JP7475902B2 - Inspection apparatus and method for generating reference image - Google Patents

Description

The present invention relates to an inspection device and a method for generating a reference image.

In the semiconductor device manufacturing process, there are two types of inspection equipment for masks and the like: die-to-die inspection and die-to-database inspection. In die-to-die inspection, a captured image is compared with a reference image to inspect for pattern defects. In die-to-database inspection, a reference image is generated from a database (DB) image corresponding to the design data of the mask. Then, the reference image is compared with the captured image to inspect the pattern.

Patent Document 1 discloses an inspection device that detects defects in a reticle. This inspection device scans a master reticle that is deemed to be substantially defect-free, and stores the scanned image (master image) in memory. The inspection device then detects defects by comparing the image of the reticle being inspected with the master image.

JP 2002-544555 A

As mentioned above, the inspection device performs inspection by comparing a reference image with a captured image. The integrity of the reference image has a significant impact on the inspection accuracy. However, when generating a reference image from a captured image, abnormalities may occur. For example, unexpected abnormalities such as variations in stage accuracy during imaging, fluctuations in the optical system, and data corruption may occur. If such abnormalities occur, it becomes impossible to generate an appropriate reference image. This can lead to problems such as normal patterns being detected as defects (false defects) and reduced sensitivity in defect detection.

This disclosure was made in consideration of these problems, and provides an inspection device that can generate more appropriate reference images, and a method for generating reference images.

The inspection device according to one aspect of this embodiment includes a stage for holding a sample, a light source for generating illumination light for illuminating the sample, a photodetector having a plurality of pixels and detecting detection light from a detection region illuminated by the illumination light, a drive mechanism for moving the relative position of the detection region with respect to the sample, and a processing unit for controlling the drive mechanism to image the sample using the photodetector, and for performing inspection by comparing the imaged image of the sample with a reference image. The processing unit drives the drive mechanism alternately in the arrangement direction of the pixels and in a cross direction intersecting the arrangement direction to obtain a capture image including a plurality of stripe regions whose longitudinal direction is the cross direction, and the image is obtained by driving the drive mechanism in the cross direction. An image of the stripe region captured in the cross direction is acquired as a first verification image, and the captured image and the first verification image are compared for the same stripe region to obtain a first comparison result. If the first comparison result is good, the reference image is generated based on the captured image of the stripe region. If the first comparison result is not good, an image of the stripe region captured again by driving the drive mechanism in the cross direction is acquired as a second verification image, and the captured image and the second verification image for the same stripe region are compared to obtain a second comparison result. If the second comparison result is good, the reference image is generated based on the captured image of the stripe region.

In the above inspection device, if the second comparison result is not good, the first verification image and the second verification image may be compared to obtain a third comparison result, and if the third comparison result is good, the reference image may be generated based on the first verification image or the second verification image.

In the above inspection device, if the third comparison result is not good, the imaging of the striped region may be repeated until the comparison result of the two verification images for the striped region becomes good.

In the above inspection device, if the first comparison result or the second comparison result is good, the processing unit may drive the driving mechanism in the array direction to capture a verification image of the next stripe region.

In the above inspection device, the drive mechanism may perform alignment after taking the capture image and before taking the first verification image.

A method for generating a reference image according to one aspect of this embodiment is a method for generating a reference image used in an image comparison inspection in an inspection device that includes a stage for holding a sample, a light source that generates illumination light for illuminating the sample, a photodetector that includes a plurality of pixels and detects detection light from a detection area illuminated by the illumination light, and a drive mechanism that moves the relative position of the detection area with respect to the sample, and includes the steps of: driving the drive mechanism alternately in the arrangement direction of the pixels and in a cross direction that crosses the arrangement direction to obtain a capture image including a plurality of stripe areas whose longitudinal direction is the cross direction; obtaining an image of the stripe area captured by driving the drive mechanism in the cross direction as a first verification image; and The method includes a step of comparing the capture image and the first verification image for the stripe region to obtain a first comparison result, a step of generating the reference image based on the capture image of the stripe region if the first comparison result is good, a step of obtaining an image of the stripe region captured again by driving the drive mechanism in the crossing direction as a second verification image if the first comparison result is not good, a step of comparing the capture image and the second verification image for the same stripe region to obtain a second comparison result, and a step of generating the reference image based on the capture image of the stripe region if the second comparison result is good.

In the above-mentioned method for generating a reference image, if the second comparison result is not good, the first verification image and the second verification image may be compared to obtain a third comparison result, and if the third comparison result is good, the reference image may be generated based on the first verification image or the second verification image.

In the above-mentioned method for generating a reference image, if the third comparison result is not good, the imaging of the striped region may be repeated until the comparison result of the two verification images for the striped region becomes good.

