JPH10326739A - Positioning method and exposure method - Google Patents

Abstract

(57) Abstract: The present invention relates to an alignment method and an exposure method for aligning a mask and a photosensitive substrate, and more particularly to an alignment method and an exposure method that improve throughput while maintaining desired alignment accuracy. It is intended to provide a method. In an alignment method for detecting alignment marks (7, 8) formed on substrates (1, 3) and aligning the substrates (1, 3), at least a part of the alignment marks (7, 8) is registered as a template. Step S1 for taking in the images of the alignment marks 7, 8; Steps S3, S5, S10 for obtaining the similarity between the alignment marks 7, 8 and the template; and images for the alignment marks 7, 8 in accordance with the obtained correlation. It is configured to include a step S7 for setting the number of times of taking in and a step S14 for performing alignment of the substrates 1 and 3 based on the images of the alignment marks 7 and 8 for the set number of times.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure apparatus used in a photolithography process for manufacturing a semiconductor device or a liquid crystal display device, for aligning a mask on which an exposure pattern is drawn with a photosensitive substrate. The present invention relates to an alignment method and an exposure method.

[0002]

2. Description of the Related Art Conventionally, when a semiconductor device, a liquid crystal display device, or the like is manufactured using a photolithography technique, a pattern formed on a photomask or a reticle (hereinafter, referred to as a mask) is exposed to a photoresist through a projection optical system. An exposure apparatus that projects and exposes each shot area on a photosensitive substrate (hereinafter, referred to as a plate) such as a semiconductor wafer or a glass plate coated with a photosensitive agent such as a glass plate is used.

As an exposure apparatus, a plate is placed on a plate stage movable two-dimensionally in a plane perpendicular to the optical axis of the projection optical system, and the plate stage is stepped by the plate stage to form a mask. A step-and-repeat type exposure apparatus that sequentially repeats an operation of collectively exposing a pattern image to each shot area on a plate,
In particular, a reduction projection type exposure apparatus is frequently used.

On the other hand, recently, one chip such as a semiconductor element has been increasing in size, and it has been required to project a pattern having a larger area onto a plate by exposure.
Also, in manufacturing a liquid crystal display device, it is required to project and expose a large-area pattern onto a plate as the display becomes larger. Therefore, a scanning exposure type exposure apparatus capable of exposing a shot area wider than an effective exposure area of the projection optical system by synchronously scanning the mask and the plate with respect to the projection optical system has been developed. .

[0005] As the exposure apparatus of the scanning exposure system, there is an exposure apparatus for sequentially projecting and exposing the entire pattern of one mask over the entire surface of one plate, and a pattern projection to each shot area on the plate by reduction projection. In addition, a step-and-scan type exposure apparatus is known which performs scanning exposure and moves between shots by stepping.

A semiconductor element or a liquid crystal display device is formed by superposing a large number of circuit patterns on a plate. Therefore, when projecting and exposing the circuit patterns of the second and subsequent layers onto the plate, any one of the exposure devices is used. In this case, it is necessary to perform high-precision alignment of each shot area on the plate on which the circuit pattern is already formed with the pattern image of the mask to be exposed, that is, alignment between the mask and the plate.

[0007] As described in Japanese Patent Application Laid-Open No. 60-130742, for example,
A method of detecting a diffracted light generated when a laser beam is irradiated on a diffraction grating mark of a plate with a photodetector, and processing a signal waveform obtained to determine a position of an alignment mark,
A predetermined area including an alignment mark formed on a mask or a plate is illuminated, and an image of the alignment mark is CC
There is a method of capturing an image with a two-dimensional image sensor such as a D camera and performing predetermined image processing on the obtained image to determine the position of an alignment mark.

[0008] Among these alignment methods, particularly in the alignment method based on image processing, the degree of accuracy of the mark position detection reproducibility differs depending on the contrast between the mark portion of the alignment mark and the surrounding area. For a mark with low contrast, the detection reproducibility of the mark position deteriorates due to the electric noise of the CCD camera, etc., so in order to secure the desired alignment accuracy, it is necessary to detect the mark position many times and perform averaging processing. There is. Note that the contrast changes depending on the material and thickness of the film forming the alignment mark.

For this reason, the number of times of mark detection and the averaging process have been optimized with reference to a mark having a low contrast so that necessary alignment accuracy can be secured for a mark having any contrast.

