therefore, there is a problem that the amount of light leakage increases and the operating characteristics of the active matrix liquid crystal panel are impaired.

[0010]

SUMMARY OF THE INVENTION In view of the above-mentioned problems of the prior art, an object of the present invention is to enable automatic alignment by image recognition processing with a minimum required window frame opening area. The following measures have been taken in order to achieve this object.
That is, the substrate alignment method of the laminated optical panel according to the present invention is performed by the following procedure. First, as a first procedure, one substrate having an alignment mark formed of a window frame-shaped opening pattern in the light-shielding region is set under the camera. Next, as the second procedure, the other substrate on which the alignment mark formed of the geometric pattern of the center-directed shape is formed is set so as to overlap the one substrate. Then, as a third procedure, the camera is operated to display both overlapping alignment marks in the image recognition visual field. Next, as a fourth procedure, the position of one of the substrates is detected by image-recognizing the projected opening pattern. Further, as a fifth step, the position of the other substrate is detected by recognizing the image of the geometric pattern portion projected in the opening pattern and performing arithmetic processing to calculate the center point of the center-directed shape. In addition, the above-mentioned fourth
The order of execution of the procedure and this fifth procedure does not matter. Finally, in a sixth step, relative alignment of both substrates is performed based on the detected positions. Preferably, the geometric pattern has a concentric or radial centered shape. The opening pattern has, for example, a circular or rectangular window frame shape.

The present invention includes not only the above-mentioned substrate alignment method, but also a laminated optical panel having a characteristic alignment mark. That is, the laminated optical panel according to the present invention is composed of a pair of substrates bonded to each other while being aligned with each other. As a characteristic feature, one of the substrates includes a light-shielding region, and an alignment mark composed of a window frame-shaped opening pattern is formed therein. The other substrate has an alignment formed of a geometric pattern with a center-directed shape. A laminated optical panel having such a structure includes, for example, a substrate on which pixel electrodes and switching elements are integrally formed, a substrate on which a color filter is formed, and a liquid crystal held in a gap between the active matrix liquid crystal panel. Make up. However, the present invention is not limited to this, and can be applied to a laminated optical panel in which at least one substrate has a light shielding region.

[0012]

According to the present invention, the alignment mark formed on one of the substrates is a window frame-shaped opening pattern provided in the light shielding region. That is, it is sufficient that the window frame itself is used as an alignment mark and the size of the opening pattern is projected within the image recognition field of view, and the minimum required size. Therefore, the amount of light leakage is extremely small, and the operational characteristics of the laminated optical panel are not substantially adversely affected. The opening pattern has a relatively simple rectangular or circular window frame shape, and the center point can be detected by a simple algorithm by image recognition. On the other hand, the other alignment mark has a large area enough to cover the mechanical pre-alignment accuracy, and a part of it is always projected in the opening pattern. The alignment mark is composed of, for example, concentric or radial center directional geometric patterns, and only a part of the alignment mark can be image-recognized and arithmetically processed to calculate the center point of the alignment mark from the geometrical relationship. That is, the position of the substrate can be detected even if the entire image of the alignment mark is not displayed. The algorithm for calculating the center point of this geometric pattern is relatively simple. It becomes possible to align both substrates with high accuracy based on the center points of the opening pattern and the geometric points detected in this way.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 shows an outline of a substrate alignment apparatus used to carry out the substrate alignment method according to the present invention, and an example of alignment marks provided on the substrate. As shown in (A), the alignment apparatus includes a lower stage 1 and an upper stage 2 arranged in parallel with each other. The lower stage 1 is movable along the x axis, the y axis, and the z axis, and is rotatable about the z axis perpendicular to the optical axis. On the other hand, the upper stage 2 is fixed.

The laminated optical panel 3 which is the work to be aligned has a pair of substrates A and B. The laminated optical panel 3 is, for example, an active matrix liquid crystal display panel. In this case, the substrate A is integrally formed with pixel electrodes, switching elements and the like. Further, the peripheral portion thereof is covered with a light shielding film to form a light shielding region. On the other hand, on the substrate B, a color filter composed of RGB primary color segments and a counter electrode are formed. Both substrates A and B are bonded in a state of being aligned with each other. Therefore, for example, the substrate A is set on the lower stage 1 and the substrate B is set on the upper stage 2 to perform mutual alignment.

