JP2006226676A - Apparatus and method for mark recognition - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus and a method for mark recognition capable of stably and accurately recognizing the positions of a plurality of alignment marks, even when the plurality of alignment marks are present, in an overlapped state in image data. <P>SOLUTION: A mark vertical sequential position determination part 32 automatically determines the vertical sequential positions of the plurality of alignment marks and recognizes the position of an alignment mark A of upper position. A mask processing image generating part 34 creates mask processing image data through the use of a mask region of the alignment mark of the upper position. A lower-position mark position recognizing part 35 recognizes the position of the alignment mark of a lower position through the use of the mask processing image data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a mark recognition apparatus and a mark recognition method for recognizing an alignment mark formed on a glass substrate, and in particular, a mark recognition apparatus and a mark recognition for recognizing the position of each alignment mark formed on an overlapped glass substrate. Regarding the method.

FIG. 14 is a flowchart for explaining a conventional mark recognition method.
As a mark recognition method for recognizing a predetermined mark from captured image data, a method for recognizing based on the edge of the mark, a method for recognizing from the center of gravity of the mark, and a method for recognizing by pattern matching have been conventionally used. Yes.

  As for a method for recognizing based on the edge of the mark, referring to FIG. 14, an image including the mark is captured (step B1), and the edge of the object to be recognized is extracted from the captured image data (step B2). 3 candidate points are extracted from the detected edges (step B3), and the center point of the predicted circle is extracted as the target point (step B4), thereby recognizing the position of the mark (for example, Patent Document 1).

  Further, the method of recognizing from the center of gravity position of the mark is to binarize the image data, detect the center of gravity position, recognize the mark position based on the center of gravity position, and further recognize by pattern matching. The method is to recognize a mark position by calculating a correlation value between a registered mark and a template image and searching for coordinates.

  However, when recognizing the position of the alignment mark formed on each glass substrate in a state where the two glass substrates are overlapped, the two alignment marks may overlap. When the two alignment marks overlap in this way, the alignment mark formed on the upper glass substrate has an effect when recognizing the alignment mark formed on the lower glass substrate with respect to the camera viewing direction. However, there is a problem that the recognition accuracy is lowered.

  That is, in the method of recognizing based on the edge of the mark, when the shape is limited to a circle and the edge point of another mark is selected in the selection of the edge point, an error occurs in the position of the recognized mark. Recognition accuracy is reduced. Also, in the method of recognizing from the center of gravity position of the mark, the mark shape becomes incomplete due to the influence of the shielding part, the center of gravity position does not match the center position, and an error occurs in the position of the recognized mark. The recognition accuracy will be reduced. Furthermore, in the method of recognizing by pattern matching, the correlation value with the registered template image is lowered due to the influence of the shielding part, and the detection accuracy is lowered, so that the recognition accuracy is lowered.

In addition, as a method of recognizing the position of the alignment mark formed on each glass substrate in a state where the two glass substrates are overlapped, after the image data of the alignment mark formed on one glass substrate is taken, 2 There has been proposed a method of recognizing the position of each alignment mark by capturing image data of alignment marks that are overlapped in a state where a plurality of glass substrates are stacked (see, for example, Patent Document 2). However, in this method, it is necessary to capture image data in two steps, and the work related to the alignment mark recognition process becomes complicated.
JP-A-9-178427 JP-A-7-306007

  The present invention has been made in view of such a problem, and an object of the present invention is to stably position each alignment mark even when a plurality of alignment marks are overlapped in image data. The present invention is to provide a mark recognition device and a mark recognition method that can be recognized easily and accurately.

