Measurement methods and systems, devices, and storage media

Abstract

A measurement method, system, device, and storage medium are disclosed. The method includes: acquiring a detection image and a template image of a marker to be measured, wherein the template image has a detection range and includes a first region of interest (ROI) with a preset positional relationship relative to the detection range; performing ROI localization processing on the detection image using the template image, wherein the ROI localization processing includes: matching the detection image with the template image within the detection range, identifying a region in the detection image with the highest similarity to the template image within the detection range as a matching region, and using the detection image of the matching region as the matching image; and determining a second ROI in the detection image based on the preset positional relationship between the first ROI in the template image and the detection range, and using the detection image within the second ROI as the ROI image. This invention is beneficial for improving the localization accuracy of the marker to be measured.

Description

Technical Field

[0001] The present invention relates to the field of optical detection technology, and in particular to a measurement method and system, device and storage medium. Background Technology

[0002] In the manufacturing process of semiconductor devices, in order to detect the target under test, it is necessary to first locate the position of the target under test to determine the region of interest (ROI), and then use the image of the region of interest to obtain the parameters to be tested.

[0003] Especially in the process of detecting alignment errors, in order to ensure that patterns formed by different film layers or different processes can be accurately aligned, it is necessary to detect the alignment error of the overlay mark. However, with the miniaturization of semiconductor devices, higher requirements are placed on the positioning accuracy of the overlay mark.

[0004] Currently, the accuracy of mechanical positioning of the target under test is low, which is difficult to meet actual needs. Summary of the Invention

[0005] The problem solved by the embodiments of the present invention is to provide a measurement method, system, device and storage medium that is beneficial to improving the positioning accuracy of the marker to be measured.

[0006] To address the aforementioned problems, this invention provides a measurement method for measuring a marker to be tested, the marker having a region of interest (ROI). The measurement method includes: acquiring a detection image and a template image of the marker to be tested, the template image having a detection range and including a preset positional relationship between a first ROI and the detection range, the first ROI being the region of the ROI of the marker to be tested within the template image; performing ROI localization processing on the detection image using the template image, the ROI localization processing including: matching the detection image with the template image within the detection range, in the detection image, acquiring the region with the highest similarity to the template image within the detection range as a matching region, the detection image of the matching region being used as a matching image; determining a second ROI in the detection image based on the preset positional relationship between the first ROI in the template image and the detection range, the second ROI being the region of the ROI of the marker to be tested within the detection image, the detection image within the second ROI being used as an ROI image, the relative positional relationship between the ROI image and the matching image being the same as the preset positional relationship.

[0007] Accordingly, this embodiment of the invention also provides a measurement system for measuring a marker to be tested, the marker to be tested having a region of interest (ROI). The measurement system includes: an image acquisition module for acquiring a detection image and a template image of the marker to be tested, the template image having a detection range and including a preset positional relationship of a first ROI relative to the detection range, the first ROI being the region of the ROI of the marker to be tested within the template image; a positioning module for performing ROI positioning processing on the detection image using the template image, the positioning module including: a matching unit for matching the detection image with the template image within the detection range, acquiring the region with the highest similarity to the template image within the detection range in the detection image as the matching region, the detection image of the matching region being used as the matching image; and a ROI determination unit for determining a second ROI in the detection image based on the preset positional relationship of the first ROI in the template image relative to the detection range, the second ROI being the region of the ROI of the marker to be tested within the detection image, the detection image within the second ROI being used as the ROI image, the relative positional relationship between the ROI image and the matching image being the same as the preset positional relationship.

[0008] Accordingly, embodiments of the present invention also provide an apparatus, including at least one memory and at least one processor, wherein the memory stores one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the measurement method described in the embodiments of the present invention.

[0009] Accordingly, embodiments of the present invention also provide a storage medium storing one or more computer instructions, which are used to implement the measurement method described in the embodiments of the present invention.

[0010] Compared with the prior art, the technical solution of the embodiments of the present invention has the following advantages:

[0011] In the measurement method provided by this invention, a template image is first used to perform region of interest (ROI) localization processing on the detection image of the marker under test. By matching the template image on the detection image, a matching region is obtained. After the matching is completed, a corresponding second ROI can be determined in the detection image based on the preset positional relationship between the first ROI in the template image and the detection range. The second ROI is the region of the ROI of the marker under test in the detection image, thereby obtaining an image of interest. Compared with the mechanical localization scheme of the marker under test, this invention, through template matching, not only facilitates faster localization of the marker under test in the detection image, but also facilitates more accurate determination of the position of the second ROI in the detection image, thereby improving the localization accuracy of the marker under test. Correspondingly, the accuracy of the image of interest is improved, and the measurement accuracy when measuring the marker under test is enhanced.

[0012] In an optional embodiment, the detection range includes multiple sub-detection ranges, each of which is located in a local area of ​​the template image. Each sub-detection range covers a portion of the image of the target marker, and the region of interest is located in the detection image using the template image within each sub-detection range. Therefore, by adopting a method of template matching by region, this embodiment of the invention can accurately determine the position of the second region of interest in the detection image even if there are large positional deviations between different types of target markers or if the size of the target marker is deformed due to process issues. Attached Figure Description

[0013] Figure 1 This is a flowchart of an embodiment of the measurement method of the present invention;

[0014] Figure 2 This is a top view of the overlay markings in a measurement method according to an embodiment of the present invention;

[0015] Figure 3 yes Figure 2 The image shown is a detection image of one embodiment of the overprinted markings;

[0016] Figure 4 yes Figure 2 A template image of one embodiment of the overprinted markings shown;

[0017] Figure 5 This is a flowchart of an embodiment of the measurement method of the present invention, specifically the region of interest localization process.

[0018] Figure 6 This is a schematic diagram of a detection image after matching processing in a measurement method according to an embodiment of the present invention;

[0019] Figure 7 This is a schematic diagram of an embodiment of the measurement method of the present invention, which obtains the alignment error between the first mark and the second mark;

[0020] Figure 8 This is a schematic diagram of the detected image in another embodiment of the measurement method of the present invention;

[0021] Figure 9 This is a schematic diagram of a template image in another embodiment of the measurement method of the present invention;

[0022] Figure 10 This is a schematic diagram of the first template image after occluding the image information within the first sub-detection range of the template image in another embodiment of the measurement method of the present invention;

[0023] Figure 11 This is a schematic diagram of the detection image after matching the detection image with the first template image in another embodiment of the measurement method of the present invention;

[0024] Figure 12 This is a schematic diagram of the second template image after extracting the second sub-detection range of the template image in another embodiment of the measurement method of the present invention;

[0025] Figure 13 This is a schematic diagram of the detection image after matching the detection image with the second template image in another embodiment of the measurement method of the present invention;

[0026] Figure 14 This is a schematic diagram of a template image in another embodiment of the measurement method of the present invention;

[0027] Figure 15 This is a functional block diagram of an embodiment of the measurement system of the present invention;

[0028] Figure 16 This is a hardware structure diagram of a device provided in an embodiment of the present invention. Detailed Implementation

[0029] As can be seen from the background technology, the current positioning accuracy of markers is low and cannot meet practical needs.

[0030] To address the aforementioned technical problem, embodiments of the present invention provide a measurement method. (See reference...) Figure 1 The flowchart illustrates an embodiment of the measurement method of the present invention. The measurement method described in this embodiment is used to measure a marker to be measured, the marker having a region of interest, and the measurement method includes the following basic steps:

[0031] Step S1: Obtain the detection image and template image of the mark to be tested. The template image has a detection range and includes a preset positional relationship of a first region of interest relative to the detection range. The first region of interest is the region of interest of the mark to be tested in the template image.

[0032] Step S2: Use the template image to perform region of interest localization processing on the detected image.

[0033] In this embodiment of the invention, a template image is first used to perform region of interest (ROI) localization processing on the detection image of the marker under test. By matching the template image on the detection image, a matching region is obtained. After matching is completed, a corresponding second ROI can be determined in the detection image based on the preset positional relationship between the first ROI in the template image and the detection range. The second ROI is the region of the ROI of the marker under test in the detection image, thereby obtaining an image of interest. Compared with the mechanical localization of the marker under test, this embodiment of the invention, through template matching, not only facilitates faster localization of the marker under test in the detection image, but also facilitates more accurate determination of the position of the second ROI in the detection image, thereby improving the localization accuracy of the marker under test. Consequently, the accuracy of the image of interest is improved, and the measurement accuracy when subsequently measuring the marker under test is enhanced.

[0034] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0035] Reference Figures 1 to 4 , Figure 2 This is a top view of an embodiment of the overlay marking of the present invention. Figure 3 yes Figure 2 The image shown is a detection image of one embodiment of the overprinted marking. Figure 4 yes Figure 2 The template image of one embodiment of the overprinted mark is shown. Step S1 is executed to obtain the detection image 200 and the template image 300 of the mark to be tested 100. The template image 300 has a detection range 300S and includes a first region of interest 300R with respect to the detection range 300S in a preset positional relationship. The first region of interest 300R is the region of the region of interest of the mark to be tested 100 in the template image 300.

[0036] In this embodiment, the test mark 100 includes a first mark 110 and a second mark 120. Correspondingly, the first mark 110 and the second mark 120 each have their own corresponding region of interest (not shown).

[0037] Specifically, the mark to be tested 100 is an overlay mark 100. The alignment error of the overlay mark 100 is obtained by detecting the alignment error between the first mark 110 and the second mark 120. To ensure precise alignment between patterns formed by different film layers or different processes, it is necessary to detect the alignment error between the overlay marks to assess the reliability of the process.