A method for generating a reference image described in any one of claims 6 to 8, wherein, in the above-mentioned reference image generating method, if the first comparison result or the second comparison result is good, the driving mechanism is driven in the array direction to capture a verification image of the next stripe region.
This may be done.

In the above-mentioned method for generating a reference image, after taking the capture image and before taking the first verification image, the driving mechanism may align the capture image with the first verification image.

The present disclosure provides an inspection device that can generate more appropriate reference images, and a method for generating reference images.

1 is a schematic diagram showing an overall configuration of an inspection device according to an embodiment of the present invention; FIG. 2 is a schematic diagram showing an inspection area and a stripe area of a sample. FIG. 2 is a control block diagram showing a configuration of the processing device. 1 is a flowchart showing a method for generating a reference image. FIG. 11 is a diagram for explaining a process for generating a reference image. FIG. 11 is a diagram for explaining a process for generating a reference image. FIG. 11 is a diagram for explaining a process for generating a reference image. FIG. 2 is a diagram for explaining a reference image.

The following describes an embodiment of the present invention with reference to the drawings. The following description shows a preferred embodiment of the present invention, and the scope of the present invention is not limited to the following embodiment. In the following description, parts with the same reference numerals indicate substantially the same contents.

The inspection device according to this embodiment performs inspection based on an image of a sample (hereinafter also referred to as a sample image). The inspection device inspects samples such as photomasks and semiconductor wafers on which fine patterns are formed. In other words, the inspection device performs defect inspection for foreign matter adhering to a patterned substrate.

In the following description, a photomask used in a photolithography process of a semiconductor device is used as a sample. For example, the inspection device performs mask-to-mask inspection by image comparison. Specifically, the inspection device captures an image of the entire inspection area of the photomask before use to generate a reference image. The inspection device performs a comparative inspection by capturing a sample image of the photomask after use and comparing the reference image with the sample image. By comparing a reference image captured before use of a physically identical mask with a sample image captured after use, deterioration over time can be evaluated. In addition, mask-to-mask inspection is not limited to comparing captured images of physically identical masks, but may also compare captured images of two masks having the same design pattern.

The inspection device and the method for generating a reference image according to this embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram showing a schematic diagram of the overall configuration of the inspection device 100. The inspection device 100 includes a stage 10, an imaging optical system 20, and a processing device 50. The processing device 50 is a computer including a processor, a memory, and the like.

In FIG. 1, for clarity of explanation, a three-dimensional orthogonal coordinate system of XYZ is shown. The Z direction is the vertical direction, and is parallel to the thickness direction of the sample 40. Therefore, the Z direction is the height direction. A pattern 41 such as a light-shielding film is formed on the upper surface of the sample 40. The Z direction is the normal direction of the pattern formation surface (main surface) of the sample 40. The X and Y directions are horizontal directions, and are parallel to the pattern direction of the sample 40. The Z direction is the thickness direction of the sample 40. Since the sample 40 is a photomask, it is a rectangular glass substrate. The X and Y directions are parallel to the edges of the sample 40.

The sample 40 to be imaged is placed on the stage 10. As described above, the sample 40 is a photomask, and is held on the stage 10. On the stage 10, the sample 40 is held parallel to the XY plane. The stage 10 is a three-dimensional drive stage having a drive mechanism 11. The processing device 50 controls the drive mechanism 11, causing the stage 10 to be driven in the XYZ directions.

The imaging optical system 20 includes a light source 21, a beam splitter 22, an objective lens 23, and a photodetector 24. Note that the imaging optical system 20 shown in FIG. 1 has been simplified as appropriate. The imaging optical system 20 may be provided with optical elements, lenses, optical scanners, mirrors, filters, beam splitters, and the like other than those described above. For example, the imaging optical system 20 may be a confocal optical system.

The light source 21 generates illumination light L11. The light source 21 may be a lamp light source, an LED (Light Emitting Diode) light source, a laser light source, or the like. The illumination light L11 from the light source 21 is incident on the beam splitter 22. The beam splitter 22 is, for example, a half mirror, and reflects approximately half of the illumination light L11 toward the sample 40. The illumination light L11 reflected by the beam splitter 22 is incident on the objective lens 23. The objective lens 23 focuses the illumination light L11 on the sample 40. This allows the surface (pattern formation surface) of the sample 40 to be illuminated. The optical axis OX of the objective lens 23 is parallel to the Z direction. The illumination light L11 illuminates a linear illumination area on the sample 40 with the X direction as the longitudinal direction.

The reflected light L12 reflected by the surface of the sample 40 is collected by the objective lens 23 and enters the beam splitter 22. The beam splitter 22 transmits approximately half of the reflected light L12. The reflected light L12 that passes through the beam splitter 22 enters the photodetector 24. This allows the photodetector 24 to image the sample 40. The objective lens 23 projects an enlarged image of the sample 40 onto the photodetector 24. A lens or the like may also be provided to focus the reflected light L12 on the light receiving surface of the photodetector 24.