In order to solve the above problems, the present invention has the following configurations.
The gist of the invention described in claim 1 is a mark recognition for recognizing the positions of a plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in a vertical order. An upper mark position recognizing unit for recognizing the position of the uppermost alignment mark, and using the recognition information of the uppermost alignment mark by the upper mark position recognizing unit as the upper mark recognition information, The mark recognition apparatus includes lower mark position recognition means for recognizing the position of the lower mark which is the lower alignment mark.
The lower mark position recognizing means recognizes the lower mark whose position has been recognized when the lower alignment mark exists than the lower mark whose position has been recognized. 2. The mark recognition apparatus according to claim 1, wherein information is used as recognition information of the upper mark to recognize a position of the alignment mark below.
The gist of the invention described in claim 3 is the mark according to claim 1 or 2, further comprising an up / down order determining means for determining the up / down order of the plurality of alignment marks using the two-dimensional image data. It exists in the recognition device.
According to a fourth aspect of the present invention, the vertical order determination means determines the vertical order of the alignment marks using each correlation value obtained from the result of individually recognizing the plurality of alignment marks. A mark recognition apparatus according to any one of claims 1 to 3, wherein
According to a fifth aspect of the invention, the lower mark position recognizing means excludes the upper mark portion on the two-dimensional image data from the correlation value calculation target of the recognition process as the upper mark recognition information. 5. A mark recognition apparatus according to claim 1, wherein a mask area is used.
The gist of the invention described in claim 6 is that the area corresponding to the size of the template image used for position recognition of the upper mark is used as the mask area. Exist.
The gist of the invention described in claim 7 resides in the mark recognition apparatus according to claim 5, wherein the upper mark area is used as the mask area.
The gist of the invention according to claim 8 resides in the mark recognition apparatus according to claim 7, wherein an area that matches the shape of the upper mark is used as the mask area.
The gist of the invention according to claim 9 is the mark recognition apparatus according to claim 7, wherein a region obtained by enlarging the region that matches the shape of the upper mark is used as the mask region. Exist.
The tenth aspect of the present invention is the mark according to the fifth aspect, wherein the mask region is a region in which a template image of a contour edge used for position recognition of the upper mark is filled. It exists in the recognition device.
The gist of the invention described in claim 11 is a mark for recognizing the position of each of the plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks that are stacked in an up-down order. A recognition method for recognizing the position of the uppermost alignment mark and using the recognized recognition information of the uppermost alignment mark as upper mark recognition information, The mark recognition method is characterized by recognizing the position of the mark.
According to a twelfth aspect of the present invention, when there is an alignment mark that is lower than the position-recognized lower mark, the position-recognized information of the position-recognized mark is used as the position-recognition information. The mark recognition method according to claim 11, wherein the position of the alignment mark below is recognized.
The gist of the invention described in claim 13 resides in the mark recognition method according to claim 11 or 12, wherein the two-dimensional image data is used to determine the order of the plurality of alignment marks.
The gist of the invention described in claim 14 is that the vertical order of the alignment marks is determined using the correlation values obtained from the result of individually recognizing the plurality of alignment marks. Item 14. The mark recognition method according to any one of Items 11 to 13.
The gist of the invention described in claim 15 is that a mask area for excluding the upper mark portion on the two-dimensional image data from the object of correlation value calculation of recognition processing is used as the recognition information of the upper mark. The mark recognition method according to any one of claims 11 to 14, wherein the mark recognition method is provided.
The gist of the invention described in claim 16 resides in the mark recognition method according to claim 15, wherein an area that matches the size of the template image used for position recognition of the upper mark is used as the mask area. .
The gist of the invention according to claim 17 resides in the mark recognition method according to claim 15, wherein the upper mark area is used as the mask area.
The gist of the invention according to claim 18 resides in the mark recognition method according to claim 17, characterized in that an area that matches the shape of the upper mark is used as the mask area.
The gist of the invention according to claim 19 resides in the mark recognition method according to claim 17, characterized in that an area obtained by enlarging the area that matches the shape of the upper mark is used as the mask area.
The gist of the invention according to claim 20 is the mark recognition method according to claim 15, wherein an area obtained by filling a template image of a contour edge used for position recognition of the upper mark is used as the mask area. Exist.
The gist of the invention described in claim 21 is a mark for recognizing the position of each of the plurality of alignment marks by using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in an up-down order. A program for causing a computer to execute a recognition method, which uses upper mark position recognition processing for recognizing the position of the uppermost alignment mark, and uses the recognized uppermost alignment mark recognition information as upper mark recognition information. Thus, the present invention resides in a program for causing a computer to execute a lower mark position recognition process for recognizing the position of a lower mark, which is the lower alignment mark.
According to a twenty-second aspect of the present invention, there is provided a mark for recognizing the position of each of the plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in a vertical order. A program for causing a computer to execute a recognition method, which uses upper mark position recognition processing for recognizing the position of the uppermost alignment mark, and uses the recognized uppermost alignment mark recognition information as upper mark recognition information. Then, the position of the lower mark, which is the lower alignment mark, is recognized, and if the lower alignment mark exists than the lower mark where the position is recognized, the lower mark is recognized when the position is recognized. Using the information as recognition information for the upper mark, the lower alignment Resides in a program, characterized in that to execute the recognizing lower mark position recognition process the position of the mark to the computer.
The subject matter of claim 23 is a mark for recognizing the position of each of the plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in an up-down order. A program for causing a computer to execute a recognition method, wherein a vertical order determination process for determining the vertical order of the plurality of alignment marks using the two-dimensional image data, and the highest order determined by the vertical order determination process The upper mark position recognition process for recognizing the position of the alignment mark and the recognized uppermost alignment mark recognition information are used as upper mark recognition information to recognize the position of the lower mark, which is the lower alignment mark. The lower mark position recognition processing to be executed is executed by a computer. It resides in the program.