A camera block 4 is provided on the upper stage 2.
And is equipped with a pair of imaging cameras 5. The camera 5 captures an image of the alignment mark provided on the substrates A and B through the aperture provided on the upper stage 2 and displays it in the image recognition visual field. The pair of substrates A and B are mechanically pre-aligned in advance, and the alignment marks provided on both substrates are imaged in an overlapping state. Since the alignment marks are provided on both sides of each substrate, the camera block 4 also includes a pair of cameras 5. Further, in order to adjust the imaging range, the camera block 4 can move along the x-axis, the y-axis, and the z-axis, and can rotate about the z-axis.

FIG. 1B shows an alignment mark provided on the substrate A side. The alignment mark comprises a window frame-shaped opening pattern 7 provided in the light-shielding region 6. The opening pattern 7 has a relatively simple figure, and for example, a rectangle or a circle can be adopted.

(3) represents an alignment mark formed on the other substrate B. This alignment mark is composed of a geometric pattern 8 having a center-directed shape. In the illustrated example, the geometric pattern 8 has a radial shape. Instead of this, a concentric shape may be adopted.

FIG. 2 shows an alignment mark projected in the image recognition visual field 9. As shown in the drawing, a part of the geometric pattern provided on the substrate B side is projected through the opening pattern provided on the substrate A side. In this example, the opening pattern has a square window frame shape. FIG. 3 shows another example of the alignment mark projected in the image recognition visual field 9. Here, the opening pattern has a circular window frame shape. In the present invention, it is not necessary to show the whole geometric pattern provided on the substrate B side, and it is sufficient if a part of the geometric pattern can be recognized.
Therefore, the area of the opening pattern provided on the substrate A side can be set to be relatively small, and the amount of light leakage can be small.
Further, it suffices that the opening pattern on the substrate A side be within the image recognition visual field 9, and it is not necessary to consider the positional deviation of the geometric pattern on the substrate B side. Therefore, in the present invention, the image recognition visual field 9 can be determined within the variation range on the substrate A side. That is, it is only necessary to consider the dimensional tolerance of the substrate A and the pre-alignment accuracy. Therefore, the imaging range can be narrowed in advance, and the magnification of the image recognition visual field can be increased correspondingly, and the resolution can be improved.

Continuing to refer to FIGS. 1 and 2, the basic procedure of the substrate alignment method according to the present invention will be described in detail. First, as the first procedure, one substrate A having an alignment mark formed of a window frame-shaped opening pattern 7 in the light shielding region 6 is set under the camera 5. Specifically, it is adsorbed and fixed on the upper surface side of the lower stage 1. As the second procedure, the other substrate B on which the alignment mark formed of the geometric pattern 8 having the center-directed shape is formed is set so as to be superimposed on the above-mentioned one substrate A. Specifically, the substrate B is suction-fixed to the lower surface side of the upper stage 2. At this stage, mechanical pre-alignment is performed, and both alignment marks overlap each other. Next, as a third procedure, the camera 5 is operated to display both alignment marks overlapping each other in the image recognition visual field 9. As a fourth procedure, the position of the substrate A is detected by image-recognizing the projected opening pattern 7. For example, the center point of the rectangular opening pattern 7 may be calculated based on the image recognition result. Further, as a fifth step, the position of the substrate B is detected by recognizing the image of the geometric pattern 8 projected in the opening pattern 7 and performing arithmetic processing to calculate the center point of the center-directed shape. Finally, as a sixth procedure, relative alignment of both substrates A and B is performed based on the detected center point coordinates. Specifically, the lower stage 1 may be moved so that the center points of both substrates coincide with each other. The above-mentioned image recognition processing, arithmetic processing, drive control of the lower stage, etc. are all performed by a computer.

Next, with reference to FIG. 4, an outline of the algorithms of the image recognition processing and the arithmetic processing performed in the above-mentioned fourth procedure and fifth procedure will be described. In this example, a rectangular opening pattern 7 and a radial geometric pattern 8 are included in the image recognition visual field 9.
A part of is shown. In order to simplify the illustration,
The edge of the geometric pattern 8 appearing inside the opening pattern 7 is shown as 10. First, the opening pattern 7 is image-recognized with the light-shielding region 6 as the background. The center point of the rectangular window frame is calculated based on this recognition result. Thereby, the position of the substrate A can be detected. Next, the intersections a, b, ..., Chi between the edge 10 and each side of the opening pattern are image-recognized. Then, the corresponding pairs of intersections are detected and connected by straight lines. For example, one straight line can be obtained by connecting the intersection points C and D. Also, another straight line can be obtained by connecting the intersection e and f. At least 2
If you set up simultaneous linear equations for the straight line of the book and solve it,
The intersection point P of two straight lines is obtained. This intersection point P is the center point of the radial geometric pattern 8 and the position of the substrate B is detected.