In FIG. 1, the inspection device 100 is a microscope using a bright-field illumination method, but the illumination method of the inspection device 100 is not particularly limited. For example, when a transmitted illumination method is used, the photodetector 24 detects transmitted light that has passed through the sample 40. The detection light detected by the photodetector 24 is not limited to reflected light reflected by the sample 40, and may be transmitted light that has passed through the sample 40.

The photodetector 24 has an imaging element for capturing an image of the sample 40. The photodetector 24 is a CCD (Charge Coupled Device) camera or a CMOS (Complementary Metal Oxide Semiconductor) sensor. The photodetector 24 detects reflected light L12 from a detection area illuminated with illumination light L11.

The photodetector 24 has a number of pixels arranged in the X direction. Here, a TDI (Time Delay Integration) sensor is used as the photodetector 24. A number of pixels are arranged along the X direction on the light receiving surface of the photodetector 24. Of course, the photodetector 24 is not limited to a TDI sensor. The photodetector 24 may be a line sensor in which a number of pixels are arranged in a row.

The reflectance of the illumination light varies depending on whether the pattern 41 is present or not. For example, in the case of a photomask, the reflectance is high where the pattern 41 is present, and low where the pattern 41 is not present. Therefore, the amount of light received by the photodetector 24 changes depending on whether the pattern 41 is present or not. The photodetector 24 outputs a detection signal (detection data) corresponding to the amount of light received for each pixel to the processing device 50.

The stage 10 is a drive stage, and can move the sample 40 in the X and Y directions. A drive control unit 56 (see FIG. 3) of the processing device 50 controls the drive mechanism 11. The drive mechanism 11 moves the detection area on the sample 40 relatively. The drive control unit 56 moves the stage 10 in the X and Y directions, thereby changing the illumination position on the sample 40 illuminated by the illumination light L11.

As a result, any position on the sample 40 can be imaged, and almost the entire surface of the sample 40 can be inspected. Of course, the drive control unit 56 may drive the imaging optical system 20 instead of the stage 10. In other words, it is sufficient that the relative position of the imaging optical system 20 with respect to the stage 10 is movable. Alternatively, the illumination light L11 may be scanned using an optical scanner or the like.

Specifically, the stage 10 can move the sample 40 in the Y direction. The illumination light L11 illuminates a linear area along the X direction on the sample 40. The pixel arrangement direction on the photodetector 24 is the X direction. In other words, the pixel arrangement direction and the driving direction of the stage 10 are perpendicular to each other. In this way, a stripe-shaped area is imaged.

Figure 2 is a schematic diagram for explaining a captured image of an inspection area 400 of a sample 40. The inspection area 400 is an area where a comparison inspection is performed, and corresponds to the effective area of the sample 40. For example, in the sample 40, the inspection area 400 is an area where a pattern to be transferred in a photolithography process is formed. When the sample 40 is rectangular, the inspection area 400 is an area excluding the frame area around the periphery of the sample 40. Of course, the shape of the inspection area 400 is not particularly limited.

The captured image of the inspection area 400 is divided into a number of stripe areas 71 to 81. Each of the stripe areas 71 to 81 is a rectangular (band-like) area with the Y direction as the longitudinal direction and the X direction as the transverse direction. The transverse direction of the stripe area 81 is parallel to the arrangement direction of the pixels of the photodetector 24. The size of the transverse direction of the stripe area 81 (also called the stripe width) corresponds to the field of view of the imaging optical system 20. For example, the size of the transverse direction of the stripe area 81 is about 110 μm. The number of pixels of the photodetector 24 in the X direction is, for example, 2560 pixels.

The longitudinal direction of stripe regions 71 to 81 is the driving direction of the stage 10. For example, the stage 10 moves in the Y direction, causing the photodetector 24 to image the stripe region 71. Once an image of the stripe region 71 has been acquired, the stage 10 moves in the X direction to image the stripe region 72. In other words, the stage 10 moves in the -X direction by a distance corresponding to the stripe width. This shifts the illumination position and detection position in the X direction. Then, the stage 10 moves again in the Y direction, causing the photodetector 24 to image the stripe region 72.

As described above, the stage 10 alternates between movement in the Y direction and movement in the X direction. In this way, images are captured in the order of stripe region 71, stripe region 72, ..., stripe region 81. The processing device 50 sequentially acquires captured images of the multiple stripe regions. The processing device 50 combines the images of stripe regions 71 to 81 to generate an image of the entire inspection region 400. Note that the images of stripe regions 71 to 81 may have some overlap at their edges.