  The mark recognition apparatus and mark recognition method of the present invention recognizes the upper mark position from the upper and lower order of a plurality of alignment marks, uses the result as a mask area, and eliminates the upper mark information for lower mark position recognition. By performing the configuration in a state, even when a plurality of alignment marks are overlapped in the image data, the position of each alignment mark can be recognized stably and accurately. .

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing the configuration of the first embodiment of the mark recognition apparatus according to the present invention, and FIG. 2 is a block diagram showing the configuration of the image processing apparatus shown in FIG.

  Referring to FIG. 1, the first embodiment includes a camera 1, an illumination 2 that irradiates a subject with light, and an image processing device 3 to which image data captured by the camera 1 is input. Range in which the alignment mark A and the alignment mark B are included from the upper glass substrate 4 by the camera 1 in a state where the upper glass substrate 4 on which A is formed and the lower glass substrate 5 on which the alignment mark B is formed are overlapped. And the image processing device 3 performs image processing on the video signal captured by the camera 1 to recognize the positions of the alignment marks A and B stably and accurately. FIG. 1 shows an example in which two layers of alignment marks A and B are imaged by the camera 1, but the present invention recognizes the positions of alignment marks of three or more layers. it can.

  The image processing apparatus 3 is an information processing apparatus such as a computer that operates under the control of a program. Referring to FIG. 2, a two-dimensional image conversion unit 31, an image memory 311, a mark up / down order determination unit 32, a mask area The generation unit 33, the mask processing image generation unit 34, the image memory 341, the lower mark position recognition unit 35, and the mark position data output unit 36 are included.

  The two-dimensional image conversion unit 31 converts a video signal input from the camera 1 into two-dimensional image data, and stores the converted two-dimensional image data (hereinafter referred to as input image data) in the image memory 311.

  The mark vertical order determination unit 32 determines the vertical order of a plurality of alignment marks existing in the input image data for the input image data stored in the image memory 311, and determines the position of the highest alignment mark in the vertical order. The recognition process is executed based on the program.

  The mask area generation unit 33 executes a process of generating mask area data from the position and shape data of the alignment mark positioned at the upper level based on a program.

  The mask processing image generation unit 34 uses the mask region data generated by the mask region generation unit 33 and recognizes a mark corresponding to a region corresponding to the region of the alignment mark positioned higher in the input image data stored in the image memory 311. A process of setting code values that are sometimes not targeted is executed based on a program, and the generated image (hereinafter referred to as mask processed image data) is stored in the image memory 341.

  The lower mark position recognizing unit 35 executes a process for recognizing the position of the alignment mark positioned lower in the mask processed image data stored in the image memory 341 based on the program.

  The mark position data output unit 36 executes a process of outputting position data of a plurality of alignment marks recognized by the mark up / down order determination unit 32 and the lower mark position recognition unit 35 based on a program.

Next, the alignment mark position recognition processing operation in the first embodiment will be described with reference to FIG.
FIG. 3 is a flowchart for explaining the operation of the first embodiment of the mark recognition apparatus according to the present invention.

  First, an image in a range including a plurality of alignment marks is captured by the camera 1 (step A1). The captured video signal is input to the two-dimensional image conversion unit 31, converted into input image data by the two-dimensional image conversion unit 31, and the converted input image data is stored in the image memory 311.

  Next, the mark up / down order determination unit 32 executes position recognition processing for each alignment mark on the input image data stored in the image memory 311 using the template images of the plurality of alignment marks to be recognized. Compare the resulting correlation values, determine the vertical order of multiple alignment marks (step A2), recognize the position of the recognized uppermost alignment mark, and mark the position data of the recognized uppermost alignment mark The data is output to the position data output unit 36.

