Alignment device and alignment method

Abstract

To provide a technique capable of adjusting one or more of a lens magnification of an imaging device possessed by an exposure device and a position of a stage possessed by the exposure device, even when only one mark is photographed on the end of a photographic image.SOLUTION: An alignment method comprises: acquiring a photographic image including a mark provided on an object held on a stage possessed by an exposure device, which is imaged by an imaging device possessed by the exposure device; and detecting the mark from the photographic image. The method corrects the position and the size of the detected mark, to a position and a size of the mark obtained by template matching using the region of the mark in the photographic image. The method adjusts one or more of the position of the stage and the lens magnification of the imaging device, on the basis of the corrected position and size.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an alignment technique. [Background technology]

[0002] In the photolithography process, the exposure equipment must precisely align the mask and the substrate (such as a semiconductor wafer or glass substrate) before exposure. With the miniaturization of semiconductor elements in recent years, there is a demand for higher precision in alignment.

[0003] The mask and substrate have alignment marks of different patterns (hereinafter referred to as mask marks and substrate marks, respectively), and it is common to capture images of both marks simultaneously and then perform alignment based on those images.

[0004] In many cases, alignment marks (hereinafter simply referred to as marks) are prepared in advance with a predetermined pattern. From the viewpoints of ease of image processing and good visibility, rectangular patterns are often used as the mark patterns.

[0005] Based on the captured image, marks in the captured image are detected by image processing to obtain position information of the marks, etc., and based on that, the mask position and substrate position are moved to perform alignment. Since accurate detection of marks in an image leads to improved alignment accuracy, various methods are being considered.

[0006] In the technology disclosed in Patent Document 1, a mark is detected using template matching, a regression line is calculated based on the position of the mark, and template matching is further performed in a limited area near the edge of the image, thereby enabling accurate detection of marks not only in the central area of ​​the image but also in the edge area of ​​the image. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] JP 2010-198089 A Summary of the Invention [Problem to be solved by the invention]

[0008] Forming a plurality of marks regularly as in Patent Document 1 is more costly than forming only one mark, and therefore, from the viewpoint of cost, it is more desirable to use an alignment method that can be implemented even with only one mark.

[0009] In the technology disclosed in Patent Document 1, the mark positions at the image edge are estimated according to the arrangement rules of multiple marks in the captured image. Therefore, alignment cannot be performed if only one mask mark or one board mark, or both, are captured at the image edge of the captured image.

[0010] The present invention provides a technology that makes it possible to adjust one or more of the lens magnification of an imaging device of an exposure apparatus and the position of a stage of the exposure apparatus, even in cases where only one mark is captured at the edge of a captured image. [Means for solving the problem]

[0011] One aspect of the present invention is characterized in that it comprises a detection means for acquiring an image captured by an imaging device possessed by an exposure apparatus, the image including a mark provided on an object held on a stage possessed by the exposure apparatus, and detecting the mark from the captured image, a correction means for correcting a position and size of the mark detected by the detection means to a position and size of the mark obtained by template matching using an area of ​​the mark in the captured image, and an adjustment means for adjusting one or more of a position of the stage and a lens magnification of the imaging device based on the position and size corrected by the correction means. Effect of the Invention

[0012] According to the present invention, even in a case where only one mark is captured at the edge of an image, it is possible to provide a technology that makes it possible to adjust one or more of the lens magnification of an imaging device of an exposure apparatus and the position of a stage of the exposure apparatus. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing an example of the arrangement of an exposure apparatus. [Diagram 2] 10 is a diagram showing an example of a captured image obtained by the image capturing unit 105 when the position of the mask stage 102, the position of the substrate stage 104, and the lens magnification of the image capturing unit 105 are properly adjusted. FIG. [Diagram 3] 13 is a diagram showing an example of a captured image obtained by the image capturing unit 105 in a state in which adjustment of one or more of the position of the mask stage 102, the position of the substrate stage 104, and the lens magnification of the image capturing unit 105 is required. FIG. [Figure 4] FIG. 2 is a block diagram showing an example of the functional configuration of a main control unit 108. [Diagram 5] 4 is a diagram showing an example of the bounding boxes of the mask mark 302 and the board mark 303 detected by the detection unit 401 from the captured image 301 of FIG. 3. [Figure 6] FIG. 1 shows an example of a template of three substrate marks of different sizes. [Figure 7] 4 is a diagram for explaining an example of the operation of a correction unit 402. [Figure 8] 13A is a flowchart showing the operation of the main control unit 108, and FIG. 13B is a flowchart showing details of the process in step S892. [Figure 9] FIG. 2 is a block diagram showing an example of the configuration of a main control unit 108. [Figure 10] FIG. 13 is a diagram for explaining an example of the operation of a selection unit 902. [Figure 11] 13 is a flowchart showing an example of a process that can be applied to step S892. [Figure 12] 4 is a flowchart of the operation of the main control unit 108. [Figure 13] 4 is a flowchart of the operation of the main control unit 108. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, the embodiments will be described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe a number of features, not all of these features are essential to the invention, and the features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0015] [First embodiment] In this embodiment, an exposure apparatus equipped with an alignment device that performs object detection on alignment marks and corrects the obtained information on the mark position and mark size by template matching will be described. With such an alignment device, alignment can be performed even if only one mask mark or one substrate mark, or both, are captured at the edge of a captured image.

[0016] First, an example of the configuration of an exposure apparatus that performs alignment including position adjustment of a mask and a substrate by an alignment apparatus according to this embodiment will be described with reference to the block diagram of FIG.

