Fast alignment method, apparatus, device, and storage medium

Abstract

The application provides a fast alignment method, device, equipment and storage medium, the fast alignment method comprises: determining a target area graph from a design layout according to a scanning electron microscope image; processing the scanning electron microscope image to obtain a first contour graph of a target format; determining a plurality of candidate alignment areas from the target area graph according to first feature information of the first contour graph and second feature information of the target area graph; calculating a first similarity between the plurality of candidate alignment areas and the first contour graph respectively; determining a target alignment result according to the first similarity between the plurality of candidate alignment areas and the first contour graph. In this way, through the two alignment operations of initial alignment and accurate alignment, the alignment accuracy and efficiency of the SEM image and the design layout are improved, and the accurate and fast alignment effect is achieved.

Description

Technical Field

[0001] This application belongs to the field of chip manufacturing technology, and in particular relates to a rapid alignment method, a rapid alignment device, an electronic device, and a computer-readable storage medium. Background Technology

[0002] In the current chip manufacturing process, as chip manufacturing technology continues to advance, the defects generated during production are becoming smaller and smaller, making electron beam defect detection particularly important. Traditional electron beam defect detection mainly adopts a "die-to-die" inspection method. However, this method is ineffective against systematic defects and aperiodic regions, and the defect detection rate is overly dependent on the alignment between the inspection image and the reference image. To address these issues, design-based electron beam defect detection methods are becoming increasingly important.

[0003] However, in electron beam defect detection, the alignment accuracy between the design layout and the scanning electron microscope (SEM) image directly affects the precision of subsequent inspections. Currently, the alignment methods between the design layout and the SEM image are time-consuming and prone to large alignment errors. Therefore, developing a precise and rapid alignment method has become a pressing need. Summary of the Invention

[0004] This application provides a rapid alignment method, rapid device, electronic device, and computer-readable storage medium, which can improve the alignment accuracy and efficiency of SEM images and design layouts through two alignment operations: initial alignment and precise alignment, achieving accurate and rapid alignment results.

[0005] According to a first aspect of this application, a rapid alignment method is provided, the rapid alignment method comprising:

[0006] Based on scanning electron microscope images, the target area map is determined from the design layout;

[0007] The scanning electron microscope image is processed to obtain the first contour map in the target format;

[0008] Based on the first feature information of the first contour map and the second feature information of the target region map, multiple candidate alignment regions are determined from the target region map;

[0009] Calculate the first similarity between multiple candidate alignment regions and the first contour map;

[0010] The target alignment result is determined based on the first similarity between multiple candidate alignment regions and the first contour map.

[0011] According to a second aspect of this application, a rapid alignment device is provided, comprising:

[0012] The area map determination module is used to determine the target area map from the design layout based on scanning electron microscope images;

[0013] The contour map acquisition module is used to process the scanning electron microscope image to obtain a first contour map in the target format;

[0014] The candidate region determination module is used to determine multiple candidate alignment regions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map;

[0015] The first similarity calculation module is used to calculate the first similarity between multiple candidate alignment regions and the first contour map respectively;

[0016] The target result determination module is used to determine the target alignment result based on the first similarity between multiple candidate alignment regions and the first contour map.

[0017] According to a third aspect of this application, an electronic device is provided, comprising: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements any of the above-described fast alignment methods.

[0018] According to a fourth aspect of this application, a computer-readable storage medium is provided, characterized in that computer program instructions are stored on the computer-readable storage medium, which, when executed by a processor, implement any of the above-described fast alignment methods.

[0019] In summary, the rapid alignment method, rapid alignment device, electronic device, and computer-readable storage medium provided in this application can determine the corresponding target area map from the design layout based on a scanning electron microscope image, thereby initially determining the alignment range of the layout. By processing the scanning electron microscope image, a first contour map of the target format is obtained, thereby converting the scanning electron microscope image into a first contour image, which facilitates subsequent alignment operations. Furthermore, based on the first feature information of the first contour map and the second feature information of the target area map, multiple candidate alignment areas are determined from the target area map. Thus, preliminary alignment operations can be performed using feature information, and multiple candidate alignment areas with high alignment probabilities can be selected from the target area map, while invalid areas are eliminated. This reduces the need for similarity calculations on invalid positions in subsequent fine alignment operations, significantly improving alignment efficiency. Additionally, a first similarity between the multiple candidate alignment areas and the first contour map is calculated respectively. Based on the first similarity between the multiple candidate alignment areas and the first contour map, the target alignment result is determined. Therefore, a similarity calculation can be performed on each candidate area through a sliding operation to achieve fine alignment operations. Furthermore, by measuring the target alignment result based on multiple first similarities, the accuracy of the alignment result can be guaranteed to the greatest extent. Thus, by performing two alignment operations—initial alignment and precise alignment—the alignment accuracy and efficiency between the SEM image and the design layout are improved, achieving a precise and fast alignment effect. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of this application or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating an alignment method provided by the prior art is shown;

[0022] Figure 2 This diagram illustrates a flowchart of a rapid alignment method provided in an embodiment of this application.