[0038] In this embodiment, the first mark 110 and the second mark 120 are formed by different processes, or the first mark 110 and the second mark 120 are located on different layers. Specifically, the mark to be tested 100 is located in a substrate (not shown), and the first mark 110 and the second mark 120 are located at different heights on the substrate, or the first mark 110 and the second mark 120 are formed in different processes. As an example, the substrate can be a wafer or a chip. The fact that the first mark 110 and the second mark 120 are located at different heights on the substrate means that, along the normal direction of the substrate surface, the first mark 110 and the second mark 120 are spatially located on different layers of the substrate.

[0039] As an example, the orthographic projection of the first mark 110 onto the layer containing the second mark 120 is located inside the second mark 120.

[0040] In this embodiment, the mark to be tested 100 includes multiple mark patterns 150 located on the same layer. Correspondingly, the first mark 110 includes multiple mark patterns 150, and the second mark 120 also includes multiple mark patterns 150. Specifically, the mark pattern 150 corresponding to the first mark 110 is the inner alignment pattern (i.e., inner bar) in the overprinted mark 100, and the mark pattern 150 corresponding to the second mark 120 is the outer alignment pattern (i.e., outer bar) in the overprinted mark 100.

[0041] In this embodiment, the first mark 110 includes a first direction (e.g., along the first direction). Figure 2 Extending along the X direction (as shown in the middle) and along the second direction (as shown in the middle X direction) Figure 2 The second mark 120 includes a pair of first alignment patterns 111 arranged in parallel along the Y direction, and a pair of second alignment patterns 112 extending along the second direction and arranged in parallel along the first direction; the second mark 120 includes a pair of third alignment patterns 121 extending along the first direction and arranged in parallel along the second direction, and a pair of fourth alignment patterns 122 extending along the second direction and arranged in parallel along the first direction.

[0042] As an example, the first mark 110 is centrally symmetrical, and the second mark 120 is centrally symmetrical.

[0043] In this embodiment, an imaging system is used, and the relative height between the imaging system and the overlay mark 100 is at a preset focusing height position. The imaging system is used to capture an image of the overlay mark 100 to obtain the detection image 200.

[0044] The template image 300 is a high-quality image that meets the design requirements, so as to use the template image 300 as a reference image to determine the position of the mark to be tested 100 in the detection image 200.

[0045] It should be noted that after capturing the test mark 100 to obtain the detection image 200, the size of the detection image 200 may be larger than the size of the area where the test mark 100 is located. For example, the detection image 200 may also contain background information. Meanwhile, the size of the template image 300 is the same as the size of the area where the test mark 100 is located, thus allowing the location of the test mark 100 to be searched from the detection image 200 using the template image 300. For example, the size of the detection image 200 may be 1200*1600, while the size of the template image 300 may be 600*600.

[0046] As an example, the design image corresponding to the marker to be tested 100 can be obtained as the template image 300. The design image can be a binarized image or a design layout.

[0047] In other embodiments, obtaining the template image corresponding to the mark to be tested includes: obtaining an image of a standard mark as the template image, wherein the standard mark is identical to the mark to be tested and conforms to design requirements (e.g., in the case that the mark to be tested is an overprinted mark, the standard mark conforms to alignment requirements). In other embodiments, obtaining the template image corresponding to the mark to be tested includes: providing a theoretical model of the interaction between the sample and light, wherein the theoretical model represents the relationship between the intensity distribution of the emitted light after the light interacts with the sample and the physical parameters of the sample; substituting the physical parameters of the mark to be tested into the theoretical model to obtain a simulated image, which serves as the template image.

[0048] Therefore, in this embodiment, the template image 300 includes an image 310 corresponding to the first mark 110 and an image 320 corresponding to the second mark 120.

[0049] Specifically, the template image 300 defines a detection range 300S, and the template image 300 includes a preset positional relationship between a first region of interest 300R and the detection range 300S. The first region of interest 300R is the region of the region of interest of the target marker 100 in the template image 300.

[0050] In one specific embodiment, the entire region of the template image 300 is used as the detection range 300S. That is, the entire region of the template image 300 is subsequently used to match the detection image 200. Accordingly, the pixels in the entire region of the template image 300 are used to participate in the calculation of the matching degree.

[0051] In this embodiment, the center C of the detection range 300S coincides with the center of the test mark 100 to be processed for region of interest localization; the preset positional relationship of the first region of interest 300R relative to the detection range 300S is the positional relationship between the center of each first region of interest 300R and the center C of the detection range 300S.

[0052] It should also be noted that the center C of the detection range 300S is easy to obtain. Therefore, a first region of interest 300R can be set for the template image 300 to obtain the positional relationship between the center of each first region of interest 300R and the center C of the detection range 300S.

[0053] For example, if the first mark 110 is centrally symmetrical, then in the template image 300, the center C of the detection range 300S is obtained, and a first region of interest 300R is selected at any position of the mark graphic 150 of the first mark 110. Then, the selected first region of interest 300R is rotated with the center C of the detection range 300S as the reference point, thereby determining the first region of interest 300R on the remaining mark graphic 150. Then, a first region of interest 310R corresponding to the first mark 110 is obtained in the template image 300, and the first region of interest 310R is centrally symmetrical.

[0054] In a similar manner as described above, a first region of interest 320R corresponding to the second marker 120 is obtained in the template image 300, and the first region of interest 320R corresponding to the second marker 120 is also centrally symmetrical.

[0055] Continue to refer to Figures 1 to 4 Then, perform step S2, using the template image 300 to perform region of interest localization processing on the detection image 200.

[0056] The location of the marker 100 to be tested is determined in the detection image 200 by region of interest localization processing.

[0057] Reference Figure 5 and Figure 6 , Figure 5 This is a flowchart of an embodiment of region of interest localization processing. Figure 6This is a schematic diagram of an embodiment of the detected image after matching processing. The region of interest localization processing includes: performing step S21, matching the detected image 200 with the template image 300 within the detection range 300S, and obtaining the region in the detected image 200 that has the highest similarity to the template image 300 within the detection range 300S as the matching region 200M (e.g., ...). Figure 6 As shown), the detected image of the matching region is used as the matching image (not shown).

[0058] It should be noted that, Figure 6 Only the image corresponding to the matching region 200M in the detection image 200 is shown (i.e., the matching image).

[0059] By matching the template image 300 on the detection image 200, a matching region 200M is obtained. After the matching is completed, a second region of interest 200R can be determined in the detection image 200 according to the preset positional relationship between the first region of interest 300R in the template image 300 and the detection range 300S.

[0060] By employing template matching, it is not only easier to locate the position of the marker 100 to be tested in the detection image 200 more quickly, but also easier to determine the position of the second region of interest 200R in the detection image 200 more accurately, thereby improving the positioning accuracy of the marker 100 to be tested.

[0061] In this embodiment, the template image 300 covering the entire region is used to perform region of interest localization processing on the detection image 200.

[0062] As an example, the matching process includes: using the same matching window as the detection range 300S, traversing each pixel of the detection image 200, calculating the correlation score between the region where the matching window is located in the detection image 200 and the template image 300 within the detection range 300S, which is used as the similarity score. The correlation score is negatively correlated with the variance or standard deviation of the grayscale values ​​of each pixel in the matching window and the template image 300 within the detection range 300S; obtaining the region where the matching window with the maximum correlation score is located as the matching region 200M, and each pixel in the template image 300 within the detection range 300S has a one-to-one correspondence with each pixel in the matching region 200M.

[0063] Specifically, the matching window slides across the detection image 200 in a preset direction, and after each slide, the correlation score between the template image 300 and the region where the current matching window is located is calculated, thus obtaining multiple correlation scores. The region where the matching window with the highest correlation score is located is selected as the matching region 200M that matches the template image 300. For example, the matching window can slide from the upper left corner of the detection image 200 to the right, with each slide being the size of one column of pixels. After sliding to the rightmost side, it slides down, with each downward slide being the size of one row of pixels. Then, it starts sliding from the leftmost side of the detection image 200 to the left, and so on, until the matching window has traversed every pixel of the detection image 200.

[0064] It should be noted that the greater the similarity, the smaller the variance or standard deviation of the grayscale of each pixel in the template image 300 within the matching window region and the detection range 300S. Therefore, the correlation score is negatively correlated with the variance and standard deviation of the grayscale of each pixel in the template image 300 within the matching window and the detection range 300S.

[0065] The correlation score between the detected image 200 and the template image 300 within the detection range 300S can be calculated using methods including Mean Absolute Differences (MAD), Sum of Absolute Differences (SAD), Sum of Squared Differences (SSD), Mean Square Differences (MSD), Normalized CrossCorrelation (NCC), Sequential Similarity Detection Algorithm (SSDA), or Hadamard Transform. As an example, the normalized crosscorrelation algorithm is used to calculate the correlation score.

[0066] Specifically, the formula for the normalized cross-correlation algorithm is:

[0067]

[0068] Where I1 is the template image 300, I2 is the detection image 200, Wp is the matching window, and the value of NCC is in the range of [-1, 1].

[0069] Continue to refer to Figure 5 and Figure 6In step S22, based on the preset positional relationship between the first region of interest 300R in the template image 300 and the detection range 300S, a second region of interest 200R is determined in the detection image 200. The second region of interest 200R is the region of interest of the target mark 100 in the detection image 200. The detection image 200 within the second region of interest 200R is designated as the image of interest 250. The relative positional relationship between the image of interest 250 and the matching image is the same as the preset positional relationship.