  Next, the mask area generation unit 33 uses the position data of the alignment mark (hereinafter referred to as the upper mark) for which position recognition has been performed most recently, to the upper mark area, that is, the area including the upper mark. Mask area data for masking the portion is generated (step A3), and the generated mask area data is output to the mask processed image generation unit.

  Next, the mask processing image generation unit 34 sets the non-recognition information in the input image data stored in the image memory 311 using the mask region data input from the mask region generation unit 33 (step A4). That is, the mask processed image generation unit 34 describes information that excludes the region of the upper mark from the input image data from the correlation value calculation target using the mask region data input from the mask region generation unit 33. Mask processing image data is generated, and the generated mask processing image data is stored in the image memory 341.

  Next, the lower mark position recognizing unit 35 receives the mask processing image data stored in the image memory 341, and performs position recognition processing on an alignment mark (hereinafter referred to as a lower mark) one layer below the upper mark. By executing this, the position of the lower mark is recognized (step A5), and the position data of the recognized lower mark is output to the mark position data output unit.

  Next, it is determined whether or not the lower mark recognized in step A5 is the lowest alignment mark (step A6). If the lower mark recognized in step A5 is not the lowest alignment mark, step Returning to A3, the processes in steps A3 to A6 are repeated until the lower mark recognized in step A5 becomes the lowest alignment mark, and the positions of the alignment marks are sequentially recognized from the highest to the lowest.

  In step A6, when the lower mark recognized in step A5 is the lowest alignment mark, the mark position data output unit 36 recognizes the position by the mark up / down order determination unit 32 and the lower mark position recognition unit 35. The position data of all the alignment marks is output (step A7), and the alignment mark position recognition process is terminated.

Next, referring to FIG. 4 to FIG. 8 for the operation on the input image data in which the alignment marks in the first embodiment are overlapped, that is, the lower mark is blocked by the uppermost or upper mark. I will explain.
FIG. 4 is a diagram showing an example of input image data converted by the two-dimensional image conversion unit shown in FIG. 2, and FIG. 5 is a template used by the mark up / down order determination unit and the lower mark position recognition unit shown in FIG. 6 is a diagram illustrating an example of an image, FIG. 6 is a diagram illustrating an example of a mask region generated by the mask region generation unit illustrated in FIG. 2, and FIG. 7 is generated by the mask processing image generation unit illustrated in FIG. FIG. 8 is a diagram illustrating an example of mask processing image data. FIG. 8 is a diagram illustrating an example of a region in which a correlation value is calculated by the lower mark position recognition unit illustrated in FIG.

Information on the shape and brightness of the plurality of alignment marks to be recognized can be obtained from the template image used during the recognition process. On the other hand, the vertical order relationship of the plurality of alignment marks in the input image data may be unknown. In addition, the alignment marks may partially overlap and shielding may occur.

FIG. 4 shows an example of input image data in the case where the vertical order relation of the plurality of alignment marks is unknown, and the alignment marks partially overlap and blockage occurs. In the input image data shown in FIG. 4, two alignment marks A and alignment marks B are included, and the two alignment marks A and alignment marks B partially overlap with each other. Shielding has occurred.

  In the conventional mark recognition method, when the positions of the alignment mark A and the alignment mark B in the input image data shown in FIG. 4 are recognized, if the method of recognizing based on the edge of the mark is adopted, the edge point of the alignment mark A is determined. When selecting, the edge point of the alignment mark B which overlaps may be selected, and the position of the recognized alignment mark will shift | deviate. Also, when adopting a method of recognizing from the center of gravity position of the mark using binarization processing, the shielding portion of the alignment mark B has an effect, and the center of gravity position of the area resulting from binarization and the position of the alignment mark A are shifted. In addition, even when a method of recognizing by pattern matching based on the brightness of the template image is adopted, the overlapping portion of the alignment mark A is affected at the time of recognition of the alignment mark B, and the obtained correlation value becomes low, and it is shifted from the correct position. May be recognized.