[0023] Figure 3 This diagram illustrates a scanning electron microscope image provided in an embodiment of this application.

[0024] Figure 4 This diagram illustrates a target area map provided in an embodiment of this application.

[0025] Figure 5 This is a schematic diagram showing a first contour view of a target format provided in an embodiment of this application;

[0026] Figure 6 This diagram illustrates a first contour map of a marker for first feature information provided in an embodiment of this application.

[0027] Figure 7 This diagram illustrates a target region map showing candidate alignment positions and candidate alignment regions according to an embodiment of this application.

[0028] Figure 8 This diagram illustrates a sliding window method within a candidate alignment area according to an embodiment of this application.

[0029] Figure 9 This diagram illustrates an alignment position provided in an embodiment of this application.

[0030] Figure 10 This illustration shows a flowchart of another rapid alignment method provided in an embodiment of this application;

[0031] Figure 11 This diagram illustrates the structure of a rapid alignment device according to an embodiment of this application.

[0032] Figure 12 This diagram illustrates the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0033] To make the above and other features and advantages of this application clearer, the application is further described below with reference to the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explanation to those skilled in the art, and are exemplary only, not restrictive.

[0034] In the following description, numerous specific details are set forth to provide a thorough understanding of this application. However, it will be apparent to those skilled in the art that the specific details are not required to practice this application. In other instances, well-known steps or operations have not been described in detail to avoid obscuring this application.

[0035] Currently, the alignment method between SEM images and design layouts first requires acquiring SEM images at designated locations. After converting the SEM image outline into the same format as the design layout and saving it as a template, a sliding window of a certain size needs to be defined on the design layout near the SEM image capture location. The similarity between the design layout and the SEM image outline is calculated frame-by-frame by sliding the template within the window; specifically, the area similarity between the shapes. When the similarity index calculated at a certain sliding position exceeds a preset threshold, the sliding stops, and the current position is output as the final alignment position.

[0036] Specifically, Figure 1The diagram illustrates a flow chart of an alignment method provided by the prior art, such as... Figure 1 As shown, the specific steps of this alignment method are as follows:

[0037] S11: Acquire scanning electron microscope images at a specified location, extract the contours from the SEM images, and convert the SEM image contours into the same format as the design layout.

[0038] S12, determine a certain alignment range on the design layout based on the image acquisition accuracy.

[0039] S13, within the alignment range, calculate the similarity between the design layout and the SEM outline frame by frame in a sliding manner.

[0040] S14, determine whether the similarity index calculated for the current sliding position is greater than the preset threshold; if it is, proceed to S15. If the area similarity of all sliding positions is not greater than the preset threshold, proceed to S16.

[0041] S15 indicates that the alignment is successful, stops sliding, and outputs the current position as the final alignment position.

[0042] S16 indicates that the alignment has failed.

[0043] However, the alignment method described above requires sliding frame by frame across the design layout within the set alignment range and calculating the similarity between the SEM image outline and the design layout. Therefore, when the machine stage or eBeam accuracy is low, the SEM image capture position is far from the specified capture position, resulting in a larger alignment range. Consequently, similarity calculations need to be performed at more sliding positions, significantly reducing alignment efficiency. Furthermore, in applications requiring higher alignment accuracy, obtaining a more accurate alignment position also necessitates similarity calculations at more locations, further decreasing alignment efficiency.

[0044] At the same time, even in scenarios where the alignment range is small and the alignment accuracy requirement is not high, the above alignment method, which uses area similarity as a similarity index to determine whether alignment has been achieved, is still at risk of failure. Especially when the SEM image outline or design layout is simple but has a large area, a high similarity cannot fully represent the similarity of the graphic shape. Only considering the similarity in terms of area and ignoring the similarity in terms of shape required for successful alignment can lead to returning the wrong alignment position in some cases.

[0045] Furthermore, in the aforementioned alignment method, the alignment process stops and returns to the current position once the calculated similarity at a certain sliding position exceeds a preset threshold. This preset threshold needs to be set based on different design layouts or different layers within the design layout. A low threshold might cause the process to exit prematurely, and the returned position might not be the optimal alignment position. Conversely, a high threshold might result in the correct alignment position not meeting the threshold similarity requirement after similarity calculations are completed at all sliding positions, leading to a false alarm of alignment failure. In other words, the threshold setting significantly impacts the accuracy of the alignment position, potentially causing misjudgments and reducing alignment precision.