[0070] Subsequently, the image of interest 250 was used to measure the marker 100 to be tested.

[0071] In this embodiment, the target marker 100 includes a first marker 110 and a second marker 120. Therefore, in the detection image 200, the first marker 110 and the second marker 120 are processed to locate regions of interest (ROIs). Specifically, based on the first ROI 310R corresponding to the first marker 110, a second ROI 210R corresponding to the first marker 110 is determined in the detection image 200; and based on the first ROI 310R corresponding to the second marker 120, a second ROI 220R corresponding to the second marker 120 is determined in the detection image 200.

[0072] Specifically, the alignment error between the first mark 110 and the second mark 120 is then obtained using the image of interest 250 (i.e., image of interest 251) corresponding to the first mark 110 and the image of interest 250 (i.e., image of interest 252) corresponding to the second mark 120.

[0073] Based on the foregoing analysis, the template matching method improves the positioning accuracy of the target marker 100, which in turn improves the accuracy of the image of interest 250, and further improves the measurement accuracy of the alignment error between the first marker 110 and the second marker 120.

[0074] In this embodiment, determining the position of the second region of interest 200R in the detection image 200 based on the preset positional relationship between the first region of interest 300R in the template image 300 and the detection range 300S includes: obtaining the position of the matching center C' of the matching region; and obtaining the second region of interest 200R in the detection image 200 based on the preset positional relationship and the position of the matching center C', such that the relative positional relationship between the center of the second region of interest 200R and the matching center C' is the same as the preset positional relationship.

[0075] The matching region is obtained through template matching. The accuracy of the matching center C' of the matching region is relatively high, which helps to ensure that the second region of interest 200R has sufficient positional accuracy. This makes it easier to obtain the true center position of the image located in the second region of interest 200R in the detection image 200 more quickly and accurately by performing image processing on the image located in the second region of interest 200R.

[0076] As an example, obtaining the matching center position of the matching region includes: mapping the center of the detection range 300S onto the detection image 200 to obtain the matching center C' position of the matching region.

[0077] In other embodiments, the matching center position of the matching region can also be obtained based on the edge contour of the matching region. It should be noted that the edge contour here refers to the outline of the graphic within the matching region.

[0078] It should be noted that in this embodiment, the entire area of ​​the template image 300 is used as the detection range 300S, and the entire area of ​​the template image 300 is used to perform region of interest localization processing on the detection image 200.

[0079] In other embodiments, the detection range includes multiple sub-detection ranges, each of which is located in a local region of the template image, and any one of the sub-detection ranges covers a portion of the image of the marker to be tested; during the process of locating the region of interest in the detection image using the template image, the template image within each sub-detection range is used to perform the region of interest location processing on the detection image.

[0080] Accordingly, the region of interest localization processing of the detection image using template images within each sub-detection range includes: occluding the sub-detection range that does not currently need to be matched in the template image; matching the detection image with a first template image within the unoccluded sub-detection range; or, extracting the sub-detection range currently to be matched from the template image; and matching the detection image with the extracted first template image within the sub-detection range.

[0081] The detection range includes multiple sub-detection ranges, each located in a local area of ​​the template image. Each sub-detection range covers a portion of the image of the target marker. The region of interest is located in the detection image using the template image within each sub-detection range. Therefore, in this embodiment, by using a method of template matching by region, even if there are large positional deviations between different types of target markers or if the size of the target marker is deformed due to process issues, the position of the second region of interest can still be accurately determined in the detection image.

[0082] Continue to refer to Figure 6 and in conjunction with references Figure 7 After performing region of interest localization processing on the detection image 200 using the template image 300, the measurement method further includes: executing step S3, using the region of interest image 250 (i.e., region of interest image 251) corresponding to the first mark 110 and the region of interest image 250 (i.e., region of interest image 252) corresponding to the second mark 120 to obtain the alignment error between the first mark 110 and the second mark 120.

[0083] By acquiring the alignment error between the first mark 110 and the second mark 120, process monitoring is achieved, thereby determining process reliability. Since this method facilitates precise positioning of the second region of interest 200R in the detection image 200, the accuracy of the image of interest 250 used to measure the alignment error is high, correspondingly improving the measurement accuracy of the alignment error between the first mark 110 and the second mark 120.

[0084] In this embodiment, obtaining the alignment error includes: performing image processing on the image of interest 251 corresponding to the first mark 110 and the image of interest 252 corresponding to the second mark 120 respectively; obtaining the center C1 of the first mark 110 through the image of interest 251 corresponding to the first mark 110; and obtaining the center C2 of the second mark 120 through the image of interest 252 corresponding to the second mark 120; calculating the positional deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120 to obtain the alignment error between the first mark 110 and the second mark 120.

[0085] For example, the positional deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120 includes: the distance deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120 in a first direction; and the distance deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120 in a second direction.

[0086] Compared to directly using the matching center C' of the matching region as the final center of the first and second markers based on the matching results, the center C1 of the first marker 110 is obtained by image processing of the image of interest 251 corresponding to the first marker 110, and the center C2 of the second marker 120 is obtained by image processing of the image of interest 251 corresponding to the second marker 120. Therefore, the positional accuracy of the center C1 of the first marker 110 and the center C2 of the second marker 120 is higher.

[0087] In this embodiment, image processing is performed on the images of interest 250 corresponding to the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122, respectively.

[0088] In this embodiment, among the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122, the alignment pattern to be image processed is used as the target alignment pattern (not shown). The image processing includes: projecting the image of interest of the target alignment pattern along the projection direction of the image corresponding to the target alignment pattern, obtaining the correspondence between the positions of multiple points in the position arrangement direction and the projection values ​​as projection data, wherein the position arrangement direction is perpendicular to the projection direction, the projection value is the weighted value of the pixel grayscale of one or more points along the projection direction, and the projection direction is parallel to the first direction or the second direction; performing cross-correlation calculation on the projection data and the symmetry relationship, obtaining the time-shift variable with the largest cross-correlation value to obtain the optimized displacement, wherein the symmetry relationship has a symmetry axis perpendicular to the position arrangement direction, and the symmetry axis passes through the matching center C'; obtaining the center of the target alignment pattern along its arrangement direction based on the optimized displacement and the position of the symmetry axis.

[0089] Specifically, when the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122 are all bar-shaped patterns, the projection direction is parallel to the extension direction of the corresponding target alignment pattern.

[0090] By performing cross-correlation calculations on the projection data and the symmetry relationship, and obtaining the time-shift variable with the maximum cross-correlation value to obtain the optimized displacement, it is possible to determine the true center of the target alignment pattern along its arrangement direction through the optimized displacement.

[0091] Specifically, the symmetry relationship is the mirror image relationship of the projected data based on the axis of symmetry; obtaining the time-shifted variable with the maximum cross-correlation value to obtain the optimized displacement includes: selecting multiple positions to be fitted on both sides of the position corresponding to the maximum cross-correlation value from the multiple cross-correlation values ​​obtained by the cross-correlation calculation, and performing Gaussian fitting on the data points of the position corresponding to the maximum cross-correlation value and the positions to be fitted to obtain a fitting curve; obtaining the position corresponding to the peak of the fitting curve as the optimized displacement; correspondingly, obtaining the center of the target alignment pattern along its arrangement direction based on the optimized displacement and the position of the axis of symmetry includes: using formula X r =X0+X shift / 2 Obtain the center of the target alignment pattern along its arrangement direction, where X r X0 represents the center of the target alignment pattern along its arrangement direction, and X' represents the position of the matching center C'. shift This represents the optimized displacement.

[0092] By taking multiple sampling points on both sides of the position corresponding to the largest cross-correlation value and performing Gaussian fitting, a higher precision positional deviation can be obtained (the magnitude of the positional deviation can be less than the size of a pixel), thereby improving the accuracy of the center position of the final target alignment pattern along its arrangement direction.

[0093] In other words, the above image processing is performed on the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122 respectively, so as to obtain the center C1 of the first mark 110 in the second direction (e.g., using the image of interest 250 of the first alignment pattern 111). Figure 2 The position of the center C1 of the first mark 110 in the first direction (as shown in the Y direction) is obtained using the image of interest 250 of the second alignment pattern 112. Figure 2 The position of the center C2 of the second mark 120 in the second direction is obtained using the image of interest 250 of the third alignment pattern 121; the position of the center C2 of the second mark 120 in the first direction is obtained using the image of interest 250 of the fourth alignment pattern 122.

[0094] Figures 8 to 13 This is a schematic diagram corresponding to another embodiment of the method for measuring the alignment error of overlay marks.

[0095] The similarities between this embodiment and the previous embodiments will not be repeated here. The difference between this embodiment and the previous embodiments is that the template image 300 is used to perform region of interest localization processing on the images of the first and second marked objects respectively.

[0096] Reference Figure 8 and Figure 9 , Figure 8 This is a schematic diagram of the detected image in this embodiment. Figure 9 This is a schematic diagram of the template image in this embodiment. A detection image 600 and a template image 400 of the mark to be tested (not shown) are obtained. The template image 400 has a detection range (not marked), and the template image 400 includes a preset positional relationship of a first region of interest relative to the detection range. The first region of interest is the region of the region of interest of the mark to be tested in the template image 400.