[0017] A mask 101 is placed on the mask stage 102, and a substrate 103 to be exposed is placed on the substrate stage 104. The mask stage 102 is configured to hold the mask 101 and move, and the substrate stage 104 is configured to hold the substrate 103 and move.

[0018] A pattern to be transferred to a substrate 103 is drawn on the mask 101. Light from an illumination optical system 106 is irradiated onto the substrate 103 via the mask 101 and a projection optical system 107, whereby the pattern drawn on the mask 101 is projected onto the substrate 103. In addition, the mask 101 is provided with mask marks that are marks for alignment, and the substrate 103 is provided with substrate marks that are marks for alignment.

[0019] The main controller 108 controls the operation of the entire exposure apparatus, and also executes or controls various processes described as processes performed by the exposure apparatus. The main controller 108 has a processor such as a CPU or MPU, and a memory for storing computer programs and data. The processor executes processes using the computer programs and data stored in the memory, allowing the main controller 108 to execute or control various processes.

[0020] The main control unit 108 also functions as an alignment device, controls the image capturing unit 105 to acquire an image, and performs one or more of adjustments (alignment) of the position of the mask stage 102, the position of the substrate stage 104, and the lens magnification of the image capturing unit 105 based on the captured image. After the alignment, the main control unit 108 controls the illumination optical system 106 to perform exposure.

[0021] 2 shows an example of a captured image obtained by the image capturing unit 105 when the position of the mask stage 102, the position of the substrate stage 104, and the lens magnification of the image capturing unit 105 are adjusted correctly. In a captured image 201 shown in FIG. 2, a mask mark 202 and a substrate mark 203 are captured at a desired position (prescribed position) with a desired size (prescribed size). The exposure apparatus according to this embodiment adjusts one or more of the position of the mask stage 102, the position of the substrate stage 104, and the lens magnification of the image capturing unit 105 so that the mask mark and the substrate mark in the captured image are captured at the desired position with a desired size.

[0022] It is assumed that the desired size and the desired position are set in advance for each lens magnification of the image capturing unit 105. The "desired position" of the mask mark 202 and the board mark 203 is set to the same position for all lens magnifications, for example, the center position of the captured image (center pixel position). On the other hand, the "desired size" of the mask mark 202 and the board mark 203 is set to a larger size for a larger lens magnification and a smaller size for a smaller lens magnification.

[0023] Next, an example of a captured image obtained by the image capturing unit 105 in a state where adjustment of one or more of the position of the mask stage 102, the position of the substrate stage 104, and the lens magnification of the image capturing unit 105 is required is shown in Fig. 3. In the captured image 301 shown in Fig. 3, the mask mark 302 and the substrate mark 303 are shown at a size smaller than the desired size, and the mask mark 302 and the substrate mark 303 are shown at a position deviated from the center position of the captured image 301. When such a captured image 301 is obtained, the exposure apparatus adjusts the positions of the mask stage 102 and the substrate stage 104, respectively, so that the mask mark 302 and the substrate mark 303 are shown at the center position of the captured image 301 (the positions of the mask mark 202 and the substrate mark 203, respectively), and adjusts the lens magnification of the image capturing unit 105 so that the mask mark 302 and the substrate mark 303 are shown in the captured image 301 at the desired size (the sizes of the mask mark 202 and the substrate mark 203, respectively).

[0024] Next, an example of the functional configuration of the main control unit 108 functioning as an alignment device will be described with reference to the block diagram of FIG. 4. The detection unit 401 acquires the captured image output from the image capturing unit 105, detects a mark from the captured image, and obtains the position (mark position) and size (mark size) of the mark. More specifically, when the detection unit 401 detects a mark from the captured image, the detection unit 401 obtains the position and size of a bounding box, which is a rectangular area that contains the mark, as the mark position and the mark size, respectively. The position of the bounding box is any position in the bounding box, and may be, for example, the center position of the bounding box or any position of the four corners of the bounding box. The size of the bounding box includes the horizontal size and vertical size of the bounding box. Note that, when a part of the mark protrudes from the captured image and is included in the captured image, the detection unit 401 obtains the position and size of the bounding box that contains the "part of the mark" included in the captured image. The position and size of the bounding box are merely one example of information for defining the area of ​​the mark in the captured image, and any information capable of defining the area of ​​the mark in the captured image may be used. Then, the detection unit 401 determines (recognizes) whether the detected mark is a mask mark or a board mark. Various methods can be applied to the method for determining the position and size of the bounding box of the mask mark and the position and size of the bounding box of the board mark from the captured image. In this embodiment, the detection unit 401 determines the position and size of the bounding box of the mask mark and the position and size of the bounding box of the board mark from the captured image using a trained neural network.

[0025] An example of the bounding boxes of the mask mark 302 and the board mark 303 detected by the detection unit 401 from the captured image 301 of FIG. 3 will be described with reference to FIG.

[0026] 5, a bounding box 501 is a bounding box that includes the mask mark 302. In addition, in the captured image 301 of FIG. 5, a part of the board mark 303 protrudes from the captured image 301, and a part of the board mark 303 is included in the lower right corner of the captured image 301. In this case, the detection unit 401 determines the position and size of a bounding box 502 that includes the "part of the board mark 303" included in the captured image 301.

[0027] Here, the bounding box of the mark detected from the captured image by the detection unit 401 is an estimated area of ​​the mark estimated from the captured image by the detection unit 401. However, the position and size of the bounding box often do not match the position and size of the mark in the captured image. For alignment, it is necessary to more accurately determine the position and size of the mark in the captured image.