Further, a concrete example of the above-described algorithm will be described in detail with reference to FIGS. First, as shown in FIG. 5, by utilizing the fact that the brightness of the opening pattern 7 greatly changes with respect to the background composed of the light-shielding region 6, the brightness is distributed in the x-axis and y-axis directions over the entire image recognition visual field 9. Obtain a cross-sectional waveform (luminance projection waveform). Further differentiate this waveform,
The minimum value x1 and the maximum value x2 of the x coordinate and the minimum value y1 and the maximum value y2 of the y coordinate are detected. As described above, the four vertices of the rectangular opening pattern 7 can be identified and the opening pattern 7 can be recognized. Further, the center coordinates of the rectangular opening pattern are xA = (x
It is given by 1 + x2) / 2, yA = (y1 + y2) / 2.

Next, as shown in FIG. 6, a luminance sectional waveform is obtained along each of the four sides a, b, c, d of the rectangular opening pattern 7. Further, its differential waveform is also calculated. Since the unevenness appears in the luminance cross-sectional waveform on the side a, a peak appears in the differential waveform, and the intersection points B and D can be recognized. Similarly, the intersection points a and c can be recognized along the side c. On the other hand, side b,
For d, since the luminance cross-section waveform does not change, the differential waveform becomes zero level and it is recognized that there is no intersection. In this way, the intersection point a of the geometric pattern 8 and the opening pattern 7,
B, C, and D are required.

Finally, as shown in FIG. 7, the recognized intersection points a,
Find two straight lines from B, C and D. That is, the coordinate systems of the four sides a, b, c and d are arranged in one line. Then, by searching for a peak in the direction of the side b from the intersection b at the maximum position of the side a, the intersection a can be found, and it can be seen that the pair of intersections a and b are connected by a straight line. Similarly, it can be seen that the other pair of intersections C and D are connected by a straight line. Next, the linear equations of the two straight lines Elo and Hani are obtained from the coordinates of their intersections, a simultaneous equation is set up, and the intersections of these two straight lines are mathematically solved to specify the position of the substrate B. It becomes possible to do. In this way, the image processing itself can be executed by a simple algorithm, and the positions of both substrates A and B can be recognized only by mathematical calculation.

For reference, the method of solving simultaneous equations is shown below.

[Equation 1]

FIG. 8 shows a processing algorithm when a plurality of radial geometric patterns (four in this example) are projected in the rectangular opening pattern 7. For example, side a
When the peak is searched from the intersection a at the maximum position in the direction of the adjacent side b, the corresponding intersection b can be found. Next, a and b whose coordinate positions are smaller than a and b
An intersection b larger than the coordinate position of can be searched as a pair. Similarly, the relationship between the straight line and the intersection can be grasped by sequentially searching for a, a, ...

FIG. 9 shows another specific example of the alignment mark. As shown in (A), the substrate A is provided with a circular opening pattern 7. On the other hand, as shown in (B),
The substrate B is provided with a concentric geometric pattern 8.

FIG. 10 shows an image recognition visual field 9 shown with the pair of alignment marks shown in FIG. 9 overlapping each other.
Is represented. As shown, a part of the concentric geometric pattern appears inside the circular opening pattern. The portion other than the opening pattern is covered with the light shielding region. FIG. 11 shows another example of the alignment mark appearing in the image recognition visual field 9. The geometric pattern is concentric, while the opening pattern has a rectangular shape.

FIG. 12 shows an algorithm applied to the image recognition of the alignment mark shown in FIG. First, the rectangular opening pattern 7 is recognized with the light shielding region 6 as the background. Next, intersection points a, b, c, and d of the window frame of the opening pattern 7 and the circular arc of the concentric geometric pattern 8 are detected. Then, the straight lines Elo and Hani are obtained, and their midpoints E and F are calculated. Furthermore, draw a vertical line from each midpoint e and f. An intersection P of a pair of perpendicular lines is obtained. This intersection point P is the center point of the concentric geometric pattern 8.