[0097] In this embodiment, the marker to be tested includes a first marker (not shown) and a second marker (not shown), and the orthographic projection of the first marker onto the layer containing the second marker is located inside the second marker. Therefore, the detection image 600 includes an image 610 corresponding to the first marker and an image 620 corresponding to the second marker. For a detailed description of the marker to be tested and its detection image 600, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0098] like Figure 9 As shown, in this embodiment, the detection range includes a first sub-detection range 401S and a second sub-detection range 402S. The first sub-detection range 401S contains the image with the first mark, and the second sub-detection range 402S contains the image with the second mark. The first sub-detection range 401S and the second sub-detection range 402S respectively cover the first region of interest 300R (i.e., region of interest 310R) corresponding to the first mark and the first region of interest 300R (i.e., region of interest 320R) corresponding to the second mark.

[0099] Taking the example that the orthographic projection of the first mark on the layer where the second mark is located is located inside the second mark, the second sub-detection range 402S is a ring-shaped area, and the second sub-detection range 402S surrounds the first sub-detection range 401S.

[0100] For a detailed description of the template image 400, please refer to the corresponding description in the foregoing embodiments, which will not be repeated here.

[0101] Reference Figures 10 to 13 The template image 400 is used to perform region of interest localization processing on the detection image 600.

[0102] In this embodiment, the template image 400 within the first sub-detection range 401S and the template image 400 within the second sub-detection range 402S are used respectively to perform region of interest localization processing on the detection image 600. Accordingly, there are multiple matching regions.

[0103] Reference Figure 10 and Figure 11 The process of locating the region of interest in the detection image 600 using the template image 400 within the second sub-detection range 402S includes: matching the detection image 600 with the template image 400 within the second sub-detection range 402S, and obtaining the region in the detection image 600 that has the highest similarity to the template image 400 within the second sub-detection range 402S as the matching region 601M corresponding to the second sub-detection range 402S.

[0104] Specifically, matching the detected image 600 with the template image 400 within the second sub-detection range 402S includes: occluding the image information within the first sub-detection range 401S of the template image 400, with the unoccluded portion serving as the first template image 410 (e.g., ...). Figure 10 (as shown); the detected image 600 is matched with the first template image 410.

[0105] Accordingly, after the matching is completed, a matching region 601M that matches the first template image 410 is obtained in the detection image 600; after obtaining the matching region 601M, the matching center C3 of the matching region 601M is obtained.

[0106] Accordingly, based on the preset positional relationship and the position of the matching center C3, a second region of interest (unmarked) corresponding to the second marker is determined in the detection image 600.

[0107] It should be noted that the image information within the first sub-detection range 401S of the template image 400 is occluded. Therefore, when performing similarity calculation, the occluded pixels are not included in the calculation, thereby matching the detection image 600 with the template image 400 within the second sub-detection range 402S. Moreover, by using the occlusion method, the first template image 410 remains a complete image.

[0108] For a detailed description of the matching process, please refer to the corresponding description in the foregoing embodiments, which will not be repeated here.

[0109] Reference Figure 12 and Figure 13 The process of locating the region of interest in the detection image 600 using the template image 400 within the first sub-detection range 401S includes: matching the detection image 600 with the template image 400 within the first sub-detection range 401S, and obtaining the region in the detection image 600 that has the highest similarity to the template image 400 within the first sub-detection range 401S as the matching region 602M corresponding to the first sub-detection range 401S.

[0110] Specifically, matching the detected image 600 with the template image 400 within the first sub-detection range 401S includes: extracting the first sub-detection range 401S of the template image 400 as the second template image 420 (e.g., ...). Figure 12 (as shown); the detected image 600 is matched with the second template image 420.

[0111] In other embodiments, the image information within the second sub-detection range of the template image may be occluded, with the unoccluded portion serving as the second template image. Similarly, occluded pixels are not included in the similarity calculation.

[0112] Accordingly, after the matching is completed, a matching region 602M that matches the second template image 420 is obtained in the detection image 600. Figure 13 (as shown); after obtaining the matching region 602M, obtain the matching center C4 of the matching region 602M.

[0113] Accordingly, based on the preset positional relationship and the position of the matching center C4, a second region of interest (unmarked) corresponding to the first marker is determined in the detection image 600.

[0114] It should be noted that the matching process between the detection image 600 and the template image 400 within the first sub-detection range 401S further includes: before matching the detection image 600 with the second template image 420, selecting a search range (not shown in the figure) in the detection image 600, the search range covering the first marked image 610 (as shown in Figure 8), and the search range being located inside the second marked image 620 (as shown in Figure 8); and matching the detection image 600 within the search range with the second template image 420.

[0115] In the detected image 600, there may be a noisy image that is the same as the first marked image 610, and the noisy image is not an overprinted mark. Therefore, by first selecting a search range, matching can be performed within that search range, thereby reducing the probability of mismatch with other noisy images.

[0116] For a detailed description of the matching process and the subsequent acquisition of alignment error, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0117] In this embodiment, by utilizing the template image 400 within the first sub-detection range 401S and the template image 400 within the second sub-detection range 402S respectively, the position of the corresponding second region of interest can be determined during their respective matching processes, thereby further improving the positioning accuracy of the marker to be tested.

[0118] For example, even when the offset between the first and second marks is large (such as... Figure 8 (As shown), or, if the pattern size of either the first mark or the second mark changes, matching is performed on the image 610 corresponding to the first mark and the image 620 corresponding to the second mark in the detection image 600, respectively. That is, in the process of matching the image 610 corresponding to the first mark, only the image 610 corresponding to the first mark needs to be considered, and the influence of the image 620 corresponding to the second mark is small or non-existent. Similarly, in the process of matching the image 620 corresponding to the second mark, only the image 620 corresponding to the second mark needs to be considered, and the influence of the image 610 corresponding to the first mark is small or non-existent. In this way, the positions of the second regions of interest of the first mark and the second mark can be located respectively, and the applicability of the measurement method to various situations is improved.

[0119] Figure 14 This is a schematic diagram of template image 300 in another embodiment of the method for measuring the alignment error of overlay marks.

[0120] The similarities between this embodiment and the previous embodiments will not be repeated here. The difference between this embodiment and the previous embodiments is that: template image 500 is used to perform image matching processing on each marked graphic in the detection image.

[0121] By performing matching processing on the images of each marker graphic in the detection image, it is possible to perform matching processing on the image of any one marker graphic to reduce the influence of other marker graphics, thereby further improving the localization accuracy of the second region of interest of the marker graphic.

[0122] In this embodiment, in the template image 500, the detection range (not shown) includes multiple sub-detection ranges (not shown), each sub-detection range contains an image of the marked graphic and corresponds one-to-one with the image of the marked graphic, and each sub-detection range covers the region of interest of the corresponding marked graphic.

[0123] Specifically, the first alignment pattern and the second alignment pattern each have a corresponding sub-detection range 501S, and the third alignment pattern and the fourth alignment pattern each have a corresponding sub-detection range 502S.

[0124] It should be noted that, for ease of illustration, Figure 14 The diagram only illustrates a sub-detection range 501S corresponding to a first alignment pattern and a sub-detection range 502S corresponding to a third alignment pattern.

[0125] Accordingly, the region of interest localization processing is performed on the detection image using the template image 500 within each of the sub-detection ranges. Each region of interest localization processing includes: matching the detection image with the template image 500 within the sub-detection range, and obtaining the region in the detection image that has the highest similarity to the template image 500 within the sub-detection range as the matching region.

[0126] For example, if the first marker includes four alignment patterns and the second marker includes four alignment patterns, then eight region of interest localization processes will be performed accordingly.

[0127] In this embodiment, each region of interest localization process further includes: before matching the detected image with the template image 500 within the sub-detection range, selecting a search range in the detected image, wherein the search range is located at the position of the marker graphic to be matched and covers the image of the marker graphic to be matched; accordingly, matching the detected image within the search range with the template image within the sub-detection range.

[0128] By first selecting a search range, the image position of the marker graphic to be matched is defined in the detected image, so that matching can be performed within this search range, thereby reducing the probability of false matches with other similar images. For example, the first marker includes a pair of first aligned graphics extending along a first direction and arranged parallel to each other along a second direction. Since the first aligned graphics are relatively close to each other, when matching the image of one of the first aligned graphics, it is easy to make a false match with the image of the other first aligned graphics. Therefore, by first selecting a search range, other similar images are excluded from the search range, making it easier to achieve accurate matching.

[0129] For a detailed description of the matching process and the subsequent acquisition of alignment error, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0130] Accordingly, embodiments of the present invention also provide a measurement system. (See reference) Figure 12 The diagram shows a functional block diagram of an embodiment of the measurement system of the present invention.

[0131] Reference Figures 2 to 4 ,as well as Figures 6 to 7The measurement system is used to measure a mark 100 to be tested, which has a region of interest. The measurement system includes: an image acquisition module 10, used to acquire a detection image 200 and a template image of the mark 100, wherein the template image has a detection range and includes a preset positional relationship of a first region of interest relative to the detection range, the first region of interest being the region of interest of the mark in the template image; and a positioning module 20, used to perform region of interest localization processing on the detection image 200 using the template image 300, the positioning module 20 including: a matching unit 21, used to match the detection image 200 with the template image 300 within the detection range 300s. In the detection image 200, a matching process is performed to obtain the region with the highest similarity to the template image 300 within the detection range 300S as the matching region 200M, and the detection image of the matching region is used as the matching image. The region of interest determination unit 22 is used to determine a second region of interest 300R in the detection image according to a preset positional relationship between the first region of interest 300R in the template image 300 and the detection range. The second region of interest is the region of the marked region of interest in the detection image, and the detection image 200 within the second region of interest 300R is used as the image of interest 250. The relative positional relationship between the image of interest and the matching image is the same as the preset positional relationship.