[0028] Therefore, the correction unit 402 sets an area including the bounding box of the mark and the surrounding area of ​​the bounding box as a search area. Then, the correction unit 402 arranges a template of the mark at each position in the search area, and obtains a similarity between the arranged template and an image area overlapping the template in the search area. A general method such as structural similarity (SSIM, Structural Similarity Index Measure) or normalized cross correlation (NCC, Normalized Cross Correlation) is used as a method for obtaining the similarity. However, other methods may be adopted as long as they can evaluate the similarity between the template and the image area. Then, the correction unit 402 corrects (updates) the mark position obtained by the detection unit 401 to the "arrangement position of the template at which the maximum similarity is obtained" and corrects (updates) the mark size obtained by the detection unit 401 to the "size of the template". In other words, the correction unit 402 identifies a more accurate area of ​​the mark from the candidate areas of the mark.

[0029] The correction unit 402 performs such processing for each of the mask mark and the board mark, thereby obtaining the mark position and mark size of the mask mark and the mark position and mark size of the board mark. That is, the correction unit 402 specifies, from the candidate region of the mask mark, a region having the mark position and mark size of the mask mark as the region of the mask mark. The correction unit 402 also specifies, from the candidate region of the board mark, a region having the mark position and mark size of the board mark as the region of the board mark.

[0030] The operation of the correction unit 402 will be described with reference to FIG. 7 as an example. The correction unit 402 sets a region including the bounding box 501 of the mask mark 302 in the captured image 301 and the surrounding region of the bounding box 501 as a search region 701. Here, as shown in FIG. 5, when the bounding box 502 including "a part of the board mark 303" is set, it is not possible to obtain the accurate mark position and mark size of the board mark 303 from the search region including the bounding box 502 and its surrounding region. Therefore, as shown in FIG. 7, since the bounding box 502 is located in the lower right corner of the captured image 301, the correction unit 402 determines a position that is a specified distance away in the lower right direction from the position of the lower right corner of the bounding box 502 as a virtual position. Then, correction unit 402 sets the position of the upper left corner of bounding box 502 as the position of its upper left corner, sets the position of the virtual position as the position of its lower right corner as an extended bounding box, and sets an area including the extended bounding box and its surrounding area as a search area 702. Note that there are various possible methods for setting search area 702 for board mark 303 based on bounding box 502 that includes "part of board mark 303", and the method is not limited to a specific method.

[0031] Then, the correction unit 402 performs template matching with the template of the mask mark 302 at each position in the search area 701, and determines the position and size of the template at which the similarity is maximized as the mark position and mark size of the mask mark 302. Similarly, the correction unit 402 performs template matching with the template of the board mark 303 at each position in the search area 702, and determines the position and size of the template at which the similarity is maximized as the mark position and mark size of the board mark 303. Note that if the template protrudes from the captured image when calculating the similarity between the template and the image area, the protruding portion is ignored, and the similarity is calculated using only the portion of the template that overlaps with the captured image.

[0032] The adjustment execution unit 403 uses the mark position and mark size of the mask mark and the mark position and mark size of the substrate mark to determine one or more of the adjustment amount for the position of the mask stage 102, the adjustment amount for the position of the substrate stage 104, and the adjustment amount for the lens magnification of the image capturing unit 105.

[0033] 3 as an example, the mask mark 302 (an area having the mark position and mark size of the mask mark 302) is shifted to the upper left from the center position of the captured image 301. Therefore, the adjustment execution unit 403 obtains a set of a horizontal shift amount Δx from the center position of the captured image 301 to the center position of the mask mark 302 and a vertical shift amount Δy from the center position of the captured image 301 to the center position of the mask mark 302 as adjustment amounts of the position of the mask mark 302. Then, based on the adjustment amounts of the position of the mask mark 302, the adjustment execution unit 403 adjusts the position of the mask stage 102 so that the center position of the mask mark 302 is located at the center position of the captured image 301. It is assumed that the amount by which the mask stage 102 is moved according to the adjustment amount of the position of the mask mark 302 is set in advance.

[0034] Furthermore, the board mark 303 (an area having the mark position and mark size of the board mark 303) is shifted to the lower right from the center position of the captured image 301. Therefore, the adjustment execution unit 403 obtains a set of a horizontal shift amount Δx from the center position of the captured image 301 to the center position of the board mark 303 and a vertical shift amount Δy from the center position of the captured image 301 to the center position of the board mark 303 as adjustment amounts for the position of the board mark 303. Then, based on the adjustment amounts for the position of the board mark 303, the adjustment execution unit 403 adjusts the position of the board stage 104 so that the center position of the board mark 303 is located at the center position of the captured image 301. It is assumed that the amount by which the board stage 104 is moved according to the adjustment amount for the position of the board mark 303 is set in advance.

[0035] Also, the size of the mask mark 302 and the size of the board mark 303 are smaller than the size of the mask mark 202 and the size of the board mark 203 in the captured image 201, respectively. Therefore, the correction unit 402 determines an adjustment amount of the lens magnification such that the size of the mask mark 302 becomes the size of the mask mark 202 (the size of the board mark 303 becomes the size of the board mark 203) based on the size ratio of the size of the mask mark 302 to the size of the mask mark 202 (the size ratio of the size of the board mark 303 to the size of the board mark 203). Then, the adjustment execution unit 403 adjusts the lens magnification of the image capturing unit 105 according to the determined adjustment amount of the lens magnification so that the size of the mask mark 302 becomes the size of the mask mark 202 (so that the size of the board mark 303 becomes the size of the board mark 203).