[0010]

As described above, a method for improving the alignment accuracy is to perform position detection for the same alignment mark a plurality of times and to average the results of the plurality of position detections. And a method of increasing the number of alignment marks on the plate. In these methods, the number of times of mark detection needs to be increased as the alignment accuracy is increased. For this reason, if the number of averaging processes is optimized for a low-contrast mark, an unnecessary averaging process is performed even on a high-contrast mark that can be detected with high accuracy. However, the throughput is reduced by the amount.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art, and an object of the present invention is to provide an alignment method and an exposure method which improve throughput while maintaining desired alignment accuracy. To provide.

[0012]

The object of the present invention will be described with reference to FIGS. 1 to 11, which show one embodiment of the present invention. The above-mentioned object is achieved by alignment marks (7, 8) formed on substrates (1, 3). Detecting an image of the alignment mark (7, 8) in the alignment method for aligning the substrates (1, 3) (Step S1)
Setting the number of times the image of the alignment mark (7, 8) is taken in accordance with the taken image (step S7); and setting the number of images of the alignment mark (7, 8) in the substrate based on the set number of times. And a step (step S14) of performing the alignment of (1, 3).

In a positioning method for detecting the alignment marks (7, 8) formed on the substrates (1, 3) and positioning the substrates (1, 3), the alignment marks (7, 8) Registering at least a part of the template as a template, and an alignment mark (7,
8) Step of capturing the image (Step S1) and Step of calculating the correlation (similarity) between the alignment mark (7, 8) and the template (Steps S2 to S5, S)
10), a step of setting the number of times of taking the images of the alignment marks (7, 8) in accordance with the obtained correlation (step S7), and a step of setting the number of times of taking the images of the alignment marks (7, 8). And a step (step S14) of positioning the substrates (1, 3).

In the positioning method, when the correlation is smaller than a predetermined value, the positioning of the substrate based on the image may be stopped. Further, the above object is achieved by an exposure method for exposing an image of a pattern of a mask onto a photosensitive substrate having an alignment mark formed thereon, wherein the exposure method comprises positioning the photosensitive substrate using the above-described alignment method. You. Further, the above object is achieved by an exposure method, wherein an alignment mark is formed on a mask in the above exposure method, and the mask is positioned using the above alignment method.

According to the present invention, at least a part of the alignment mark (7, 8) is registered as a template, and the alignment is performed according to the obtained correlation (similarity) using image processing by a template matching method. Since the number of times of taking in the images of the marks (7, 8) is set, the alignment of the substrate can be performed without lowering the throughput while maintaining the desired alignment accuracy.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A positioning method and an exposure method according to one embodiment of the present invention will be described with reference to FIGS. First, a schematic configuration of an exposure apparatus using the alignment method and the exposure method according to the present embodiment will be described with reference to FIG. The exposure apparatus shown in FIG. 1 is a scanning type exposure apparatus having a projection optical system 2 for projecting an erect erect image at the same magnification.
In FIG. 1, the illumination optical system 30 illuminates a partial area of the mask 1 with a substantially uniform illuminance in a predetermined illumination area. Although not shown in detail, the illumination optical system 3
Reference numeral 0 denotes a light source such as an ultra-high-pressure ice-silver lamp that emits illumination light such as g-line and h-line, and an optical system including a fly-eye lens, a field stop, and a condenser lens through which illumination light from the light source passes. have.

The illuminated pattern image of the mask 1 is transferred to a glass plate 3 as a photosensitive substrate as an equal-size and erect image via a projection optical system 2 having a predetermined imaging characteristic. ing. The mask 1 is mounted on a fine movement stage 4, and the fine movement stage 4 is mounted on one side surface of a U-shaped carriage 6. The fine movement stage 4 is finely moved by the fine movement actuator 5 with respect to the carriage 6 in the X direction (the direction perpendicular to the paper surface in the drawing), the Y direction (the vertical direction in the paper surface in the drawing), and the rotation direction in the XY plane. You can do it. The plate 3 is mounted on the other side surface of the carriage 6 facing the mounting surface of the fine movement stage 4.

The carriage 6 is driven by a drive mechanism (not shown) so as to be movable in the X direction relative to the illumination optical system 30 and the projection optical system 2. With the movement of the carriage 6, the mask 1 and the plate are moved. 3 scans in synchronization with the projection optical system 2.