[0132] In the measurement system, the template image 300 is first used to perform region of interest (ROI) localization processing on the detection image 200 of the mark 100 to be tested. By matching the template image 300 on the detection image 200, a matching region 200M is obtained. After the matching is completed, the corresponding second region of interest 200R in the detection image 200 can be determined according to the preset positional relationship between the first region of interest in the template image 300 and the detection range 300S, thereby obtaining the image of interest 250. Compared with the mechanical positioning scheme of the mark to be tested, this embodiment uses template matching, which not only helps to locate the position of the mark 100 to be tested in the detection image 200 more quickly, but also helps to determine the position of the second region of interest 200R in the detection image 200 more accurately, thereby improving the positioning accuracy of the mark 100 to be tested. Correspondingly, the accuracy of the image of interest 250 is improved, and the measurement accuracy when measuring the mark to be tested is improved.

[0133] In this embodiment, the marker to be tested 100 includes a first marker 110 and a second marker 120. Correspondingly, the first marker 110 and the second marker 120 each have their own corresponding regions of interest (not shown).

[0134] Specifically, the mark to be tested 100 is an overlay mark, so that the alignment error of the overlay mark 100 can be obtained by detecting the alignment error between the first mark 110 and the second mark 120.

[0135] Specifically, the overlay mark 100 is located in the substrate (not shown), and the first mark 110 and the second mark 120 are located at different heights on the substrate, or the first mark 110 and the second mark 120 are formed in different processes. As an example, the substrate can be a wafer or a chip.

[0136] As an example, the orthographic projection of the first mark 110 onto the layer containing the second mark 120 is located inside the second mark 120.

[0137] In this embodiment, each of the test marks 100 includes multiple mark patterns 150. Specifically, the mark pattern 150 corresponding to the first mark 110 is the inner alignment pattern (i.e., inner bar) in the overprinted mark 100, and the mark pattern 150 corresponding to the second mark 120 is the outer alignment pattern (i.e., outer bar) in the overprinted mark 100.

[0138] In this embodiment, the first mark 110 includes along a first direction (e.g., Figure 2 Extending along the X direction (as shown in the middle) and along the second direction (as shown in the middle X direction) Figure 2 The first alignment pattern 110 comprises a pair of first alignment patterns 111 arranged in parallel along the Y direction, and a pair of second alignment patterns 112 extending along the second direction and arranged in parallel along the first direction; the second alignment pattern 120 comprises a pair of third alignment patterns 121 extending along the first direction and arranged in parallel along the second direction, and a pair of fourth alignment patterns 122 extending along the second direction and arranged in parallel along the first direction. As an example, the first alignment pattern 110 is centrally symmetrical, and the second alignment pattern 120 is centrally symmetrical.

[0139] The template image 300 is a high-quality image that meets the design requirements, so as to use the template image 300 as a reference image to determine the position of the mark to be tested 100 in the detection image 200.

[0140] It should be noted that the size of the detection image 200 may be larger than the size of the area where the test mark 100 is located. For example, the detection image 200 may also contain background information, while the size of the template image 300 is the same as the size of the area where the test mark 100 is located. Thus, the position of the test mark 100 can be searched from the detection image 200 through the template image 300.

[0141] As an example, template image 300 is a design image corresponding to the mark to be tested 100, such as a binarized image or design layout. In other embodiments, the template image may also be an image of a standard mark that is identical to the mark to be tested and conforms to design requirements (e.g., in the case that the mark to be tested is an overlay mark, the standard mark conforms to alignment requirements). The template image may also be a simulated image obtained through a theoretical model.

[0142] Therefore, in this embodiment, the template image 300 includes an image 310 corresponding to the first marker 110 and an image 320 corresponding to the second marker 120. Specifically, the template image 300 defines a detection range 300S, and the template image 300 has a preset positional relationship between a first region of interest 300R and the detection range 300S. The first region of interest 300R is the region of interest of the marker to be tested 100 in the template image 300.

[0143] In one specific embodiment, the entire region of the template image 300 is used as the detection range 300S. That is, the entire region of the template image 300 is subsequently matched with the detection image 200. Accordingly, the pixels in the entire region of the template image 300 are used to participate in the calculation of the matching degree in order to statistically analyze the overall similarity of the template images 300.

[0144] In this embodiment, the center C of the detection range 300S coincides with the center of the test mark 100 to be processed for region of interest localization; the preset positional relationship of the first region of interest 300R relative to the detection range 300S is the positional relationship between the center of each first region of interest 300R and the center C of the detection range 300S.

[0145] The positioning module 20 determines the position of the image of the marker 100 to be tested in the detection image 200.

[0146] Reference Figure 5 and Figure 6 By matching the template image 300 on the detection image 200, a matching region 200M is obtained (e.g., ...). Figure 6 As shown, this allows for the determination of a second region of interest 200R in the detection image 200 based on the preset positional relationship between the first region of interest 300R in the template image 300 and the detection range 300S after matching is completed. Template matching not only facilitates faster localization of the target marker 100 in the detection image 200 but also allows for more precise determination of the second region of interest 200R in the detection image 200, thereby improving the localization accuracy of the target marker 100.

[0147] As an example, the matching unit 21 includes: a correlation score acquisition subunit, used to traverse each pixel of the detection image 200 using the same matching window as the detection range 300S, and calculate the correlation score between the region where the matching window is located in the detection image 200 and the template image 300 within the detection range 300S, which is used as a similarity score. The correlation score is negatively correlated with the variance or standard deviation of the grayscale of each pixel of the matching window and the template image 300 within the detection range 300S; and a filtering subunit, used to acquire the region where the matching window with the maximum correlation score is located, as the matching region 200M, and each pixel of the template image 300 within the detection range 300S has a one-to-one correspondence with each pixel of the matching region 200M.

[0148] Specifically, the correlation score acquisition subunit slides the matching window in the detection image 200 according to a preset sliding direction, and calculates the correlation score between the template image 300 and the area where the current matching window is located after each slide, thereby obtaining multiple correlation scores; the filtering subunit selects the area where the matching window corresponding to the maximum correlation score is located as the matching area 200M that matches the template image 300.

[0149] The correlation score acquisition subunit calculates the correlation score between the detected image 200 and the template image 300 within a detection range of 300 seconds using the mean absolute difference algorithm, the sum of absolute errors algorithm, the sum of squared errors algorithm, the mean sum of squared errors algorithm, the normalized cross-correlation algorithm, the sequential similarity detection algorithm, or the Hadamard transform algorithm. As an example, the normalized cross-correlation algorithm is used to calculate the correlation score.

[0150] Specifically, the formula for the normalized cross-correlation algorithm is:

[0151]

[0152] Where I1 is the template image 300, I2 is the detection image 200, Wp is the matching window, and the value of NCC is in the range of [-1, 1].

[0153] Continue to refer to Figure 5 and Figure 6 The region of interest determination unit 22 is used to determine a second region of interest 300R in the detection image 200 based on the preset positional relationship of the first region of interest 300R in the template image 300 relative to the detection range 300S. The second region of interest 300R is the region of the region of interest of the target mark 100 in the detection image 200. The detection image 200 within the second region of interest 200R is used as the image of interest 250. The relative positional relationship between the image of interest 250 and the matching image is the same as the preset positional relationship.

[0154] Subsequently, the test mark 100 is measured using the image of interest 250. In this embodiment, the test mark 100 includes a first mark 110 and a second mark 120. Therefore, the region of interest determination unit 22 performs region of interest localization processing on the first mark 110 and the second mark 120. Specifically, based on the first region of interest 310R corresponding to the first mark 110, a second region of interest 210R corresponding to the first mark 110 is determined in the detection image 200; based on the first region of interest 310R corresponding to the second mark 120, a second region of interest 220R corresponding to the second mark 120 is determined in the detection image 200.

[0155] Specifically, the alignment error between the first mark 110 and the second mark 120 is then obtained by using the image of interest 250 (i.e., image of interest 251) corresponding to the first mark 110 and the image of interest 250 (i.e., image of interest 252) corresponding to the second mark 120.

[0156] Based on the foregoing analysis, the template matching method improves the positioning accuracy of the target marker 100, which in turn improves the accuracy of the image of interest 250, and further improves the measurement accuracy of the alignment error between the first marker 110 and the second marker 120.

[0157] Specifically, the region of interest determination unit 22 includes: a matching center acquisition unit, used to acquire the matching center C' of the matching region 200M; and a region of interest determination subunit, used to acquire a second region of interest 200R in the detection image 200 according to the preset positional relationship and the position of the matching center C', such that the relative positional relationship between the center of the second region of interest 200R and the position of the matching center C' is the same as the preset positional relationship.

[0158] The matching region is obtained through template matching. The accuracy of the matching center C' of the matching region is relatively high, which helps to ensure that the second region of interest 200R has sufficient positional accuracy. This makes it easier to obtain the true center position of the image located in the second region of interest 200R in the detection image 200 more quickly and accurately by performing image processing on the image located in the second region of interest 200R.

[0159] As an example, the matching center acquisition unit maps the center of the detection range 300S to the detection image 200 to obtain the position of the matching center C' of the matching region.