The longitudinal direction (Y direction) of the stripe regions 71 to 81 is the main scanning direction, and the lateral direction (X direction) of the stripe regions 71 to 81 is the sub-scanning direction. In FIG. 3, the number of stripe regions included in the inspection region 400 is 11, but the number of stripe regions is not particularly limited. In other words, the number of stripe regions is determined according to the size and stripe width of the inspection region 400. In addition, the main scanning direction is not limited to a direction perpendicular to the pixel arrangement direction, but may be any direction that intersects with it.

In two adjacent stripe regions, the movement direction in the Y direction is opposite. In other words, the main scanning direction is reversed in adjacent stripe regions. For example, in odd-numbered stripe regions 71, 73, 75, 77, 79, and 81, the stage 10 moves in the +Y direction to obtain the respective captured images. On the other hand, in even-numbered stripe regions 72, 74, 76, 78, and 80, the stage 10 moves in the -Y direction to obtain the respective captured images. In this way, the drive mechanism 11 moves the stage 10 in a zigzag manner, allowing the photodetector 24 to capture an image of the entire inspection region 400.

Figure 3 is a control block diagram showing the configuration of the processing device 50. The processing device 50 includes a first image acquisition unit 51, a second image acquisition unit 52, a comparison unit 53, a reference image storage unit 54, a reference image generation unit 55, and a drive control unit 56. The drive control unit 56 controls the stage 10 as described above. The drive control unit 56 then outputs the drive position of the stage 10 to the comparison unit 53. The comparison unit 53 performs inspection by comparing the first image with the second image.

The reference image storage unit 54 has a memory and the like, and stores a reference image of the entire inspection area 400. The reference image stored in the reference image storage unit 54 is read by the first image acquisition unit 51. In other words, the first image acquisition unit 51 acquires the reference image stored in the reference image storage unit 54 as the first image. The second image acquisition unit 52 acquires the image newly captured by the photodetector 24 as the second image.

For example, the comparison unit 53 finds the difference in gradation values between the first image (reference image) and the second image (captured image) and compares the difference with a threshold value. The comparison unit 53 detects defects such as pattern abnormalities and foreign matter based on the comparison result between the difference value and the threshold value. That is, at defective locations where foreign matter or the like is attached, the difference value becomes larger than the threshold value. The comparison unit 53 outputs the defective locations in association with their position coordinates. The position coordinates of the defective locations are specified by the drive position of the drive control unit 56, etc. The comparison unit 53 finds the XY coordinates in the inspection area 400 based on the drive position of the stage 10 and the pixel position in the photodetector 24.

In this way, the comparison unit 53 performs a comparative inspection of the images. Note that the processing in the comparison unit 53 can use various processing methods as in the past. Therefore, detailed explanations are omitted. The comparison unit 53 compares images for each stripe region. That is, in the comparison by the comparison unit 53, one stripe region is used as a unit. For example, as shown in FIG. 2, the comparison unit 53 compares a reference image R71 corresponding to the stripe region 71 with a captured image I71. Hereinafter, the comparative inspection of the reference image and the sample image will be referred to as an actual inspection.

The reference image generating unit 55 generates a reference image prior to the actual inspection. The reference image generating unit 55 writes the generated reference image to the reference image storage unit 54. A method for generating a reference image will be described with reference to FIGS. 4 to 7 as well as FIGS. 1 to 3. FIG. 4 is a flowchart showing the method for generating a reference image. The processing device 50 performs a process for verifying whether or not there is a sudden abnormality in the initially captured image. FIGS. 5 to 7 are schematic diagrams showing the process for generating a reference image.

First, the processing device 50 acquires a capture image (S11). Specifically, the driving mechanism 11 moves the stage 10 in a zigzag manner as described above, and the photodetector 24 captures a capture image of the entire inspection area 400. The driving control unit 56 drives the driving mechanism 11 alternately in the X direction and the Y direction to acquire a capture image including a plurality of stripe areas. The capture image is an image including captured images of a plurality of stripe areas 71 to 81 as shown in FIG. 2. Also, as shown in FIG. 5 and other figures, the images corresponding to the stripe areas 71 to 73 in the capture image C are taken as capture images C71 to C73, respectively. The capture image of the entire inspection area is stored in the reference image storage unit 54 as a reference image.

The processing device 50 then performs the verification (S12). To this end, the photodetector 24 captures a verification image of the first stripe. Specifically, the stage 10 moves linearly only in the Y direction, causing the photodetector 24 to capture a verification image of the stripe region 71. As shown in FIG. 5, the verification image of the stripe region 71 is referred to as verification image V71-1. Verification image V71-1 is also referred to as the first verification image.

The comparison unit 53 acquires a verification image V71-1 of the first stripe region 71. The comparison unit 53 compares the captured image C71 with the verification image V71-1 (first comparison process). This detects defects in the stripe region 71. As described above, the comparison unit 53 detects defects by comparing the difference value with a threshold value. The processing device 50 regards the result of the first comparison process as the first comparison result.