  On the other hand, in the first embodiment of the mark recognition apparatus according to the present invention, first, the mark vertical order determination unit 32 recognizes the positions of the alignment marks A and B for the input image data. 5A is performed using the template image of the alignment mark A shown in FIG. 5A and the template image of the alignment mark B shown in FIG. 5B, and the correlation values of the respective processing results are compared. The order is determined, the position of the recognized uppermost alignment mark is recognized, and the position data of the recognized uppermost alignment mark is output to the mark position data output unit 36. In the case of the input image data shown in FIG. 4, since the alignment mark B has a shielding portion due to the overlap with the alignment mark A, the correlation value is low, so the upper and lower order is the alignment mark A and the alignment mark B from the top. The position recognition of the alignment mark A is accurately performed by the mark vertical order determination unit 32.

  Next, the mask area generation unit 33 generates area data (mask area) having the same size as the template image of the alignment mark A at the position of the recognized upper alignment mark A shown in FIG. Next, the mask processed image generation unit 34 sets a code that is not targeted at the time of recognition processing in the mask area portion of the input image data. Pixels in the mask area are included in the path having the upper left position and the lower right position as the end coordinates of the two-dimensional image data to be recognized by the template image, but are excluded when calculating the correlation value. The As shown in FIG. 7, the generated mask-processed image is only the luminance information of the alignment mark B excluding the luminance information of the alignment mark A.

  Next, a lower alignment mark B is recognized by the lower mark position recognition unit 35. In the lower mark position recognizing unit 35, the correlation value is calculated using the template image of the alignment mark B shown in FIG. 5B. However, as shown in FIG. When the mask area of the mark A is included, it is performed in a portion excluding the mask area. As a result, a high correlation value can be obtained with the mask area removed from the template image of FIG. 5B, and the position of the alignment mark B is accurately recognized by the lower mark position recognition unit 35.

  The mark up / down order determination unit 32 and the lower mark position recognition unit 35 are configured to recognize the position of the alignment mark using matching based on the brightness of the template image. However, the edge image is extracted from the template image. You may change to the matching method by brightness | luminance.

FIG. 9 is a diagram showing an example of a template image of a contour edge used in the mark up / down order determination unit and the lower mark position recognition unit shown in FIG.
When the matching method based on the brightness of the edge image is adopted as the position recognition method in the mark up / down order determination unit 32 and the lower mark position recognition unit 35, the template image uses the image of the contour portion shown in FIG. As the mask area generated in the mask area generation unit 33, as shown in FIG. 9B, only the upper mark part becomes the mask area by using the area where the contour image is subjected to the hole filling process, and the recognition accuracy is improved. Can be raised.

FIG. 10 is a diagram illustrating another example of the mask region generated in the mask region generation unit illustrated in FIG. 2, and FIG. 11 illustrates mask processing image data generated by the mask processing image generation unit illustrated in FIG. 2. FIG. 12 is a diagram illustrating another example, and FIG. 12 is a diagram illustrating an example in which enlargement processing is performed on the mask region illustrated in FIG. 10.
Further, as the mask area generated in the mask area generation unit 33, the template image size area of the upper mark shown in FIG. 6 is used. Instead, the same area as the shape of the upper mark as shown in FIG. 10 is used. Thus, only the upper mark portion becomes a mask region, and the recognition accuracy can be improved. When the mask region shown in FIG. 10 is used, the mask processed image data generated by the mask processed image generation unit 34 is as shown in FIG.

  Further, by performing an enlargement process on the mask area shown in FIG. 10 and using the enlarged area as shown in FIG. It becomes the area | region which removes, and can raise recognition accuracy.

  As described above, according to the first embodiment, the mark vertical order determination unit 32 automatically determines the vertical order of the plurality of alignment marks A and B, and recognizes the position of the upper alignment mark A. Then, the mask processing image generation unit 34 generates mask processing image data using the mask region of the upper alignment mark A, and the lower mark position recognition unit 35 recognizes the position of the lower alignment mark B using the mask processing image data. By configuring as described above, there is an effect that the position of each alignment mark can be recognized stably and accurately even when a plurality of alignment marks are overlapped in the image data.

Next, a second embodiment of the present invention will be described with reference to FIG.
FIG. 13 is a block diagram showing the configuration of the second embodiment of the mark recognition apparatus according to the present invention.

  In the second embodiment, referring to FIG. 13, an upper mark position recognizing unit 42 is provided in place of the mark up / down order determining unit 32 having the configuration of the first embodiment.