[0036] Here, if the above-mentioned template matching is performed using a template of one size, there is a problem that if the lens magnification of the image capturing unit 105 is set incorrectly, template matching cannot be performed appropriately. In this embodiment, therefore, in order to solve such a problem, templates of different sizes are prepared in advance. Such multiple templates are stored, for example, in a memory possessed by the main control unit 108.

[0037] An example of three board mark templates of different sizes is shown in Fig. 6. Template 602 is a template having the same size as "board mark 202 in captured image 201" and is a template corresponding to lens magnification R for capturing an image of the board mark at the size of board mark 202.

[0038] The template 601 is a template having a size larger than the "board mark 202 in the captured image 201" and corresponds to a lens magnification higher than the lens magnification R.

[0039] The template 603 is a template having a size smaller than the "board mark 202 in the captured image 201" and corresponds to a lens magnification lower than the lens magnification R.

[0040] Similarly, for the mask mark templates, a template having the same size as the "mask mark 203 in the captured image 201" (template corresponding to the lens magnification R), a template having a size larger than the "board mark 202 in the captured image 201" (template corresponding to a lens magnification higher than the lens magnification R), and a template having a size smaller than the "board mark 202 in the captured image 201" (template corresponding to a lens magnification lower than the lens magnification R) are provided.

[0041] Next, the operation of the main control unit 108 will be described with reference to the flowchart in Fig. 8(a). In step S891, the detection unit 401 acquires the captured image output from the image capturing unit 105, detects a mark from the captured image, and determines the position and size of the bounding box of the mark as the mark position and mark size, respectively.

[0042] In step S892, the correction unit 402 sets a region including the bounding box of the mark (mask mark and substrate mark) and the peripheral region of the bounding box as a search region. Then, the correction unit 402 performs template matching on the search region of the mark to correct the mark position and mark size of the mark. Details of the process in step S892 will be described later with reference to FIG. 8(b).

[0043] In step S893, the adjustment execution unit 403 uses the mark positions and mark sizes of the marks (mask marks and substrate marks) to determine one or more of the adjustment amount for the position of the mask stage 102, the adjustment amount for the position of the substrate stage 104, and the adjustment amount for the lens magnification of the image capturing unit 105.

[0044] Then, the adjustment execution unit 403 adjusts the adjustment target according to the calculated adjustment amount. For example, when the adjustment execution unit 403 calculates an adjustment amount for the position of the mask stage 102, the adjustment execution unit 403 adjusts the position of the mask stage 102 based on the adjustment amount. When the adjustment execution unit 403 calculates an adjustment amount for the position of the substrate stage 104, the adjustment execution unit 403 adjusts the position of the substrate stage 104 based on the adjustment amount. When the correction unit 402 calculates an adjustment amount for the lens magnification of the image capturing unit 105, the correction unit 402 adjusts the lens magnification based on the adjustment amount.

[0045] Next, an example of processing applicable to step S892 above will be described with reference to the flowchart in Fig. 8(b). Note that the processing according to the flowchart in Fig. 8(b) is a flowchart for processing one mark. However, by performing the processing according to the flowchart in Fig. 8(b) for each of the mask mark and the substrate mark, the mark positions and mark sizes of each of the mask mark and the substrate mark can be obtained.

[0046] In step S801, when the detection unit 401 determines that the mark is a mask mark, the correction unit 402 acquires a template group prepared for each lens magnification for the mask mark. On the other hand, when the detection unit 401 determines that the mark is a board mark, the correction unit 402 acquires a template group prepared for each lens magnification for the board mark.

[0047] In step S802, correction unit 402 selects one unselected template from the group of templates acquired in step S801 as a selected template.

[0048] In step S803, the correction unit 402 sets an area including the bounding box (BBox) of the mark and the surrounding area of ​​the bounding box as a search area.

[0049] In step S804, the correction unit 402 determines whether the mark is located at an edge of the captured image and whether the mark protrudes from the captured image based on the position and size of the bounding box of the mark. For example, when the bounding box of the mark is located at an edge of the captured image, the correction unit 402 may determine that the mark is located at an edge of the captured image. Also, for example, when the bounding box of the mark protrudes from the captured image, the correction unit 402 may determine that the mark protrudes from the captured image.

[0050] If the result of this determination is that the mark is located at the edge of the captured image or that the mark extends beyond the captured image, the process proceeds to step S806. On the other hand, if the mark is not located at the edge of the captured image and does not extend beyond the captured image, the process proceeds to step S805.

[0051] In step S805, the correction unit 402 arranges the selection template at each position in the search area, and calculates the similarity between the arranged selection template and the image area in the search area that overlaps with the selection template. The correction unit 402 then specifies the arrangement position of the selection template that has obtained the maximum similarity as the mark position, and specifies the size of the selection template as the mark size.

[0052] In step S806, the correction unit 402 obtains an extended bounding box, and sets an area including the extended bounding box and its surrounding area as a search area. Then, similar to step S805, the correction unit 402 arranges a selection template at each position in the search area, and obtains a similarity between the arranged selection template and an image area in the search area that overlaps with the selection template. At that time, as described above, if the selection template protrudes from the captured image when calculating the similarity between the selection template and the image area, the protruding part is ignored, and the similarity is obtained using only the part of the selection template that overlaps with the captured image. Then, the correction unit 402 specifies the arrangement position of the selection template that has obtained the maximum similarity as the mark position, and specifies the size of the selection template as the mark size.