The processing device 50 judges whether the first comparison result is good (S13). The reference image generating unit 55 judges whether the number of defects detected in the first comparison process is equal to or greater than a predetermined value. If the number of defects is less than the predetermined value, the reference image generating unit 55 judges that the first comparison result is good because not many defects have occurred. If the number of defects is equal to or greater than the predetermined value, the reference image generating unit 55 judges that the first comparison result is not good because many defects have occurred. The judgment of whether the comparison result is good or bad may be made using not only the number of defects but also the defect size, etc.

If the first comparison result is good (YES in S13), there is no abnormality in the capture image C71 in the stripe region 71, so the process proceeds to step S20. Here, the capture image C71 is used as the reference image. In other words, the reference image generating unit 55 generates a reference image of the stripe region 71 based on the capture image C71.

If the first comparison result is not good (NO in S13), verification of the same stripe is performed (S14). Therefore, first, the processing device 50 reacquires the verification image of the first stripe. That is, the photodetector 24 captures the verification image of the first stripe region 71 again. Specifically, the driving mechanism 11 moves the stage 10 linearly only in the Y direction, and the photodetector 24 captures the verification image of the stripe region 71.

The verification image reacquired in S14 is referred to as reacquired image V71-2, as shown in FIG. 6. Reacquired image V71-2 is also referred to as the second verification image. Reacquired image V71-2 is an image of the same stripe region 71 as verification image V71-1. Comparison unit 53 compares captured image C71 with reacquired image V71-2 (second comparison process). This detects defects in stripe region 71. Processing device 50 regards the result of the second comparison process as the second comparison result.

The processing device 50 determines whether the comparison result in the second comparison process is good or not (S15). That is, the reference image generating unit 55 determines whether the second comparison result is good or not in the same manner as in S13, depending on the number of defects detected in the second comparison process, etc. If the second comparison result is good (YES in S15), there is no abnormality in the captured image C71 in the stripe region 71, so the processing device 50 proceeds to step S20.

The reference image generating unit 55 estimates that an anomaly 42 occurred when capturing the verification image V71-1, as shown in FIG. 5. The reference image generating unit 55 uses the captured image C71 as the reference image. In other words, it generates a reference image of the stripe region 71 based on the captured image C71. Note that examples of sudden anomalies 42 that occur during capturing include variations in stage precision during capturing, fluctuations in the optical system, and data corruption.

If the second comparison result is not good (NO in S15), the reference image generation unit 55 determines that there is an abnormality in the captured image C71 (S16). In this case, multiple defects occurred in both the first and second comparison processes. Therefore, as shown in FIG. 6, the reference image generation unit 55 estimates that an abnormality 42 occurred when the captured image C71 was taken. The reference image generation unit 55 does not use the captured image C71 as a reference image.

The reference image generating unit 55 replaces the capture image C71 in the stripe region 71 with the reacquired image V71-2 (S17). As shown in FIG. 7, the capture image C71 corresponding to the stripe region 71 is extracted from the capture image C of the entire inspection region 400, and replaced with the reacquired image V71-2. The reference image generating unit 55 rewrites the reference image in the reference image storage unit 54 using the reacquired image V71-2. As a result, the reference image of the stripe region 71 becomes the reacquired image V71-2.

The processing device 50 re-verifies the same stripe (S18). Specifically, the comparison unit 53 compares the reacquired image V71-2 with the verification image V71-1 (third comparison process). This detects defects in the stripe region 71. The processing device 50 sets the result of the third comparison process as the third comparison result.

The processing device 50 judges whether the third comparison result is good or not (S19). That is, the reference image generating unit 55 judges whether the third comparison result is good or not in the same manner as in S13, depending on the number of defects detected in the third comparison process, etc. If the third comparison result is good (YES in S19), the processing device 50 proceeds to step S20 because there is no abnormality in the reacquired image V71-2 in the first stripe region 71. The reference image generating unit 55 adopts the reacquired image V71-2 as the reference image. That is, the reference image generating unit 55 generates a reference image of the stripe region 71 based on the reacquired image V71-2. The reference image generating unit 55 rewrites the reference image in the reference image storage unit 54 using the reacquired image V71-2. As a result, the reference image of the stripe region 71 becomes the reacquired image V71-2.

If the third comparison result is not good (NO in S19), the process returns to step S14 and repeats. That is, the photodetector 24 captures a new verification image of the first stripe region 71. The reference image generation unit 55 sets the newly captured image of the stripe region 71 as the third verification image. Furthermore, the comparison unit 53 compares the second verification image, which is the reacquired image V71-2, with the third verification image, and sets the result as the fourth comparison result. Note that the second verification image has become the reference image (captured image C71) by the swap in S17. The reference image generation unit 55 judges whether the fourth comparison result is good or bad.