  The upper mark position recognizing unit 42 determines the order of alignment marks to be recognized based on the upper and lower order data and recognizes the position of the uppermost alignment mark when the upper and lower order of the plurality of alignment marks is known. It is. The mark up / down order determination unit 32 which is the configuration of the first embodiment recognizes the positions of all the alignment marks in order to determine the up / down order. Therefore, the lower mark needs to be recognized twice for each alignment mark before the position is determined. On the other hand, since the upper and lower order is known by the upper mark position recognizing unit 42, the positions of all the alignment marks can be recognized by one recognition.

  As described above, according to the second embodiment, the upper mark position recognition unit 42 recognizes the position of the uppermost alignment mark based on the known vertical order of a plurality of alignment marks, and performs mask processing. The image generation unit 34 generates mask processed image data using the mask region of the upper alignment mark, and the lower mark position recognition unit 35 recognizes the position of the lower alignment mark using the mask processed image data. As a result, the mark position recognition process can be performed at high speed, and stable and accurate recognition can be achieved.

  Further, according to the first and second embodiments, when a plurality of alignment marks, the lower alignment mark overlaps the upper alignment mark and is partially shielded, the position is shifted due to the influence of the upper alignment mark. The problem with the prior art is that the upper alignment mark is recognized based on the upper and lower order of the alignment mark positions, and the lower alignment mark position is recognized without the upper alignment mark information. Can be solved.

  Furthermore, according to the first embodiment, there is no need to know the upper and lower order of the positions of the plurality of alignment marks, and there is an advantage that the determination can be automatically made from the input image data. In addition, according to the second embodiment, since the vertical order of the alignment marks is known, the number of recognition processes until the positions of all the alignment marks are determined is smaller than that in the first embodiment. In order to achieve the purpose, there is a merit that can be processed at high speed.

  Note that the present invention is not limited to the above-described embodiments, and it is obvious that the embodiments can be appropriately changed within the scope of the technical idea of the present invention. In addition, the number, position, shape, and the like of the constituent members are not limited to the above-described embodiment, and can be set to a suitable number, position, shape, and the like in practicing the present invention. In each figure, the same numerals are given to the same component.

It is a block diagram which shows the structure of 1st Embodiment of the mark recognition apparatus which concerns on this invention. It is a block diagram which shows the structure of the image processing apparatus shown in FIG. It is a flowchart for demonstrating operation | movement of 1st Embodiment of the mark recognition apparatus which concerns on this invention. It is a figure which shows the example of input image data converted by the two-dimensional image conversion part shown in FIG. It is a figure which shows the example of a template image used in the mark up-down order determination part and lower mark position recognition part shown in FIG. It is a figure which shows the example of a mask area | region produced | generated by the mask area | region production | generation part shown in FIG. It is a figure which shows the mask processing image data example produced | generated by the mask processing image generation part shown in FIG. It is a figure which shows the example of an area | region which calculates a correlation value in the low-order mark position recognition part shown in FIG. It is a figure which shows the template image example of the contour edge used in the mark up-down order determination part and lower mark position recognition part shown in FIG. It is a figure which shows the other example of the mask area | region produced | generated in the mask area | region production | generation part shown in FIG. It is a figure which shows the other example of the mask process image data produced | generated by the mask process image production | generation part shown in FIG. It is a figure which shows the example which performed the expansion process to the mask area | region shown in FIG. It is a block diagram which shows the structure of 2nd Embodiment of the mark recognition apparatus which concerns on this invention. It is a flowchart for demonstrating the conventional mark recognition method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Camera 2 Illumination 3 Image processing apparatus 4 Upper glass substrate 5 Lower glass substrate 31 Two-dimensional image conversion part 32 Mark up-down order determination part 33 Mask area generation part 34 Mask process image generation part 35 Lower mark position recognition part 36 Mark position data Output unit 42 Upper mark position recognition unit 311 Image memory 341 Image memory A Alignment mark B Alignment mark

Claims (23)