A moving mirror (not shown) is fixed to an end surface of the carriage 6 in the X direction. The moving mirror reflects laser light from a laser interferometer (not shown) to move the carriage 6 in the X direction. Are measured with high resolution. The carriage 6 is accurately positioned based on the position information from the laser interferometer. At the time of exposure, the carriage 6 is caused to scan the projection optical system 2 in the X direction, so that images of desired circuit patterns drawn on the mask 1 are sequentially projected and exposed on the plate 3.

In the scanning projection exposure apparatus according to the present embodiment, a pair of alignment systems AM1 and AM2 provided at predetermined intervals in the Y direction are arranged. Further, another pair of alignment systems AM3 and AM4 are arranged outside the alignment systems AM1 and AM2. The alignment system AM1, AM2 or the alignment system AM
Before scanning exposure, the position of the alignment mark 7, 8 or the alignment mark 47, 48 formed at a predetermined position on the mask 1 and the plate 3 is detected by using one or both of AM3 and AM4. The displacement between the plate 1 and the plate 3 is measured. By driving the fine movement stage 4 based on the amount of displacement and moving the mask 1 in the XY plane, the pattern already formed on the plate 3 and the circuit pattern of the mask 1 to be exposed from now on are superimposed. Positioning is performed so as to satisfy desired overlay accuracy.

The alignment systems AM1 and AM2 illuminate a predetermined area including the alignment marks 7 and 8 formed on the mask 1 and the plate 3, and illuminate the pattern image with a C image.
An optical system for forming an image on a TV camera such as a CD (solid-state imaging device), and the captured image is processed to detect the positions of the alignment marks 7 and 8 on the mask 1 and the plate 3, respectively. And a signal processing system for calculating the amount of displacement.

The alignment systems AM3 and AM4 are the same as those described in the above-mentioned Japanese Patent Application Laid-Open No. 60-130742, and apply laser light to the diffraction grating marks 47 and 48 provided on the mask 1 and the plate 3. The diffracted light generated by the irradiation is detected by a photodetector, and the resulting signal waveform is processed to measure the amount of displacement between the diffraction grating marks 47 and 48. In the alignment in the present embodiment, alignment systems AM1 and AM2 are mainly used, and alignment systems AM3 and AM4 are used when high alignment accuracy cannot be obtained with alignment systems AM1 and AM2 as described later.

The configuration of the alignment systems AM1 and AM2 mainly used in the alignment in this embodiment will be described in more detail with reference to FIG. 2 in addition to FIG. 1 and 2, an e-ray emitted from an extra-high pressure mercury lamp,
Illumination light, such as d-line, having a wavelength that does not expose the photoresist applied to the surface of the plate 3 to the optical fiber 3
2, the light is incident on an illumination optical system for an alignment system that illuminates predetermined regions of the mask 1 and the plate 3. The illumination optical system for the alignment system includes a condenser lens 9, a field stop 10, a lens 11, and the like.
Through the beam splitter 14 and the objective lens 13,
A predetermined area including the alignment mark 7 of the mask 1 is illuminated with substantially uniform illuminance.

Here, the illumination light transmitted through the mask 1 passes through the projection optical system 2 and illuminates a predetermined area including the alignment mark 8 of the plate 3 with substantially uniform illuminance. Illuminated alignment marks 7, 8 on mask 1 and plate 3
The pattern image of the objective lens 13 and the beam splitter 1
4. An image is formed on the imaging surface of the CCD camera 16 through the imaging lens 15.

The CCD camera 16 is a two-dimensional image sensor, in which pixels are arranged two-dimensionally in a horizontal direction (scanning direction) and a vertical direction. The amount of light accumulated in each pixel of the CCD camera 16 for a predetermined time is converted into an electric signal level corresponding to the magnitude. Further, after preprocessing such as AGC17 (auto gain control), the analog electric signal
In C18 (analog / digital converter), the signal is converted into a digital signal intensity corresponding to each two-dimensional pixel position. In the image processing unit 19 including the image processing device and the image storage device, the alignment marks 7 and 8 of the mask 1 and the plate 3 are obtained from the captured two-dimensional image.
Are detected, and the amount of displacement between the mask 1 and the plate 3 is obtained.