The processes of steps S14 to S19 are repeated until the comparison result is determined to be good in step S19. In other words, the processing device 50 performs processing to repeatedly capture a verification image of the stripe region 71 and compare the verification images with each other until the comparison result in step S15 or S19 is determined to be good.

If the comparison result is good in step S13, step S15, or step S19, the reference image generating unit 55 determines whether all stripes have been completed (S20). If not all stripes have been completed (NO in S20), the next stripe is verified (S21). In other words, verification is performed on the second stripe region 72.

Therefore, first, the processing device 50 acquires a verification image of the stripe region 72. The stage 10 moves in the X direction from the end position of the first stripe by the stripe width. Furthermore, the driving mechanism 11 moves the stage 10 in the Y direction, and the photodetector 24 images the second stripe region 72. The captured image C72 of the second stripe region 72 is compared with the verification image. Then, the process returns to S13, and the same process as for the first stripe region 71 is performed. The process is repeated until all stripes are completed (YES in S20).

As described above, the processing device 50 verifies the capture image for each stripe region. The processing device 50 determines whether or not there is an abnormality in the capture image by performing image comparison for each stripe region. The reference image generating unit 55 replaces the capture image with the verification image only in the stripe region where an abnormality has occurred, and updates the reference image. In other words, the reference image generating unit 55 uses the capture image as is in the stripe region where no abnormality has occurred to generate a reference image.

Figure 8 shows an example of a reference image R generated by the reference image generating unit 55. Here, an example is shown in which an abnormality 42 has occurred in the captured images C71 and C79 in two of the stripe regions 71 and 79. That is, the reference image generating unit 55 generates the reference image R based on the reacquired image V71-2, the captured images C72 to C78, the reacquired image V79-2, and the captured images C80 and C81. The portions of the reference image corresponding to the stripe regions 71 and 79 are rewritten to the reacquired image V71-2 and the reacquired image V79-2. The reference image generating unit 55 generates the reference image R of the entire inspection region by synthesizing the reacquired image V71-2, the captured images C72 to C78, the reacquired image V79-2, and the captured images C80 and C81.

In this way, the processing device 50 can generate an appropriate reference image. In other words, if a sudden abnormality 42 occurs when taking a capture image, the reference image of the stripe region where the abnormality 42 occurs is replaced with the verification image. Therefore, a reference image without or with fewer abnormalities 42 can be generated. This makes it possible to suppress the occurrence of false defects in actual inspection, and defects can be detected with high detection sensitivity. Therefore, actual inspection can be performed with high accuracy.

Also, if the comparison result is good in step S13, step S15, or step S19, the process proceeds to the verification of the next stripe. Therefore, if the comparison result is bad, the driving mechanism 11 drives the stage 10 in the Y direction, and the photodetector 24 recaptures the verification image of the same stripe region. Therefore, the reacquisition operation of the verification image for the entire surface of the inspection region 400 can be omitted. In other words, since it is necessary to reacquire the verification image only for the stripe region in which the abnormality 42 occurs, it is not necessary to reacquire the verification image of the stripe region in which the abnormality 42 does not occur. This makes it possible to shorten the processing time for generating the reference image. For example, if the inspection region 400 has 1000 stripe regions, it takes a long time, about one hour, to image the entire inspection region. The method of this embodiment makes it possible to omit such a long reacquisition operation.

The verification process may also be performed in real time. For example, a comparison inspection of the nth stripe region (n is an integer equal to or greater than 1) may be performed before capturing the verification image of the nth stripe region is completed. This allows the verification to be performed without delay. Furthermore, the stage 10 does not move in the X direction until the verification of each stripe region is completed. This allows the processing device 50 to quickly generate an appropriate reference image for each stripe region.

There is variation in the drive control of the stage 10. Therefore, the stage 10 may be aligned before capturing the verification image. For example, a slight positional deviation may occur in the stage 10 between capturing the capture image C71 and capturing the verification image V71-1. Specifically, after capturing the entire capture image C, when the stage 10 returns to the position of the stripe region 71, there may be a deviation of several pixels in the stage 10. Therefore, the processing device 50 aligns the stage 10 before capturing the verification image V71-1. In other words, by adjusting the position of the stage 10, it is possible to reduce the positional deviation to one pixel or less by aligning the capture image C71 and the verification image V71-1. The alignment of the stage 10 is performed, for example, by comparing two images.