[0053] By performing such processing, even if a part of the mark protrudes from the captured image, it is possible to more accurately determine the mark position and mark size of the mark.

[0054] In step S807, correction unit 402 determines whether or not all templates in the template group have been selected as selected templates. If the result of this determination is that all templates in the template group have been selected as selected templates, the process proceeds to step S808. On the other hand, if there are any templates remaining in the template group that have not yet been selected as selected templates, the process proceeds to step S802.

[0055] In step S808, the correction unit 402 specifies the mark position and mark size specified by the above process for the template with the highest similarity among the mark positions and mark sizes specified by the above process for each template in the template group as the final mark position and final mark size, respectively. Then, the correction unit 402 corrects the mark position determined by the detection unit 401 to the final mark position, and corrects the mark size determined by the detection unit 401 to the final mark size.

[0056] The adjustment execution unit 403 performs one or more of adjustment of the position of the mask stage 102, adjustment of the position of the substrate stage 104, and adjustment of the lens magnification of the image capturing unit 105, using the mark position and mark size corrected in step S808.

[0057] [Second embodiment] In the following embodiments including this embodiment, the differences from the first embodiment will be described, and unless otherwise noted below, it is assumed that the embodiments are the same as the first embodiment. In the first embodiment, the correction unit 402 performed template matching for all marks (mask marks and board marks) detected from the captured image. In contrast, in this embodiment, in order to reduce the number of times template matching is performed, the correction unit 402 performs template matching for marks that satisfy a condition among the marks (mask marks and board marks) detected from the captured image.

[0058] An example of the configuration of the main control unit 108 according to this embodiment will be described with reference to the block diagram of Fig. 9. In Fig. 9, the same functional units as those shown in Fig. 4 are given the same reference numerals, and descriptions of these functional units will be omitted.

[0059] The selection unit 902 selects a mark to be subjected to template matching in the correction unit 402, based on the mark position and mark size detected by the detection unit 401. The operation of the selection unit 902 will be described with reference to FIG.

[0060] A captured image 1001 includes a board mark 1002, a mask mark 1003, and a board mark 1004. A bounding box 1005 is set for the board mark 1002, a bounding box 1006 is set for the mask mark 1003, and a bounding box 1007 is set for the board mark 1004.

[0061] The size of the bounding box 1005 and the size of the bounding box 1006 are both close to the above-mentioned "desired size" (the difference in size is less than a specified value, and the size ratio is close to 1). However, the size of the bounding box 1007 is smaller than the above-mentioned "desired size". In this case, the selection unit 902 selects the bounding box 1005 and the bounding box 1006. The correction unit 402 corrects the mark position and the mark size based on the bounding box 1005 and the bounding box 1006 selected by the selection unit 902. Then, the adjustment execution unit 403 performs alignment based on the mark position and the mark size of the mark corrected by the correction unit 402.

[0062] Other conditions may also be applied to the above "conditions", for example, "a bounding box of a position whose distance from the center position of the captured image is less than a threshold distance". The above "conditions" may also include multiple conditions (for example, the above condition regarding size and the above condition regarding position).

[0063] With this configuration, according to this embodiment, the number of bounding boxes to be processed can be narrowed down, so that, for example, the number of times template matching is performed by the correction unit 402 can be reduced.

[0064] [Third embodiment] In the first embodiment, the correction unit 402 performs template matching using templates of different sizes. In contrast, in the present embodiment, the correction unit 402 performs template matching by enlarging or reducing the captured image and the bounding box in order to realize the same processing using templates of the same size.

[0065] An example of processing applicable to step S892 above will be described with reference to the flowchart of Fig. 11. Note that the processing according to the flowchart of Fig. 11 is a flowchart for processing one mark. However, by performing the processing according to the flowchart of Fig. 11 for each of the mask mark and the substrate mark, the mark positions and mark sizes of each of the mask mark and the substrate mark can be obtained.

[0066] In step S1101, when the detection unit 401 determines that the mark is a mask mark, the correction unit 402 obtains one template of the mask mark. On the other hand, when the detection unit 401 determines that the mark is a board mark, the correction unit 402 obtains one template of the board mark.

[0067] In step S1102, the correction unit 402 enlarges the captured image and the bounding box of the mark (enlargement with a magnification less than 1 is considered to be "reduction"). For example, consider a case where the processes of steps S1103 to S1106 are performed for each of the enlargement magnifications 0.5, 0.6, 0.7, ..., 1.8, 1.9, to 2.0 (16 types) in the enlargement magnification range. In this case, in the first step S1102, the correction unit 402 enlarges the captured image and the bounding box of the mark by 0.5 times. Then, in the second step S1102, the correction unit 402 enlarges the captured image and the bounding box of the mark by 0.6 times. Note that the positional relationship of the bounding box in the captured image before and after enlargement is assumed to be unchanged.

[0068] In step S1103, similarly to step S803 described above, the correction unit 402 sets an area including the enlarged bounding box (BBox) and the surrounding area of ​​the enlarged bounding box as a search area.

[0069] In step S1104, the correction unit 402 determines, based on the position and size of the enlarged bounding box, whether the bounding box is located at the edge of the enlarged captured image and whether it extends beyond the enlarged captured image, in the same manner as in step S804 above.