[0029]

As described above, according to the present invention, an alignment mark formed of a window frame-shaped opening pattern provided in a light-shielding region of one substrate and a center-directed shape provided on the other substrate. Image alignment is used to perform automatic alignment of both substrates using the alignment mark consisting of the geometric pattern. If only a part of the geometric pattern can be recognized through the opening pattern, the automatic alignment of the pair of substrates becomes possible. Therefore, there is an effect that the area of the opening pattern can be reduced as compared with the conventional case, and the deterioration of the operating characteristics of the laminated optical panel due to light leakage or the like can be prevented. Since the image recognition visual field can be determined only by the variation range of the opening pattern,
Only the dimensional tolerance and pre-alignment accuracy of one substrate need be considered. Therefore, there is an effect that the imaging range can be reduced as compared with the conventional case, and the enlargement ratio of the image recognition visual field can be increased correspondingly, and the position detection resolution is improved. Further, there is an effect that a complicated algorithm is not required for recognizing the opening pattern and the geometric pattern. It is needless to say that the substrate alignment method according to the present invention is not limited to the alignment of a pair of substrates forming the laminated optical panel, but can be applied to the positioning of one substrate.

[Brief description of drawings]

FIG. 1 is a schematic view showing an alignment device and an alignment mark used for carrying out a substrate alignment method according to the present invention.

FIG. 2 is a schematic view showing an example of an alignment mark projected in an image recognition visual field.

FIG. 3 is a schematic diagram showing another example of the alignment mark projected in the image recognition field of view.

FIG. 4 is a diagram showing an outline of an image recognition algorithm for an alignment mark.

FIG. 5 is a diagram showing a specific example of the image recognition algorithm.

FIG. 6 is a diagram showing a specific example of the image recognition algorithm.

FIG. 7 is a diagram similarly showing a specific example of the image recognition algorithm.

FIG. 8 is a diagram showing a specific example of the image recognition algorithm.

FIG. 9 is a schematic diagram showing another example of an alignment mark.

FIG. 10 is a schematic view showing an example of an alignment mark projected in the image recognition visual field.

FIG. 11 is a schematic view showing another example of the alignment mark projected in the image recognition field of view.

FIG. 12 is a diagram showing an outline of an image recognition algorithm.

FIG. 13 is a plan view showing an example of a laminated optical panel.

FIG. 14 is a partial plan view showing an example of a conventional alignment mark.

FIG. 15 is a partial plan view showing a state where the alignment marks shown in FIG. 14 are overlapped with each other.

FIG. 16 is an explanatory diagram showing a positional relationship of conventional alignment marks.

FIG. 17 is an explanatory diagram showing a positional relationship between conventional alignment marks.

FIG. 18 is an explanatory diagram showing a positional relationship of conventional alignment marks.

FIG. 19 is an explanatory diagram showing a positional relationship of conventional alignment marks.

[Explanation of symbols]

 1 Lower Stage 2 Upper Stage 3 Laminated Optical Panel 4 Camera Block 5 Camera 6 Light-shielding Area 7 Aperture Pattern 8 Geometric Pattern 9 Image Recognition Field of View A One Substrate B Another Substrate

Claims (5)

[Claims] 1. A first step of setting one substrate, which has an alignment mark formed of a window frame-shaped opening pattern in a light-shielding region, under a camera, and an alignment mark formed of a center-oriented geometric pattern. The second procedure of setting the other substrate on the one substrate overlaid on the other substrate, the third procedure of operating the camera to project both alignment marks overlapping each other in the image recognition visual field, and the image of the projected aperture pattern. The fourth procedure of recognizing and detecting the position of one substrate, and the other part by recognizing the image of the geometric pattern portion projected in the opening pattern and performing arithmetic processing to calculate the center point of the center-directed shape Substrate alignment of a laminated optical panel comprising a fifth step of detecting the position of the substrate and a sixth step of performing relative alignment of the both substrates based on the detected positions. Door way. 2. The substrate alignment method for a laminated optical panel according to claim 1, wherein the geometric pattern has a concentric or radial centering shape. 3. The substrate alignment method for a laminated optical panel according to claim 1, wherein the opening pattern has a circular or rectangular window frame shape. 4. A laminated optical panel having a pair of substrates bonded to each other in alignment with each other, wherein one of the substrates includes a light-shielding region, and an alignment mark formed of a window frame-shaped opening pattern is formed therein. On the other hand, the other substrate is provided with an alignment mark composed of a center-oriented geometric pattern. 5. The substrate according to claim 4, comprising a substrate on which pixel electrodes and switching elements are integrally formed, a substrate on which a color filter is formed, and a liquid crystal held in a gap between them. Laminated optical panel.