After the capture image is taken and before the first verification image is taken, the driving mechanism 11 aligns the capture image with the first verification image. After the driving mechanism 11 aligns the stage 10 in the X direction, it starts driving in the Y direction. The driving mechanism 11 aligns the stage 10 so that the position of the first verification image coincides with the position of the capture image. By the driving mechanism 11 aligning the stage 10, the reacquired image V71-2 and the captured image C71 can be directly swapped. In this embodiment, the stage 10 does not move in the X direction until the verification of each stripe region is completed. In other words, the position of the stage 10 in the X direction is constant from step S13 to step S19. Therefore, it is not necessary to align the stage 10 in the X direction between the time when the verification image V71-1 is taken and the time when the reacquired image V71-2 is taken. Therefore, the reacquired image can be taken more easily.

The above-described inspection device 100 allows mask-to-mask inspection to be performed with high accuracy. In mask-to-mask inspection, it is possible to monitor changes in one mask over time. For example, in EUV photolithography, masks with pellicles are currently in the development stage, so masks without pellicles are used until masks with pellicles are put into practical use. With masks without pellicles, dust and other particles adhere to the mask surface over time. By performing an inspection of the entire mask surface, it is possible to evaluate the increase in dust. Of course, the inspection device 100 may also perform inspections other than mask-to-mask inspection, such as die-to-die inspection.

Note that in S17, the captured image C71 and the reacquired image V71-2 are swapped, but the captured image C71 and the verification image V71-1 may also be swapped. Furthermore, if the third comparison result is good, the reference image generating unit 55 may generate a reference image using the verification image V71-1. The reference image generating unit 55 may update the reference image using one of the two verification images when the comparison result becomes good.

The thresholds used in the comparison inspection (S12, S14, S18) when generating the reference image may be set to be stricter than the thresholds used in the actual inspection. This allows a more complete reference image to be generated. As described above, the actual inspection is a comparison inspection between the reference image and the sample image.

The processing device 50 is not limited to being a single physical device. In other words, the processing in the processing device 50 may be distributed and performed across multiple devices. For example, the processing device that acquires the detection data from the photodetector 24 and the processing device that performs the arithmetic processing to generate the reference image may be physically different devices. In addition, the reference image generated by the processing device 50 may be transmitted to another inspection device. In other words, the reference image generated by one inspection device 100 may be used by another inspection device to perform the actual inspection.

The sample 40 used to generate the reference image is preferably a photomask immediately after a pattern is formed. In other words, it is preferable for the inspection device 100 to take a capture image before the photomask is used in the photolithography process. In this way, a capture image can be taken in a clean state before any foreign matter or the like adheres to the photomask, or in a state before any defects such as missing patterns are generated. This can improve the inspection accuracy. If a defect is detected by actual inspection, the photomask is cleaned or repaired.

A part or all of the processing of the processing device 50 described above may be executed by a computer program. The above-mentioned program can be stored and supplied to a computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R/Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be provided to the computer by various types of transient computer-readable media. Examples of transient computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transient computer-readable medium can provide the program to the computer via a wired communication path such as an electric wire or optical fiber, or via a wireless communication path.

The above describes an embodiment of the present invention, but the present invention includes appropriate modifications that do not impair its objects and advantages, and is not limited to the above embodiment.

REFERENCE SIGNS LIST 100 Inspection device 10 Stage 11 Driving mechanism 20 Imaging optical system 21 Light source 22 Beam splitter 23 Objective lens 24 Photodetector 40 Sample 41 Pattern 42 Anomaly 50 Processing device 51 First image acquisition unit 52 Second image acquisition unit 53 Comparison unit 54 Reference image storage unit 55 Reference image generation unit 56 Driving control unit 400 Inspection area

Claims (12)