A rectangular mark as shown in FIG. 3 is formed on the mask 1 as the alignment mark 7 in advance. In addition, an alignment mark 8 is provided on the plate 3.
As shown in FIG. 4, four cross-shaped marks are formed in advance. FIG. 5 shows an example of an image including the alignment marks 7 and 8 between the mask 1 and the plate 3 captured by the CCD camera 16. The image processing unit 19 processes the image illustrated in FIG. 5 to detect the positions of the alignment marks 7 and 8 of the mask 1 and the plate 3, respectively, and determines the center position of the alignment mark of the mask 1 and the alignment mark of the plate 3. The shift amount ΔX of the center position is detected. In the scanning exposure apparatus according to the present embodiment, the entire scanning exposure apparatus is controlled by a main controller (not shown).

Next, a description will be given of a positioning method according to the present embodiment based on the above-described apparatus configuration. In the image processing unit 19 in the present embodiment, the positions of the alignment marks 7 and 8 between the mask 1 and the plate 3 are detected by a template matching method (correlation method). In the template matching method, first, an image to be detected is registered in advance as a template in a storage device (not shown) in the image processing unit 19. An image having the same shape as the alignment mark 7 of the mask 1 as shown in FIG. 6 and a plate 3 as shown in FIG.
Are partially registered as templates.

Next, with respect to the image taken by the CCD camera 16 and stored in the storage device in the image processing unit 19,
The image of the template is shifted by one pixel in the X direction and the Y direction as shown in FIG.
(Correlation value) is calculated, and the position of the image that best matches the image of the template is determined from the magnitude of the similarity R.
This is the template matching method. There are various calculation methods for determining the similarity R, and one of them is a calculation method using the following [Equation 1].

[0029]

(Equation 1)

Here, I1 (x, y) is the position (x, y).
The brightness of the captured image at y), I2 (x, y) is the brightness of the template at the position (x, y), and P is the number of pixels of the template.

In [Equation 1], the difference between the luminance of each pixel in the template and the luminance of the corresponding pixel in the picked-up image is added over all the pixels of the template, and the result is standardized by the number of pixels P of the template. It is a thing. Here, when the luminance is 256 gradations, the similarity R takes a value in the range of 0 to 255. When [Equation 1] is used, the smaller the similarity R, the better the match between the detected image and the template image, and the case where the similarity R = 0 completely matches the template image. It is.

As a method of registering a template, a method of mathematically calculating the luminance of each pixel from the contrast of a mark portion and its surrounding portion to create a template, and a method of actually storing an image of a mark on a mask 1 and a plate 3 by a CCD. There is a method of taking an image with the camera 16 and registering a part of the image as a template. Since the contrast of the alignment mark slightly changes depending on the material and thickness of the film forming the mark, it is difficult to mathematically create a template, and the latter method is often used.

Generally, in the template matching method, the better the detected image matches the image of the template, the higher the reproducibility of mark position detection. FIG. 9 shows the relationship between the similarity R defined by [Equation 1] and the detection reproducibility of the mark position. In FIG. 9, the horizontal axis represents the similarity R, and the vertical axis represents the degree of reproducibility of mark position detection. FIG.
According to the above, it is understood that the smaller the value of the similarity R defined in [Equation 1], the higher the accuracy of the mark position detection reproducibility. In the present embodiment, focusing on this relationship, the detection reproducibility of the mark position is estimated from the magnitude of the similarity R between the image of the template and the alignment mark, and the alignment accuracy and the reproducibility of the mark position are estimated based on the estimated reproducibility accuracy of the mark position. An alignment method that optimizes throughput is selected.

As described above, in order to increase the alignment accuracy, for example, there is a method in which the position of the same mark is detected a plurality of times, and the average value of the detected plurality of mark positions is used as the alignment mark position. If the number of times of averaging of the mark positions is set large and fixed in advance, it is possible to accurately detect the position of a mark having a low contrast. However, since averaging is performed even on a mark having a high contrast that does not require averaging, the throughput is reduced. Therefore, in the present embodiment, the relationship between the similarity R and the reproducibility of detection of the position of the alignment mark as shown in FIG. Thus, the relationship between the magnitude of the similarity R and the number of times of averaging at the time of mark detection is determined in advance.