[0160] In other embodiments, the matching center acquisition unit may also acquire the matching center position of the matching region based on the edge contour of the matching region. It should be noted that the edge contour here refers to the outline of the graphic in the matching region.

[0161] It should be noted that in this embodiment, the positioning module 20 uses the entire area of ​​the template image 300 as the detection range 300S, and uses the entire area of ​​the template image 300 to perform region of interest localization processing on the detection image 200.

[0162] In other embodiments, the detection range includes multiple sub-detection ranges, each of which is located in a local area of ​​the template image, and any one of the sub-detection ranges covers a portion of the image of the marker to be tested; the localization module is used to perform region of interest localization processing on the detection image using the template image within each sub-detection range.

[0163] Accordingly, the matching unit is used to occlude the sub-detection range that does not need to be matched in the template image, and to match the detected image with the first template image in the unoccluded sub-detection range; or, the matching unit is used to extract the sub-detection range to be matched in the template image, and to match the detected image with the first template image in the extracted sub-detection range.

[0164] In this embodiment, the measurement system further includes an error acquisition module 30, which is used to acquire the alignment error between the first mark 110 and the second mark 120 using the image of interest 250 (i.e., image of interest 251) corresponding to the first mark 110 and the image of interest 250 (i.e., image of interest 252) corresponding to the second mark 120.

[0165] By acquiring the alignment error between the first mark 110 and the second mark 120, process monitoring is achieved, thereby determining process reliability. Specifically, because this measurement system facilitates precise positioning of the second region of interest 200R in the detection image 200, the accuracy of the image of interest 250 used to measure the alignment error is high, correspondingly improving the measurement accuracy of the alignment error between the first mark 110 and the second mark 120.

[0166] In this embodiment, the error acquisition module 30 includes: an image processing unit, used to perform image processing on the image of interest 251 corresponding to the first mark 110 and the image of interest 252 corresponding to the second mark 120, respectively, to obtain the center C1 of the first mark 110 through the image of interest 251 corresponding to the first mark 110, and to obtain the center C2 of the second mark 120 through the image of interest 252 corresponding to the second mark 120; and a deviation calculation unit, used to calculate the positional deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120, and to obtain the alignment error between the first mark 110 and the second mark 120.

[0167] For example, the positional deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120 includes: the distance deviation between the center C1 of the first mark 110 and the center C2 of the second mark 120 in a first direction and the distance deviation in a second direction.

[0168] Compared to directly using the matching center C' of the matching region as the final center of the first and second markers based on the matching results, the center C1 of the first marker 110 is obtained by image processing of the image of interest 251 corresponding to the first marker 110, and the center C2 of the second marker 120 is obtained by image processing of the image of interest 251 corresponding to the second marker 120. Therefore, the positional accuracy of the center C1 of the first marker 110 and the center C2 of the second marker 120 is higher.

[0169] In this embodiment, the image processing unit is used to perform image processing on the images of interest 250 corresponding to the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121 and the fourth alignment pattern 122, respectively.

[0170] Specifically, among the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122, the alignment pattern to be processed is used as the target alignment pattern (not shown). The image processing unit includes: a first projection subunit, which projects the image of interest of the target alignment pattern along the projection direction of the image corresponding to the target alignment pattern, and obtains the correspondence between the positions of multiple points in the position arrangement direction and the projection values ​​as projection data. The position arrangement direction is perpendicular to the projection direction, and the projection value is a weighted value of the pixel grayscale of one or more points along the projection direction. The projection direction is parallel to a first direction or a second direction; a first processing subunit, which performs cross-correlation calculation on the projection data and the symmetry relationship, and obtains the time-shift variable with the maximum cross-correlation value to obtain an optimized displacement. The symmetry relationship has a symmetry axis perpendicular to the position arrangement direction, and the symmetry axis passes through the matching center C'; and a second processing subunit, which obtains the center of the target alignment pattern along its arrangement direction based on the optimized displacement and the position of the symmetry axis.

[0171] Specifically, when the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122 are all bar-shaped patterns, the projection direction is parallel to the extension direction of the corresponding target alignment pattern.

[0172] By performing cross-correlation calculations on the projection data and the symmetry relationship, and obtaining the time-shift variable with the maximum cross-correlation value to obtain the optimized displacement, it is possible to determine the true center of the target alignment pattern along its arrangement direction through the optimized displacement.

[0173] Specifically, the symmetry relationship is the mirror image relationship of the projected data based on the axis of symmetry; the first processing subunit is used to select multiple positions to be fitted on both sides of the position corresponding to the largest cross-correlation value among the multiple cross-correlation values ​​obtained by the cross-correlation calculation, and to perform Gaussian fitting on the data points of the position corresponding to the largest cross-correlation value and the positions to be fitted to obtain a fitting curve; correspondingly, the second processing subunit is used to use formula X r =X0+X shift / 2 Obtain the center of the target alignment pattern along its arrangement direction, where X r X0 represents the center of the target alignment pattern along its arrangement direction, and X' represents the position of the matching center C'. shift This represents the optimized displacement.

[0174] In other words, the image processing unit performs the aforementioned image processing on the first alignment pattern 111, the second alignment pattern 112, the third alignment pattern 121, and the fourth alignment pattern 122 respectively, thereby obtaining the center C1 of the first mark 110 in the second direction (e.g., using the image of interest 250 of the first alignment pattern 111). Figure 2 The position of the first mark 110 (as shown in the Y direction) is obtained using the image of interest 250 of the second alignment pattern 112 in the first direction (as shown in the Y direction); Figure 2 The position of the center C2 of the second mark 120 in the second direction is obtained using the image of interest 250 of the third alignment pattern 121; the position of the center C2 of the second mark 120 in the first direction is obtained using the image of interest 250 of the fourth alignment pattern 122.

[0175] It should be noted that in this embodiment, the entire area of ​​the template image 300 is used as the detection range 300S for illustration.

[0176] Reference Figures 8 to 13 In another embodiment, the positioning module may also use the template image 300 to perform region of interest positioning processing on the first marked image and the second marked image respectively.

[0177] In this embodiment, the detection range of the template image 300 includes a first sub-detection range 401S and a second sub-detection range 402S. The first sub-detection range 401S contains an image with a first mark, and the second sub-detection range 402S contains an image with a second mark. The first sub-detection range 401S and the second sub-detection range 402S respectively cover the first region of interest 300R (i.e., region of interest 310R) corresponding to the first mark and the first region of interest 300R (i.e., region of interest 320R) corresponding to the second mark.

[0178] Taking the example that the orthographic projection of the first mark on the layer where the second mark is located is inside the second mark, the second sub-detection range 402S is a ring-shaped area, and the second sub-detection range 402S surrounds the first sub-detection range 401S.

[0179] Accordingly, the localization module 20 uses the template image 400 within the first sub-detection range 401S and the template image 400 within the second sub-detection range 402S to perform region of interest localization processing on the detection image 600.

[0180] Reference Figure 10 and Figure 11 In this embodiment, the matching unit includes a fourth matching subunit, which is used to perform matching processing on the detection image 600 and the template image 400 within the second sub-detection range 402S, and to obtain the region in the detection image 600 that has the highest similarity to the template image 400 within the second sub-detection range 402S as the matching region 601M.

[0181] Specifically, the fourth matching subunit occludes the image information within the first sub-detection range 401S of the template image 400, and the unoccluded portion is used as the first template image 410 (e.g., ...). Figure 10 As shown in the figure, the detected image 600 is matched with the first template image 410.

[0182] After matching is completed, a matching region 601M matching the first template image 410 is obtained in the detection image 600; after obtaining the matching region 601M, the matching center acquisition unit obtains the matching center C3 of the matching region 601M. Accordingly, the region of interest determination subunit determines the second region of interest (unmarked) corresponding to the second marker in the detection image 600 according to the preset positional relationship and the position of the matching center C3.

[0183] Reference Figure 12 and Figure 13 In this embodiment, the matching unit further includes a fifth matching subunit, which is used to perform matching processing on the detection image 600 and the template image 400 within the first sub-detection range 401S, and to obtain the region in the detection image 600 that has the highest similarity to the template image 400 within the first sub-detection range 401S as the matching region 602M.

[0184] Specifically, the fifth matching subunit extracts the first sub-detection range 401S of the template image 400 as the second template image 420 (e.g., ...). Figure 12As shown in the figure, the detected image 600 is matched with the second template image 420. In other embodiments, the fifth matching subunit may also occlude the image information within the second sub-detection range of the template image, using the unoccluded portion as the second template image. Similarly, occluded pixels are not included in the similarity calculation.

[0185] After matching is completed, a matching region 602M that matches the second template image 420 is obtained in the detection image 600 (e.g., ...). Figure 13 (As shown); After obtaining the matching region 602M, the matching center acquisition unit obtains the matching center C4 of the matching region 602M. Accordingly, the region of interest determination subunit determines the second region of interest (unmarked) corresponding to the first marker in the detection image 600 according to the preset positional relationship and the position of the matching center C4.

[0186] It should be noted that, in this embodiment, the matching unit further includes: a search range setting subunit, used to select a search range (not shown) in the detection image 600 before performing the matching process, the search range covering the first marked image 610 (as shown in 8), and the search range being located inside the second marked image 620 (as shown in 8); correspondingly, the fifth matching subunit performs matching processing on the detection image 600 within the search range and the second template image 420.