[0070] If it is determined that the target object is located at the edge of the enlarged captured image or that the target object extends beyond the enlarged captured image, the process proceeds to step S1106. On the other hand, if the target object is not located at the edge of the enlarged captured image and does not extend beyond the enlarged captured image, the process proceeds to step S1105.

[0071] In step S1105, the correction unit 402 specifies the arrangement position of the template where the maximum similarity is obtained as the mark position, in the same manner as in step S805 described above, and then stores in memory a set of the similarity, the mark position, and the enlargement factor in step S1102.

[0072] In step S1106, the correction unit 402 specifies the arrangement position of the template where the maximum similarity is obtained as the mark position, in the same manner as in step S806 above. Then, the correction unit 402 stores in memory a set of the similarity, the mark position, and the enlargement magnification in step S1102.

[0073] In step S1107, correction unit 402 determines whether or not the processes of steps S1102 to S1106 have been performed for all enlargement factors in the above enlargement factor range. If the result of this determination is that the processes of steps S1102 to S1106 have been performed for all enlargement factors in the above enlargement factor range, the process proceeds to step S1108. On the other hand, if there are any enlargement factors in the above enlargement factor range that have not yet been subjected to the processes of steps S1102 to S1106, the process proceeds to step S1102.

[0074] In step S1108, the correction unit 402 selects the set including the maximum "similarity" among the "similarity" in each set stored in the memory as a selected set. Then, the correction unit 402 obtains a position obtained by multiplying the mark position in the selected set by the reciprocal of the "enlargement factor" in the selected set as a final mark position. The correction unit 402 also obtains a size obtained by multiplying the size of the template by the reciprocal of the "enlargement factor" in the selected set as a final mark size. Then, the correction unit 402 corrects the mark position obtained by the detection unit 401 to the final mark position, and corrects the mark size obtained by the detection unit 401 to the final mark size. However, the correction of the mark position and the mark size may be performed by other methods as long as they are based on the position and size of the template and the enlargement factor.

[0075] The adjustment execution unit 403 performs one or more of adjustment of the position of the mask stage 102, adjustment of the position of the substrate stage 104, and adjustment of the lens magnification of the image capturing unit 105, using the mark position and mark size corrected in step S1108.

[0076] In this embodiment, when templates of different sizes are used, in addition to step S1108, the same process as in step S808 is performed to select a template and a similarity for correcting the mark position and the mark size.

[0077] [Fourth embodiment] In the first embodiment, the correction unit 402 performs template matching using a template prepared in advance. In contrast, in the present embodiment, an image in an image area having a mark position and a mark size corrected by the correction unit 402 (including a case where only one of them is corrected) is registered as a new template in the "memory of the main control unit 108."

[0078] For example, the correction unit 402 extracts an image in an image area having the mark position and mark size corrected in step S808 (including the case where only one of them is corrected) from the captured image. Then, the correction unit 402 registers the extracted image as a new template in the "memory of the main control unit 108."

[0079] As a result, when imaging continues under the same conditions, template matching can be performed using templates with similar image characteristics, which is expected to improve the accuracy of template matching.

[0080] [Fifth embodiment] In this embodiment, among the marks detected by the detection unit 401, the mark position and the mark size are output only for marks with high probability (mark probability), and the mark position and the mark size are not output for marks with low mark probability. In other words, the bounding box is set only for marks with high mark probability.

[0081] For example, the detection unit 401 acquires, as the mark probability, the output value (likelihood for the learned mark) of the softmax layer, which is the final layer in the neural network used to detect the mark from the captured image.The detection unit 401 then outputs the mark position and mark size of the mark whose mark probability is equal to or greater than a threshold, and does not output the mark position and mark size of the mark whose mark probability is less than the threshold.The method by which the detection unit 401 acquires the mark probability is not limited to the above method using a neural network.

[0082] With this configuration, alignment can be performed using the mark position and mark size obtained based on the bounding box corresponding to the mark with high mark accuracy.

[0083] If a mark with a mark accuracy equal to or greater than the threshold is not detected from the captured image, a bounding box is not set for the captured image, so the captured image is not used for alignment, and the main control unit 108 operates with the next captured image as the target. Such an operation of the main control unit 108 will be described with reference to the flowchart in FIG. 12.

[0084] In step S1201, the detection unit 401 obtains the mark probability of the mark detected from the captured image as a reliability, and judges whether the reliability is equal to or greater than a threshold. If the result of this judgment is that there is a mark whose reliability is equal to or greater than the threshold, the process proceeds to step S1202, and if there is no mark whose reliability is equal to or greater than the threshold, the process according to the flowchart in FIG. 12 ends.

[0085] In step S1202, a functional section downstream of the detection section 401 performs processing based on a bounding box corresponding to a mark whose reliability is equal to or greater than a threshold, and as a result, alignment is performed based on the mark position and mark size of the mark whose reliability is equal to or greater than a threshold.

[0086] In this way, according to this embodiment, alignment is performed based on the mark position and mark size of marks whose reliability is equal to or greater than a threshold, and marks whose reliability is less than the threshold are not involved in the alignment, thereby making it possible to prevent poor alignment based on the detection result of a mark with low reliability.

[0087] [Sixth embodiment] In this embodiment, the mark positions and mark sizes are corrected only for marks with a high maximum similarity in template matching, and alignment is performed based on the corrected mark positions and mark sizes.

[0088] For example, the correction unit 402 corrects the mark position and mark size only for marks whose maximum similarity is equal to or greater than a threshold, and does not correct the mark position and mark size for marks whose maximum similarity is less than the threshold. The adjustment execution unit 403 performs alignment based on the mark position and mark size corrected by the correction unit 402, and does not use the mark position and mark size that are not corrected by the correction unit 402 for alignment.