FIG. 10 shows an example of a table showing the relationship between the size of the similarity R and the number of times of averaging at the time of mark detection. In FIG. 10, when the similarity R is 0 to 20, the images of the template and the mark match well, and it is determined that the detection reproducibility of the mark position is high, and the averaging (averaging number = 1) is not performed. The averaging is performed three times when the similarity R is 20 to 50, and five times when the similarity R is 50 to 100. If the similarity R exceeds 100, it is determined that there is no image that matches the template in the captured image, and the processing shifts to error processing. If such a table is created in advance, it is possible to detect the mark position with the optimum number of times of averaging based on the similarity R.

Next, a processing procedure by a main controller (not shown) in the above-described positioning method according to the present embodiment will be described with reference to a flowchart shown in FIG. In FIG. 11, first, in step S1, the images of the alignment marks 7 and 8 between the mask 1 and the plate 3 are transferred to the CCD camera
And sends it to the image processing unit 19. Next, in the image processing unit 19, the similarity RM of the mask 1 is obtained by a template matching method using the template of the mask 1 with respect to the captured image, and the position where the similarity RM has the minimum value is determined by the alignment of the mask 1. It is detected as the mark position of the mark 7 (step S2). Here, the alignment mark 7 of the mask 1 is formed of a chromium (Cr) film having a high reflectivity and generally has good contrast, so that the averaging process is not performed.

Next, in step S3, it is determined whether or not the value of the similarity RM of the obtained mask 1 is larger than 20, and if the value of the similarity RM is larger than 20, the mask is included in the captured image. It is determined that there is no 1 mark, and the process shifts to step S14 to perform error processing. As described above, since the alignment mark 7 of the mask 1 has a high reflectance, if an error occurs in step S3, it is highly likely that the image of the alignment mark 7 could not be captured. For this reason, as an example of the error processing, it is conceivable to work on the image of the alignment mark 7 again.

In step S3, if the value of the similarity RM is not more than 20, it is determined that the detection of the mark position of the alignment mark 7 of the mask 1 has been performed with a predetermined accuracy, and the process proceeds to step S4. In step S4, the image processing unit 19 obtains four similarities R from the smaller one of the similarities R obtained by the template matching method using the template for the plate 3 with respect to the fetched image. Are detected as the respective mark positions of the four marks on the plate 3. Next, step S
At 5, the average similarity RP is calculated from each similarity R of these four marks.

In step S6, the number of times of averaging N corresponding to the average similarity RP is determined with reference to the table illustrated in FIG. 10, and it is determined whether the average similarity RP exceeds 100, for example. If the average similarity RP exceeds 100,
Alignment mark 8 on plate 3 in captured image
Does not exist or the desired alignment accuracy cannot be obtained, and the process proceeds to step S14 to perform error processing. As an example of the error processing here, the procedure from step S1 is newly performed using the alignment mark on the mask 1 and the plate 3 formed at a position different from the alignment marks 7 and 8 of the mask 1 and the plate 3 described above. Alternatively, the image capturing may be performed, or the image capturing may be stopped and the diffraction light generated by irradiating the diffraction grating mark 48 provided on the plate 3 with laser light using another alignment system AM3 or AM4. The alignment may be performed by detecting the light detector, processing the obtained signal waveform and measuring the position of the diffraction grating mark 48.

If the average similarity RP does not exceed 100, it is determined that the averaging process is performed with the number of averaging N corresponding to the average similarity RP obtained from the table (step S7). In step S8, it is determined whether the number of averaging N is 1 or not.
The position shift amount ΔX between the mask 1 and the plate 3 is calculated from the position of the alignment mark between the mask 1 and the plate 3 detected up to 4 (step S13). Number of averaging N is 2
In the above case, the process proceeds to step S9. Step S9
Then, the image of the alignment mark 8 of the plate 3 is captured again, and then, in step S10, the four mark positions of the plate 3 are detected by a template matching method using the template for the plate 3 with respect to the captured image.

Next, in step S11, it is determined whether or not the determined number of times of averaging N has been reached. If the number of times of averaging has not been reached, the process returns to step S9 to take in the image again. I do. When the number of times of averaging N has been reached, in step S12, the mark positions of the alignment marks 8 on the plate 3 are determined by averaging the N detected mark positions. Next, the amount of displacement ΔX between the mask 1 and the plate 3 is calculated from the detected alignment mark position between the mask 1 and the plate 3 (step S13). Position shift amount Δ obtained in step S13
The fine movement stage 4 is driven based on X to
And the plate 3 are aligned. (Step S1
4)

If the amount of displacement .DELTA.X between the mask 1 and the plate 3 is determined by the above-described processing procedure, a desired alignment accuracy can be ensured by an optimum averaging process based on the reproducibility of mark position detection. become able to. Then, the image of the circuit pattern of the mask 1 is
In the exposure step of exposing to light, the mask 1 and the plate 3 are aligned and exposed using the above-described alignment method according to the present embodiment, so that exposure with excellent throughput while maintaining desired overlay accuracy is performed. Can be realized.