[0187] In the detection image 600, there may be a noisy image that is the same as the first marked image 610, and the noisy image is not an overprinted mark. Therefore, by first selecting a search range so as to perform matching within the search range, the probability of mismatch with other noisy images is reduced.

[0188] In this embodiment, by utilizing the template image 400 within the first sub-detection range 401S and the template image 400 within the second sub-detection range 402S respectively, the position of the corresponding second region of interest can be determined during their respective matching processes, thereby further improving the positioning accuracy of the marker to be tested.

[0189] Reference Figure 14 In another embodiment, the positioning module may also use template image 500 to perform matching processing on the images of each marked graphic in the detection image.

[0190] By performing matching processing on the images of each marker graphic in the detection image, it is possible to perform matching processing on the image of any one marker graphic to reduce the influence of other marker graphics, thereby further improving the localization accuracy of the second region of interest of the marker graphic.

[0191] In this embodiment, in the template image 500, the detection range (not shown) includes multiple sub-detection ranges (not shown), each sub-detection range contains an image with a marked graphic and corresponds one-to-one with the image of the marked graphic. The sub-detection range covers the region of interest of the corresponding marked graphic.

[0192] Specifically, the first and second aligned patterns each have a corresponding sub-detection range 501S, and the third and fourth aligned patterns each have a corresponding sub-detection range 502S. It should be noted that, for ease of illustration, Figure 14 The diagram only illustrates a sub-detection range 501S corresponding to a first alignment pattern and a sub-detection range 502S corresponding to a third alignment pattern.

[0193] Accordingly, the localization module uses the template images 500 within each sub-detection range to perform region of interest (ROI) localization processing on the detection image. In each ROI localization process, the matching unit 21 matches the detection image with the template images 500 within the sub-detection range, identifying the region in the detection image with the highest similarity to the template images 500 within the sub-detection range as the matching region. For example, if the first marker includes four alignment patterns and the second marker includes four alignment patterns, then eight ROI localization processes are performed accordingly.

[0194] In this embodiment, the matching unit includes: a search range setting subunit, used to select a search range in the detection image, the search range being located at the position of the marker graphic to be matched and encompassing the image of the marker graphic to be matched; and a sixth matching subunit, used to perform matching processing between the detection image within the search range and the template image within the sub-detection range. By first selecting the search range, the image position of the marker graphic to be matched is defined in the detection image, so that matching can be performed within the search range, thereby reducing the probability of mismatches with other similar images.

[0195] This invention also provides a device that can implement the measurement method provided in this invention by loading a program in the form of the above-described measurement method.

[0196] refer to Figure 9 The diagram illustrates the hardware structure of a device according to an embodiment of the present invention. The device in this embodiment includes: at least one processor 01, at least one communication interface 02, at least one memory 03, and at least one communication bus 04.

[0197] In this embodiment, the number of processor 01, communication interface 02, memory 03 and communication bus 04 is at least one, and the processor 01, communication interface 02 and memory 03 communicate with each other through the communication bus 04.

[0198] The communication interface 02 can be an interface of a communication module used for network communication, such as the interface of a GSM module.

[0199] The processor 01 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the measurement method described in this embodiment.

[0200] The memory 03 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0201] The memory 03 stores one or more computer instructions, which are executed by the processor 01 to implement the measurement method provided in the foregoing embodiments.

[0202] It should be noted that the aforementioned terminal device may also include other devices (not shown) that may not be essential to understanding the content disclosed in the embodiments of the present invention; given that these other devices may not be essential for understanding the content disclosed in the embodiments of the present invention, the embodiments of the present invention will not describe them one by one.

[0203] This invention also provides a storage medium storing one or more computer instructions for implementing the measurement method provided in the foregoing embodiments.

[0204] In the measurement method of this invention, a template image is first used to perform region of interest (ROI) localization processing on the detection image of the marker. By matching the template image on the detection image, a matching region is obtained. After the matching is completed, a corresponding second ROI can be determined in the detection image based on the preset positional relationship between the first ROI in the template image and the detection range. The second ROI is the region of the marker's ROI in the detection image, thereby obtaining an image of interest. Compared with the mechanical positioning of the marker, this invention, through template matching, not only facilitates faster positioning of the marker in the detection image but also facilitates more accurate determination of the position of the second ROI in the detection image, thereby improving the positioning accuracy of the marker. Consequently, the accuracy of the image of interest is improved, and the measurement accuracy when measuring the marker subsequently is enhanced.

[0205] The embodiments of the present invention described above are combinations of elements and features of the present invention. Unless otherwise stated, the elements or features described are optional. Individual elements or features may be practiced without combination with other elements or features. Furthermore, embodiments of the present invention may be constructed by combining some elements and / or features. The order of operations described in the embodiments of the present invention may be rearranged. Some constructions of any embodiment may be included in another embodiment and may be replaced by corresponding constructions of another embodiment. It will be apparent to those skilled in the art that claims in the appended claims that are not expressly referenced to each other may be combined to form embodiments of the present invention, or may be included as new claims in amendments made after the filing of this application.

[0206] Embodiments of the present invention can be implemented by various means, such as hardware, firmware, software, or combinations thereof. In a hardware configuration, the method according to an exemplary embodiment of the present invention can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, etc.

[0207] In firmware or software configuration, embodiments of the present invention can be implemented in the form of modules, processes, functions, etc. Software code can be stored in a memory unit and executed by a processor. The memory unit is located inside or outside the processor and can send data to and receive data from the processor via various known means.

[0208] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is accorded the widest scope consistent with the principles and novel features disclosed herein.

[0209] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A measurement method, characterized in that, The method is used to measure a mark to be tested, wherein the mark to be tested is an overlay mark, the mark to be tested includes a first mark and a second mark, and the first mark and the second mark each have their corresponding regions of interest, the mark to be tested has a region of interest, and the measurement method includes: Acquire a detection image and a template image of the target marker. The template image has a detection range and includes a first region of interest with respect to a preset positional relationship relative to the detection range. The first region of interest is the region of interest of the target marker in the template image. Using the template image, the detection image is processed to obtain the region of interest corresponding to the first marker and the region of interest corresponding to the second marker. The region of interest localization process includes: The detected image is matched with the template image within the detection range. In the detected image, the region with the highest similarity to the template image within the detection range is obtained as the matching region, and the detected image of the matching region is used as the matching image. Based on the preset positional relationship between the first region of interest in the template image and the detection range, a second region of interest is determined in the detection image. The second region of interest is the region of interest of the target marker in the detection image. The detection image within the second region of interest is used as the image of interest. The relative positional relationship between the image of interest and the matching image is the same as the preset positional relationship. In the detected image, the region of interest (ROI) of the first marker and the second marker are respectively obtained by performing region of interest localization processing on the first marker and the second marker. After using the template image to perform region of interest localization processing on the detection image to obtain the region of interest corresponding to the first marker and the region of interest corresponding to the second marker, the alignment error between the first marker and the second marker is obtained using the region of interest corresponding to the first marker and the region of interest corresponding to the second marker.

2. The measurement method as described in claim 1, characterized in that, The entire region of the template image is used as the detection range; during the process of locating the region of interest in the detection image using the template image, the entire region of the template image is used to perform the region of interest location processing on the detection image; Alternatively, the detection range may include multiple sub-detection ranges, each of which is located in a local region of the template image, and any one of the sub-detection ranges may cover a portion of the image of the marker to be tested; During the process of locating the region of interest in the detection image using the template image, the template images within each of the sub-detection ranges are used to perform the region of interest localization process on the detection image.

3. The measurement method as described in claim 2, characterized in that, The process of locating the region of interest in the detection image using template images within each of the sub-detection ranges includes: In the template image, the sub-detection range that does not currently need to be matched is occluded; the detection image is then matched with the first template image within the unoccluded sub-detection range. Alternatively, the sub-detection range to be matched is extracted from the template image; the detection image is then matched with the first template image within the extracted sub-detection range.

4. The measurement method as described in claim 1, characterized in that, The marker to be tested includes a first marker and a second marker, and the first marker and the second marker each have their own corresponding regions of interest; the number of matching regions is multiple; The detection range includes a first sub-detection range and a second sub-detection range. The first sub-detection range contains the image with the first mark, and the second sub-detection range contains the image with the second mark. The first sub-detection range and the second sub-detection range respectively cover the first region of interest corresponding to the first mark and the first region of interest corresponding to the second mark. During the process of locating the region of interest (ROI) of the detected image using the template image, the template images within the first sub-detection range and the template images within the second sub-detection range are used respectively to perform ROI localization processing on the detected image, wherein... The process of locating the region of interest (ROI) of the detected image using a template image within the first sub-detection range includes: matching the detected image with a template image within the first sub-detection range, and then matching the detected image with the template image within the first sub-detection range. The region with the highest similarity to the template image within the first sub-detection range is selected as the region with the highest similarity to the template image within the first sub-detection range. The matching area corresponding to the measurement range; The process of locating the region of interest in the detection image using the template image within the second sub-detection range includes: matching the detection image with the template image within the second sub-detection range, and obtaining the region in the detection image that has the highest similarity to the template image within the second sub-detection range as the matching region corresponding to the second sub-detection range.