[0089] In other words, if there is no mark in the captured image whose maximum similarity is equal to or greater than the threshold, the mark position and size are not corrected, so that the captured image is not used for alignment, and the main control unit 108 operates on the captured image to be obtained next. Such an operation of the main control unit 108 will be described with reference to the flowchart in FIG.

[0090] In step S1301, the correction unit 402 determines whether the maximum similarity calculated for the marks detected from the captured image is equal to or greater than a threshold value. If the result of this determination is that there is a mark whose maximum similarity is equal to or greater than the threshold value, the process proceeds to step S1302, and if there is no mark whose maximum similarity is equal to or greater than the threshold value, the process according to the flowchart in FIG. 13 ends.

[0091] In step S1302, the adjustment execution unit 403 performs alignment based on the mark position and mark size corrected by the correction unit 402 for the marks whose maximum similarity is equal to or greater than a threshold value.

[0092] In addition, in each of the above embodiments, the case where alignment is performed using a captured image including a mask mark and a substrate mark (i.e., two marks) has been described. However, the above alignment device can also be applied to a case where alignment is performed using a captured image including one mark (only a mask mark, only a substrate mark, or only another type of mark). In this case, the alignment device acquires a captured image including a mark provided on an object held on a stage of the exposure device, which is captured by an imaging device of the exposure device, and detects the mark from the captured image. Then, the alignment device corrects the position and size of the detected mark to the position and size of the mark obtained by template matching using the area of ​​the mark in the captured image. Then, the alignment device adjusts one or more of the position of the stage and the lens magnification of the imaging device based on the corrected position and size.

[0093] In addition, the numerical values, processing timing, processing order, processing subject, data (information) acquisition method / destination / source / storage location, etc. used in each of the above embodiments are given as examples to provide a concrete explanation, and are not intended to be limited to these examples.

[0094] In addition, a part or all of the embodiments described above may be used in appropriate combination. In addition, a part or all of the embodiments described above may be used selectively.

[0095] (Other embodiments) The present invention can also be realized by a process in which a program for implementing one or more of the functions of the above-described embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., ASIC) that implements one or more of the functions.

[0096] The invention of this specification includes the following alignment apparatus, alignment method, and computer program.

[0097] (Item 1) a detection means for acquiring a captured image including a mark provided on an object held on a stage of the exposure apparatus, the captured image being captured by an imaging device of the exposure apparatus, and detecting the mark from the captured image; a correction means for correcting a position and a size of the mark detected by the detection means to a position and a size of the mark obtained by template matching using an area of ​​the mark in the captured image; an adjustment means for adjusting one or more of the position of the stage and the lens magnification of the imaging device based on the position and size corrected by the correction means; An alignment apparatus comprising:

[0098] (Item 2) The alignment device described in item 1, characterized in that the correction means sets a search area that includes a bounding box of the mark detected from the captured image and a surrounding area of ​​the bounding box, and corrects the position and size of the bounding box of the mark detected by the detection means to the position and size of the mark obtained by template matching with the search area.

[0099] (Item 3) The alignment device described in item 1 or 2, characterized in that the correction means determines whether the bounding box of the mark detected from the captured image is located at the edge of the captured image, sets an extended bounding box by extending the bounding box determined to be located at the edge, sets an area including the extended bounding box and the surrounding area of ​​the extended bounding box as a search area, and corrects the position and size of the mark to those obtained by template matching for the search area using a portion of a template that overlaps with the captured image.

[0100] (Item 4) The alignment device described in item 1, characterized in that the correction means sets a search area that includes the bounding box of a mark that satisfies a condition among the bounding boxes of marks detected from the captured image and the surrounding area of ​​the bounding box of the mark that satisfies the condition, and corrects the position and size of the bounding box of the mark that satisfies the condition to the position and size of the mark obtained by template matching with the search area.

[0101] (Item 5) The alignment device described in any one of items 2 to 4, characterized in that the correction means performs template matching on the search area using templates of different sizes and corrects the position and size of the mark to those obtained by template matching with the highest similarity.

[0102] (Item 6) The alignment device described in item 1, characterized in that the correction means enlarges the captured image and the bounding box of the mark detected from the captured image to different sizes, sets an area including the enlarged bounding box and the surrounding area of ​​the enlarged bounding box as a search area, performs template matching on the search area using one template, and corrects the position and size of the mark detected by the detection means based on the position and size of the mark obtained by the template matching.

[0103] (Item 7) moreover, 7. The alignment device according to any one of items 1 to 6, further comprising a registration means for extracting from the captured image an image within an image area having a position and size corrected by the correction means, and registering the extracted image as a new template.

[0104] (Item 8) 8. An alignment apparatus according to any one of items 1 to 7, characterized in that the marks include a substrate mark provided on a substrate and a mask mark provided on a mask on which a pattern to be transferred to the substrate is drawn.

[0105] (Item 9) The alignment apparatus described in item 8, characterized in that the adjustment means adjusts one or more of the position of a substrate stage that holds the substrate, the position of a mask stage that holds the mask, and the lens magnification of the imaging device based on the position and size of the substrate mark corrected by the correction means and the position and size of the mask mark corrected by the correction means.

[0106] (Item 10) 2. The alignment device according to item 1, wherein the correction means corrects the position and size of a mark detected by the detection means for a mark whose similarity in the template matching is equal to or greater than a threshold value to the position and size of the mark obtained by template matching using an area of ​​the mark in the captured image.