[0046] In view of this, embodiments of this application provide a fast and accurate alignment method. First, the SEM image is converted into a contour map consistent with the design layout format. Feature information of the contour map and the design layout is extracted and feature matching is performed, thus obtaining multiple candidate alignment regions for initial alignment. A more precise sliding alignment operation is performed on all candidate alignment regions, calculating the similarity between each sliding position and the contour map. The alignment result can be determined based on the similarity. Thus, through the two alignment operations of initial alignment and precise alignment, on the one hand, false alarms of alignment positions can be reduced, and the position with the highest global similarity can be used as the alignment position, thereby improving the alignment accuracy. On the other hand, the similarity calculation of invalid positions can be reduced, significantly improving alignment efficiency.

[0047] One aspect of this application provides a rapid alignment method, which can be applied to the rapid alignment device provided in this application, and the rapid alignment device can be configured in the electronic device provided in this application. Figure 2 This document illustrates a flowchart of a rapid alignment method provided in an embodiment of this application. Figure 2 As shown, this rapid alignment method includes the following steps.

[0048] S21. Based on the scanning electron microscope image, determine the target area map from the design layout.

[0049] S22, process the scanning electron microscope image to obtain the first contour map in the target format.

[0050] S23, based on the first feature information of the first contour map and the second feature information of the target region map, determine multiple candidate alignment regions from the target region map.

[0051] S24, calculate the first similarity between multiple candidate alignment regions and the first contour map respectively.

[0052] S25, determine the target alignment result based on the first similarity between multiple candidate alignment regions and the first contour map.

[0053] In the above embodiments, a target region map corresponding to the design layout can be determined from the scanning electron microscope image, thereby initially determining the alignment range of the layout. By processing the scanning electron microscope image, a first contour map of the target format is obtained, thus converting the scanning electron microscope image into a first contour image, facilitating subsequent alignment operations. Furthermore, based on the first feature information of the first contour map and the second feature information of the target region map, multiple candidate alignment regions are determined from the target region map. In this way, preliminary alignment operations can be performed using feature information, allowing multiple candidate alignment regions with high alignment probabilities to be selected from the target region map, while invalid regions are eliminated. This reduces the need for similarity calculations on invalid positions in subsequent fine alignment operations, significantly improving alignment efficiency. Additionally, a first similarity between each of the multiple candidate alignment regions and the first contour map is calculated. Based on the first similarity between the multiple candidate alignment regions and the first contour map, the target alignment result is determined. This allows for similarity calculations on each candidate region through a sliding operation to achieve fine alignment operations. Furthermore, evaluating the target alignment result based on multiple first similarities maximizes the accuracy of the alignment result. Thus, by performing two alignment operations—initial alignment and precise alignment—the alignment accuracy and efficiency between the SEM image and the design layout are improved, achieving a precise and fast alignment effect.

[0054] The steps described above are described in detail below.

[0055] In S21, the scanning electron microscope image can be an image obtained by taking a picture of a specified location on the chip using a high-precision scanning electron microscope. This specified location can be any location within the non-blank pattern area of ​​the chip. Figure 3 This diagram illustrates a scanning electron microscope image provided in an embodiment of this application. The design layout can be a chip design layout. The target area map can be a layout area corresponding to the scanning electron microscope image. Figure 4 This diagram illustrates a target region map provided in an embodiment of this application. Specifically, the chip range corresponding to the target image region and the chip range corresponding to the scanning electron microscope image both include the chip region at a specified location.

[0056] Furthermore, the chip area corresponding to the target image region is slightly larger than the chip area corresponding to the scanning electron microscope image. The target area map can be determined based on the imaging precision of the scanning electron microscope. For example, if the imaging precision of the scanning electron microscope is 3µm, the target area map can be 5µm to 10µm.

[0057] In other words, the rapid alignment device can determine a pre-sized alignment area on the design layout based on the image acquisition accuracy of the scanning electron microscope, and use this alignment area as the target area map.

[0058] In this embodiment of the application, the rapid alignment device controls a high-precision scanning electron microscope to take a picture of a specified position on the chip, thereby obtaining a scanning electron microscope image corresponding to the specified position. The rapid alignment device also extracts a region graphic containing the specified position from the chip's design layout, thereby obtaining a target region map.