A mark recognition device for recognizing the position of each of the plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in an up-down order,
Upper mark position recognition means for recognizing the position of the uppermost alignment mark;
Lower mark position recognition means for recognizing the position of the lower mark, which is the lower alignment mark, using the recognition information of the uppermost alignment mark by the upper mark position recognition means as the upper mark recognition information. The mark recognition apparatus characterized by the above-mentioned.   The lower mark position recognizing means uses the recognition information of the lower mark whose position is recognized as the recognition information of the upper mark when the alignment mark lower than the lower mark whose position is recognized exists. The mark recognition apparatus according to claim 1, wherein the position of the alignment mark below is recognized.   The mark recognition apparatus according to claim 1, further comprising a vertical order determination unit that determines the vertical order of the plurality of alignment marks using the two-dimensional image data.   4. The up / down order determining means determines the up / down order of the alignment marks using respective correlation values obtained from the result of individually recognizing the plurality of alignment marks. The mark recognition apparatus in any one of.   The lower mark position recognizing means uses, as the upper mark recognition information, a mask region for excluding the upper mark portion on the two-dimensional image data from a correlation value calculation target of recognition processing. The mark recognition apparatus according to claim 1.   6. The mark recognition apparatus according to claim 5, wherein an area that matches a size of a template image used for position recognition of the upper mark is used as the mask area.   6. The mark recognition apparatus according to claim 5, wherein the upper mark area is used as the mask area.   8. The mark recognition apparatus according to claim 7, wherein an area that matches the shape of the upper mark is used as the mask area.   8. The mark recognition apparatus according to claim 7, wherein an area obtained by enlarging an area that matches the shape of the upper mark is used as the mask area.   6. The mark recognition apparatus according to claim 5, wherein the mask region is a region in which a template image of a contour edge used for position recognition of the upper mark is subjected to hole filling processing. A mark recognition method for recognizing the positions of a plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in a vertical order,
Recognize the position of the top alignment mark,
A mark recognition method comprising recognizing the position of a lower mark, which is the lower alignment mark, by using the recognized uppermost alignment mark recognition information as upper mark recognition information.   If there is a lower alignment mark than the lower mark whose position is recognized, the recognition information of the lower mark whose position is recognized is used as the upper mark recognition information. The mark recognition method according to claim 11, wherein the position is recognized.   The mark recognition method according to claim 11 or 12, wherein the two-dimensional image data is used to determine the order of the plurality of alignment marks.   The mark according to any one of claims 11 to 13, wherein the vertical order of the alignment marks is determined using each correlation value obtained from the result of individually recognizing the plurality of alignment marks. Recognition method.   15. The mask area for excluding the upper mark portion on the two-dimensional image data from the correlation value calculation target of recognition processing is used as the upper mark recognition information. The mark recognition method described in 1.   The mark recognition method according to claim 15, wherein an area that matches a size of a template image used for position recognition of the upper mark is used as the mask area.   The mark recognition method according to claim 15, wherein the upper mark area is used as the mask area.   The mark recognition method according to claim 17, wherein an area that matches a shape of the upper mark is used as the mask area.   18. The mark recognition method according to claim 17, wherein an area obtained by enlarging an area that matches the shape of the upper mark is used as the mask area.   The mark recognition method according to claim 15, wherein an area obtained by performing a hole filling process on a template image of a contour edge used for position recognition of the upper mark is used as the mask area. A program for causing a computer to execute a mark recognition method for recognizing the positions of a plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in a vertical order. ,
An upper mark position recognition process for recognizing the position of the uppermost alignment mark;
Using the recognized recognition information of the uppermost alignment mark as the upper mark recognition information, and causing the computer to execute a lower mark position recognition process for recognizing the position of the lower mark, which is the lower alignment mark. A featured program. A program for causing a computer to execute a mark recognition method for recognizing the positions of a plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in a vertical order. ,
An upper mark position recognition process for recognizing the position of the uppermost alignment mark;
The recognition information of the recognized uppermost alignment mark is used as the upper mark recognition information, and the position of the lower mark, which is the lower alignment mark, is recognized, and further lower than the lower mark whose position is recognized. When the alignment mark is present, the lower mark position recognition process for recognizing the position of the alignment mark below it using the recognition information of the lower mark whose position is recognized as the recognition information of the upper mark. A program characterized by being executed by a computer. A program for causing a computer to execute a mark recognition method for recognizing the positions of a plurality of alignment marks using two-dimensional image data obtained by imaging a plurality of alignment marks stacked in a vertical order. ,
A vertical order determination process for determining the vertical order of the plurality of alignment marks using the two-dimensional image data;
An upper mark position recognition process for recognizing the position of the uppermost alignment mark determined by the vertical order determination process;
Using the recognized recognition information of the uppermost alignment mark as the upper mark recognition information, and causing the computer to execute a lower mark position recognition process for recognizing the position of the lower mark, which is the lower alignment mark. A featured program.

Images