5. The measurement method as described in claim 4, characterized in that, The first mark and the second mark are formed by different processes, or the first mark and the second mark are located on different layers, and the orthographic projection of the first mark on the layer where the second mark is located is located inside the second mark; The matching process between the detected image and the template image within the second sub-detection range includes: occluding the image information within the first sub-detection range of the template image, with the unoccluded portion serving as the first template image; and matching the detected image with the first template image. The matching process between the detected image and the template image within the first sub-detection range includes: extracting the first sub-detection range of the template image as the second template image, or occluding the image information within the second sub-detection range of the template image, with the unoccluded portion serving as the second template image; and matching the detected image with the second template image.

6. The measurement method as described in claim 5, characterized in that, The matching process between the detected image and the template image within the first sub-detection range further includes: before matching the detected image with the second template image, selecting a search range in the detected image, wherein the search range covers the image of the first marker and is located inside the image of the second marker; The matching process between the detected image and the second template image includes: matching the detected images within the search range with the second template image.

7. The measurement method as described in claim 1, characterized in that, The marker to be tested includes multiple marker graphics located on the same layer; The detection range includes multiple sub-detection ranges, each sub-detection range containing an image of the marked graphic and corresponding one-to-one with the image of the marked graphic. Each sub-detection range covers the region of interest of the corresponding marked graphic. During the process of locating the region of interest (ROI) of the detection image using the template image, the ROI of the detection image is located using the template images within each sub-detection range. Each ROI location process includes: matching the detection image with the template images within the sub-detection range, and obtaining the region in the detection image that has the highest similarity to the template images within the sub-detection range as the matching region.

8. The measurement method as described in claim 7, characterized in that, Each region of interest localization process further includes: before matching the detected image with the template image within the sub-detection range, selecting a search range in the detected image, wherein the search range is located at the position of the marker graphic to be matched and covers the image of the marker graphic to be matched; Matching the detected image with the template image within the sub-detection range includes: matching the detected image within the search range with the template image within the sub-detection range.

9. The measurement method as described in claim 1 or 4, characterized in that, The center of the detection range coincides with the center of the target marker to be localized (the region of interest is to be defined); the first region of interest is relative to the detection range. The preset positional relationship is the positional relationship between the centers of each of the first regions of interest relative to the center of the detection range; Determining the second region of interest in the detection image based on the preset positional relationship between the first region of interest in the template image and the detection range includes: obtaining the matching center position of the matching region; Based on the preset positional relationship and the matching center position, a second region of interest is obtained in the detected image, such that the relative positional relationship between the center of the second region of interest and the matching center position is the same as the preset positional relationship.

10. The measurement method as described in claim 9, characterized in that, Obtaining the matching center position of the matching region includes: mapping the center of the detection range onto the detection image to obtain the matching center position of the matching region; or, obtaining the matching center position of the matching region based on the edge contour of the matching region.

11. The measurement method as described in claim 1, characterized in that, The overlay mark is located in the substrate, and the first mark and the second mark are located at different heights on the substrate, or the first mark and the second mark are formed in different processes.

12. The measurement method as described in claim 11, characterized in that, Using the image of interest corresponding to the first marker and the image of interest corresponding to the second marker, the alignment error between the first marker and the second marker is obtained, including: Image processing is performed on the image of interest corresponding to the first marker and the image of interest corresponding to the second marker respectively. The center position of the first marker is obtained through the image of interest corresponding to the first marker, and the center position of the second marker is obtained through the image of interest corresponding to the second marker. Calculate the positional deviation between the center of the first mark and the center of the second mark to obtain the alignment error between the first mark and the second mark.

13. The measurement method as described in claim 12, characterized in that, The first mark includes a pair of first alignment patterns extending along a first direction and arranged in parallel along a second direction, and a pair of second alignment patterns extending along the second direction and arranged in parallel along the first direction; the second mark includes a pair of third alignment patterns extending along the first direction and arranged in parallel along the second direction, and a pair of fourth alignment patterns extending along the second direction and arranged in parallel along the first direction, wherein the first direction is perpendicular to the second direction. Image processing is performed on the images of interest corresponding to the first alignment pattern, the second alignment pattern, the third alignment pattern, and the fourth alignment pattern, respectively. The image processing includes: obtaining the position of the center of the first mark in the second direction using the image of interest of the first alignment pattern; obtaining the position of the center of the first mark in the first direction using the image of interest of the second alignment pattern; obtaining the position of the center of the second mark in the second direction using the image of interest of the third alignment pattern; and obtaining the position of the center of the second mark in the first direction using the image of interest of the fourth alignment pattern.

14. The measurement method as described in claim 13, characterized in that, Based on the preset positional relationship between the first region of interest in the template image and the detection range, the matching center of the matching region is obtained during the process of determining the second region of interest in the detection image; Among the first alignment pattern, the second alignment pattern, the third alignment pattern, and the fourth alignment pattern, the one to be performed at the moment is... The image processing includes: (1) aligning the image as the target alignment image; (2) image processing including: Projecting the image of interest of the target alignment pattern along the projection direction of the image corresponding to the target alignment pattern, obtaining the correspondence between the positions of multiple points in the position arrangement direction and the projection values ​​as projection data, wherein the position arrangement direction is perpendicular to the projection direction, the projection value is the weighted value of the pixel grayscale of one or more points along the projection direction, and the projection direction is parallel to the first direction or the second direction; Cross-correlation calculations are performed on the projection data and the symmetry relationship to obtain the time-shift variable with the maximum cross-correlation value to obtain the optimized displacement. The symmetry relationship has a symmetry axis perpendicular to the position arrangement direction, and the symmetry axis passes through the matching center. The center of the target alignment pattern along its arrangement direction is obtained based on the optimized displacement and the position of the symmetry axis.

15. The measurement method as described in claim 14, characterized in that, The symmetry relationship is the mirror relationship of the projected data based on the axis of symmetry; The process of obtaining the time-shifted variable with the largest cross-correlation value to obtain the optimized displacement includes: selecting multiple positions to be fitted on both sides of the position corresponding to the largest cross-correlation value from the multiple cross-correlation values ​​obtained by the cross-correlation calculation; performing Gaussian fitting on the data points of the position corresponding to the largest cross-correlation value and the positions to be fitted to obtain a fitting curve; and obtaining the position corresponding to the peak of the fitting curve as the optimized displacement. Obtaining the center of the target alignment pattern along its arrangement direction based on the optimized displacement and the position of the axis of symmetry includes: using the formula Xr = X0 + X shift / 2 to obtain the center of the target alignment pattern along its arrangement direction, where Xr represents the center of the target alignment pattern along its arrangement direction, X0 represents the position of the matching center, and X shift represents the optimized displacement.

16. The measurement method as described in claim 1, characterized in that, The matching process includes: using a matching window that is the same as the detection range, traversing each pixel of the detection image, calculating the correlation score between the region where the matching window is located in the detection image and the template image within the detection range, as the similarity, wherein the correlation score is negatively correlated with the variance or standard deviation of the gray levels of each pixel of the matching window and the template image within the detection range; and obtaining the region where the matching window with the maximum correlation score is located as the matching region.

17. The measurement method as described in claim 16, characterized in that, The methods for calculating the correlation score between the detected image and the template image within the detection range include the mean absolute difference algorithm, the sum of absolute errors algorithm, the sum of squared errors algorithm, the mean sum of squared errors algorithm, the normalized cross-correlation algorithm, the sequential similarity detection algorithm, or the Hadamard transform algorithm.

18. The measurement method according to claim 1, characterized in that, The step of obtaining the template image of the mark to be tested includes: obtaining the design image corresponding to the mark to be tested as the template image; Alternatively, a captured image of a standard marker can be used as a template image, wherein the standard marker is identical to the marker to be tested and the standard marker meets the design requirements; Alternatively, a theoretical model of the interaction between the sample and light can be provided, which represents the relationship between the intensity distribution of the emitted light after the light interacts with the sample and the physical parameters of the sample; the physical parameters of the marker to be tested are substituted into the theoretical model to obtain a simulated image, which serves as a template image.

19. A measurement system, characterized in that, The measurement system is used to measure a marker under test, the marker having a region of interest, and includes: An image acquisition module is used to acquire a detection image and a template image of the marker to be tested. The template image has a detection range and includes a preset positional relationship between a first region of interest and the detection range. The first region of interest is the region of interest of the marker to be tested within the template image; A localization module is used to perform region of interest (ROI) localization processing on the detection image using the template image to obtain a ROI corresponding to a first marker and a ROI corresponding to a second marker. The localization module includes: a matching unit, used to match the detection image with template images within a detection range, and in the detection image, obtain the region with the highest similarity to the template images within the detection range as the matching region, and use the detection image of the matching region as the matching image; and a ROI determination unit, used to determine a second ROI in the detection image based on a preset positional relationship between the first ROI in the template image and the detection range, wherein the second ROI is the ROI of the marker to be tested in the detection image. The region in the image, the detected image within the second region of interest is taken as the image of interest, the relative positional relationship between the image of interest and the matching image is the same as the preset positional relationship, used in the image of interest in the detected image, the region of interest is obtained by performing region of interest localization processing on the first mark and the second mark respectively; and after using the template image to perform region of interest localization processing on the detected image to obtain the region of interest corresponding to the first mark and the second mark, the alignment error between the first mark and the second mark is obtained using the region of interest corresponding to the first mark and the region of interest corresponding to the second mark.

20. A device, characterized in that, It includes at least one memory and at least one processor, the memory storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the measurement method as described in any one of claims 1 to 18.

21. A storage medium, characterized in that, The storage medium stores one or more computer instructions for implementing the measurement method as described in any one of claims 1 to 18.

Images