[0107] (Item 11) 11. The alignment apparatus according to any one of items 1 to 10, wherein the detection means detects the mark from the captured image using a neural network.

[0108] (Item 12) 12. The alignment apparatus according to any one of items 1 to 11, characterized in that the alignment apparatus is mounted on the exposure apparatus.

[0109] (Item 13) An alignment method performed by an alignment apparatus, comprising: a detection step in which a detection means of the alignment apparatus acquires a captured image including a mark provided on an object held on a stage of the exposure apparatus, the captured image being captured by an imaging device of the exposure apparatus, and detects the mark from the captured image; a correction step in which a correction means of the alignment device corrects the position and size of the mark detected in the detection step to the position and size of the mark obtained by template matching using an area of ​​the mark in the captured image; an adjustment step in which an adjustment means of the alignment device adjusts one or more of the position of the stage and the lens magnification of the imaging device based on the position and size corrected in the correction step; An alignment method comprising:

[0110] (Item 14) A computer program for causing a computer to function as each of the means of the alignment device according to any one of items 1 to 12.

[0111] The invention is not limited to the above-described embodiments, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0112] 101: Mask 102: Mask stage 103: Substrate 104: Substrate stage 105: Image capturing unit 106: Illumination optical system 107: Projection optical system 108: Main control unit

Claims

1. a detection means for acquiring a captured image, captured by an imaging device included in the exposure apparatus, including a first mark provided on a first object held on a first stage included in the exposure apparatus and a second mark provided on a second object held on a second stage included in the exposure apparatus, and for detecting the first mark and the second mark from the captured image; a correction means for correcting the position and size of each of the first mark and the second mark detected by the detection means to the position and size of the mark obtained by template matching using an area of ​​the mark in the captured image; an adjustment means for adjusting the position of the first stage based on the position and size of the first mark corrected by the correction means, adjusting the position of the second stage based on the position and size of the second mark corrected by the correction means, and adjusting the lens magnification of the imaging device based on the sizes of the first mark and the second mark corrected by the correction means; An alignment device comprising:

2. 2. The alignment device according to claim 1, wherein the correction means sets a search area that includes the bounding box of the mark and the surrounding area of ​​the bounding box, and corrects the position and size of the bounding box to the position and size of the mark obtained by template matching with the search area.

3. 2. The alignment device according to claim 1, wherein the correction means determines whether the bounding box of the mark is located at an edge of the captured image, sets an extended bounding box by extending the bounding box determined to be located at the edge, sets an area including the extended bounding box and the surrounding area of ​​the extended bounding box as a search area, and corrects the position and size of the mark to those obtained by template matching with the search area using a portion of the template that overlaps with the captured image.

4. 2. The alignment device according to claim 1, wherein the correction means sets a search area to an area including the bounding box of a mark that satisfies a condition among the bounding boxes of the marks and the surrounding area of ​​the bounding box of the mark that satisfies the condition, and corrects the position and size of the bounding box of the mark that satisfies the condition to the position and size of the mark obtained by template matching with the search area.

5. 3. The alignment apparatus according to claim 2, wherein the correction means performs template matching on the search area using templates of different sizes, and corrects the position and size of the mark to those obtained by template matching with the highest similarity.

6. 2. The alignment device according to claim 1, wherein the correction means enlarges the bounding boxes of the captured image and the mark to different sizes, sets an area including the enlarged bounding box and the surrounding area of ​​the enlarged bounding box as a search area, performs template matching on the search area using one template, and corrects the position and size of the mark based on the position and size of the mark obtained by the template matching.

7. moreover, 2. The alignment device according to claim 1, further comprising a registration means for extracting from the captured image an image within an image area having the position and size corrected by the correction means, and registering the extracted image as a new template.

8. 2. The alignment apparatus according to claim 1, wherein the second mark is a substrate mark provided on a substrate, and the first mark is a mask mark provided on a mask on which a pattern to be transferred to the substrate is drawn.

9. 2. The alignment device according to claim 1, wherein the correction means performs a process for each of the first mark and the second mark to correct the position and size of the mark detected by the detection means for a mark whose similarity in the template matching is equal to or greater than a threshold value to the position and size of the mark obtained by template matching using the area of ​​the mark in the captured image.

10. 2. The alignment apparatus according to claim 1, wherein the detection means detects the first mark and the second mark from the captured image using a neural network.

11. 2. The alignment apparatus according to claim 1, wherein the alignment apparatus is mounted on the exposure apparatus.

12. An alignment method performed by an alignment apparatus, comprising: a detection step in which a detection means of the alignment apparatus acquires a captured image, captured by an imaging device of the exposure apparatus, including a first mark provided on a first object held on a first stage of the exposure apparatus and a second mark provided on a second object held on a second stage of the exposure apparatus, and detects the first mark and the second mark from the captured image; a correction step in which a correction means of the alignment device corrects the position and size of each of the first mark and the second mark detected in the detection step to the position and size of the mark obtained by template matching using an area of ​​the mark in the captured image; an adjustment step in which an adjustment means of the alignment device adjusts the position of the first stage based on the position and size of the first mark corrected in the correction step, adjusts the position of the second stage based on the position and size of the second mark corrected in the correction step, and adjusts the lens magnification of the imaging device based on the sizes of the first mark and the second mark corrected in the correction step; An alignment method comprising:

13. A computer program for causing a computer to function as each of the means of the alignment apparatus according to any one of claims 1 to 11.

Images