The present invention is not limited to the above embodiment, but can be variously modified. For example, in the above-described embodiment, the averaging process is performed by performing the mark position detection for the same alignment mark a plurality of times, but a plurality of alignment marks provided at a plurality of positions on the mask 1 and the plate 3 are used. Thus, it is also possible to create a table for changing the number of alignment marks detected for the similarity R and perform the alignment.

In the above-described embodiment, the scanning type projection exposure apparatus in which the carriage is moved is used to carry out the positioning method of the present invention. However, the present invention is not limited to this, and the present invention is not limited to this. Of course, the present invention can be applied to an exposure apparatus of a repeat system or a step-and-scan system. Further, the present invention can be applied to a charged particle beam exposure apparatus using an electron beam or X-ray as a light source.

[0045]

As described above, according to the present invention, the number of times of taking an image of an alignment mark is set according to the obtained similarity by using the image processing by the template matching method. The alignment of the substrate can be performed without lowering the throughput while maintaining the alignment accuracy.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of a scanning projection exposure apparatus used for an alignment method according to an embodiment of the present invention.

FIG. 2 is a diagram showing a schematic configuration of an alignment system used for a positioning method according to an embodiment of the present invention.

FIG. 3 is a diagram showing a shape of an alignment mark used in one embodiment of the present invention.

FIG. 4 is a diagram showing a shape of an alignment mark used in one embodiment of the present invention.

FIG. 5 is a diagram illustrating an example of an image including alignment marks 7 and 8 of a mask 1 and a plate 3 captured by a CCD camera 16;

FIG. 6 is a diagram illustrating a shape of a template registered in a storage device in the image processing unit 19;

FIG. 7 is a diagram illustrating a shape of a template registered in a storage device in the image processing unit 19;

FIG. 8 is a diagram illustrating a template matching method.

FIG. 9 is a diagram illustrating a relationship between a similarity R defined by [Equation 1] and detection reproducibility of a mark position.

FIG. 10 is a diagram illustrating an example of a table representing a relationship between the number of times of averaging at the time of mark detection with respect to the magnitude of the similarity R.

FIG. 11 is a flowchart showing a processing procedure in a positioning method according to an embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 1 mask 2 projection optical system 3 plate 4 fine movement stage 5 fine movement actuator 6 carriage 7, 8, 47, 48 alignment mark 9 condenser lens 10 field stop 13 objective lens 14 beam splitter 15 imaging lens 16 CCD camera 17 AGC 18 ADC 19 image Processing unit 30 Illumination optical system 32 Optical fiber AM1, AM2, AM3, AM4 Alignment system

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI H01L 21/30 525X

Claims (5)

[Claims] 1. An alignment method for detecting an alignment mark formed on a substrate and aligning the substrate, wherein an image of the alignment mark is captured, and the alignment mark is provided in accordance with the captured image. Setting a number of times of capturing the image of the above, and positioning the substrate based on the set number of images of the alignment mark. 2. An alignment method for detecting an alignment mark formed on a substrate and aligning the substrate, wherein at least a part of the alignment mark is registered as a template, and an image of the alignment mark is registered. Taking in, obtaining the correlation between the alignment mark and the template, setting the number of times to take in the image of the alignment mark in accordance with the obtained correlation, and setting the number of times of the alignment mark to be set. Positioning the substrate based on an image. 3. The alignment method according to claim 2, wherein when the correlation is smaller than a predetermined value, the alignment of the substrate based on the image is stopped. 4. An exposure method for exposing an image of a pattern of a mask on a photosensitive substrate on which an alignment mark is formed, wherein the photosensitive substrate is positioned by using the alignment method according to claim 1 or 2. Exposure method. 5. The exposure method according to claim 4, wherein an alignment mark is formed on the mask, and the mask is positioned by using the alignment method according to claim 1 or 2. Exposure method.