A stage for holding a sample;
a light source that generates illumination light for illuminating the sample;
a photodetector including a plurality of pixels for detecting detection light from a detection area illuminated with the illumination light;
a drive mechanism for moving a relative position of the detection region with respect to the sample;
a processing device that controls the drive mechanism to image the sample using the photodetector, and generates a reference image used in an image comparison inspection of the sample ;
The processing device comprises:
by alternately driving the driving mechanism in the pixel arrangement direction and a cross direction crossing the pixel arrangement direction, a capture image including a plurality of stripe regions having a longitudinal direction in the cross direction is obtained;
obtaining an image of the stripe region captured by driving the driving mechanism in the intersecting direction as a first verification image;
comparing the captured image with the first verification image for the same stripe region to obtain a first comparison result;
If the first comparison result is good, the captured image of the stripe region is set as the reference image for the stripe region ;
If the first comparison result is not good, an image of the stripe region is captured again by driving the driving mechanism in the intersecting direction, and the captured image is obtained as a second verification image;
comparing the captured image with the second verification image for the same stripe region to obtain a second comparison result;
When the second comparison result is good, the captured image of the stripe region is set as the reference image for the stripe region . A stage for holding a sample;
a light source that generates illumination light for illuminating the sample;
a photodetector including a plurality of pixels for detecting detection light from a detection area illuminated with the illumination light;
a drive mechanism for moving a relative position of the detection region with respect to the sample;
a processing device that controls the drive mechanism to image the sample using the photodetector, and generates a reference image used in an image comparison inspection of the sample ;
The processing device comprises:
by alternately driving the driving mechanism in the pixel arrangement direction and a cross direction crossing the pixel arrangement direction, a capture image including a plurality of stripe regions having a longitudinal direction in the cross direction is obtained;
obtaining an image of the stripe region captured by driving the driving mechanism in the intersecting direction as a first verification image;
comparing the captured image with the first verification image for the same stripe region to obtain a first comparison result;
If the first comparison result is good, generating the reference image based on the captured image of the stripe region;
If the first comparison result is not good, an image of the stripe region is captured again by driving the driving mechanism in the intersecting direction, and the captured image is obtained as a second verification image;
comparing the captured image with the second verification image for the same stripe region to obtain a second comparison result;
generating the reference image based on the captured image of the stripe region when the second comparison result is good;
If the second comparison result is not good, the first verification image is compared with the second verification image to obtain a third comparison result;
If the third comparison result is good, an inspection apparatus generates the reference image based on the first verification image or the second verification image . If the second comparison result is not good, the first verification image is compared with the second verification image to obtain a third comparison result;
2 . The inspection apparatus according to claim 1 , wherein, if the third comparison result is good, the reference image is generated based on the first verification image or the second verification image. 4. The inspection apparatus according to claim 2, further comprising: if the third comparison result is not favorable, the imaging of the striped region is repeated until the comparison result of the two verification images for the striped region becomes favorable. The inspection device according to any one of claims 1 to 4, wherein, when the first comparison result or the second comparison result is good, the processing device drives the driving mechanism in the array direction to capture a verification image of the next stripe region. 6. The inspection apparatus according to claim 1, wherein the driving mechanism performs alignment after taking the capture image and before taking the first verification image. A stage for holding a sample;
a light source that generates illumination light for illuminating the sample;
a photodetector including a plurality of pixels for detecting detection light from a detection area illuminated with the illumination light;
a drive mechanism for moving a relative position of the detection region with respect to the sample,
a step of alternately driving the driving mechanism in an arrangement direction of the pixels and an intersecting direction intersecting the arrangement direction to obtain a capture image including a plurality of stripe regions having a longitudinal direction in the intersecting direction;
acquiring an image of the stripe region captured by driving the driving mechanism in the intersecting direction as a first verification image;
comparing the captured image with the first verification image for the same stripe region to obtain a first comparison result;
if the first comparison result is good, taking the captured image of the stripe region as a reference image for the stripe region ;
acquiring, as a second verification image, an image of the stripe region captured again by driving the driving mechanism in the intersecting direction when the first comparison result is not good;
comparing the captured image with the second verification image for the same stripe region to obtain a second comparison result;
and if the second comparison result is good, setting the captured image of the stripe region as a reference image for the stripe region . A stage for holding a sample;
a light source that generates illumination light for illuminating the sample;
a photodetector including a plurality of pixels for detecting detection light from a detection area illuminated with the illumination light;
a drive mechanism for moving a relative position of the detection region with respect to the sample,
a step of alternately driving the driving mechanism in an arrangement direction of the pixels and an intersecting direction intersecting the arrangement direction to obtain a capture image including a plurality of stripe regions having a longitudinal direction in the intersecting direction;
acquiring an image of the stripe region captured by driving the driving mechanism in the intersecting direction as a first verification image;
comparing the captured image with the first verification image for the same stripe region to obtain a first comparison result;
generating the reference image based on the captured image of the stripe region if the first comparison result is good;
acquiring, as a second verification image, an image of the stripe region captured again by driving the driving mechanism in the intersecting direction when the first comparison result is not good;
comparing the captured image with the second verification image for the same stripe region to obtain a second comparison result;
generating the reference image based on the captured image of the stripe region when the second comparison result is good;
If the second comparison result is not good, the first verification image is compared with the second verification image to obtain a third comparison result;
a reference image generating method for generating the reference image based on the first verification image or the second verification image if the third comparison result is good; If the second comparison result is not good, the first verification image is compared with the second verification image to obtain a third comparison result;
The method for generating a reference image according to claim 7 , further comprising the step of generating the reference image based on the first verification image or the second verification image if the third comparison result is good. 10. The method for generating a reference image according to claim 8, further comprising the step of repeating imaging of the striped region until a comparison result of the two verification images for the striped region becomes favorable when the third comparison result is not favorable. A method for generating a reference image described in any one of claims 7 to 10, wherein when the first comparison result or the second comparison result is good, the driving mechanism is driven in the array direction to capture a verification image of the next stripe region. A method for generating a reference image described in any one of claims 7 to 11, wherein after capturing the capture image and before capturing the first verification image, the driving mechanism aligns the capture image with the first verification image.

Images