[0059] In S22, the processing can be image processing, which may include, but is not limited to, contour extraction and format conversion. The target format can be a design layout format. Optionally, the target format can be a Graphic Design System (GDS) format or an Open Artwork System Interchange Standard (OASIS) format. The first contour map can be a graphic containing the contour of a scanning electron microscope image. Figure 5 This is a schematic diagram showing a first contour view of a target format provided in an embodiment of this application.

[0060] In this embodiment of the application, the rapid alignment device performs image processing on the scanning electron microscope image to obtain a first contour map corresponding to the scanning electron microscope image, and the format of the first contour map is consistent with the format of the design layout.

[0061] In S23, the first feature information can be feature information of the first contour map. The second feature information can be feature information of the target region map. The candidate alignment region can be a region in the target region map that has the same features as the first contour map. The area of ​​the candidate alignment region is larger than the area of ​​the first contour map. Optionally, the area of ​​the candidate alignment region can be more than twice the area of ​​the first contour map. In this embodiment, the fast alignment device searches for feature information similar to the first feature information from the second feature information, and determines the regions where these feature information is located from the target region map, and uses these regions as candidate alignment regions.

[0062] It should be noted that the candidate alignment region can be a potential alignment region of the first contour map on the target region map.

[0063] In S24, the first similarity can be the similarity between the candidate alignment region and the first contour map. In this embodiment, the fast alignment device calculates the similarity between each candidate alignment region and the first contour map, thereby obtaining multiple first similarities. Each first similarity pair corresponds to one candidate alignment region.

[0064] In the embodiments of this application, there are multiple ways to calculate the first similarity. Optionally, the first similarity can be obtained by calculating the area intersection-union ratio.

[0065] Specifically, the formula for calculating the first similarity is as follows:

[0066] Similarity = (Template_Area∩Design_Area) / (Template_Area∪Design_Area) (1)

[0067] Where Template_Area is the area of ​​the polygon in the first contour map, Design_Area is the area of ​​the polygon in the subsequent alignment area, and Similarity is the first similarity.

[0068] It should be noted that the polygons in the first contour map can be polygons formed by contours in the first contour map. The polygons in the candidate alignment region can be polygons formed by contours in the candidate alignment region.

[0069] In step S25, the target alignment result can include both successful and unsuccessful alignment. In the case of successful alignment, the target alignment result can further include the target alignment position. The target alignment position is the region in the target region map that best matches the first contour map. The target alignment position can be one of several candidate regions.

[0070] In some embodiments, S22, processing the scanning electron microscope image to obtain a first contour map in a target format may include:

[0071] Contour extraction is performed on the scanning electron microscope image to obtain a second contour map;

[0072] The second contour image is formatted to obtain the first contour image in the target format.

[0073] Here, contour extraction can employ existing contour extraction methods, such as hollowing-out algorithms and tracking algorithms. The second contour map can be in an image format, such as JPG. The target format can be a design layout format, such as GDS or OASIS. Both the first and second contour maps are contour maps corresponding to the scanning electron microscope image, containing all the contours of the scanning electron microscope.

[0074] In this embodiment, the rapid alignment device extracts the contour of a scanning electron microscope image using a contour extraction algorithm to obtain a second contour map corresponding to the scanning electron microscope image. The second contour map is then converted from an image format to a target format to obtain a first contour map.

[0075] In the above embodiments, contour extraction and format conversion of the scanning electron microscope can convert the scanning electron microscope into a contour map that can be aligned with the design layout. This facilitates subsequent alignment calculations and reduces alignment errors caused by format mismatch, thus improving alignment accuracy, since the format of the contour map is consistent with the format of the design layout.

[0076] In some embodiments, S23, determining a plurality of candidate alignment regions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map may include:

[0077] Based on the first feature information of the first contour map and the second feature information of the target region map, multiple candidate alignment positions are determined from the target region map;

[0078] Based on the position of each candidate alignment position in the target area map, determine the candidate alignment region corresponding to each candidate alignment position to obtain multiple candidate alignment regions.

[0079] Here, the first feature information can be the feature information of polygons within the first contour map. The second feature information can be the feature information of multiple polygons in the target region map.

[0080] In this embodiment of the application, the rapid alignment device uses a feature matching method to match the first feature information with the second feature information, selects feature information from the second feature information that matches the feature information of the polygon in the first contour map, and determines the area where these matched feature information is located in the target area map, and uses these areas as candidate alignment positions.

[0081] The candidate alignment position can be the location in the target region map that matches the first feature information in the first contour map. The candidate alignment region can be the region containing the candidate alignment position. Optionally, the candidate alignment region can be a region slightly larger than the bounding rectangle of the candidate alignment position. The area of ​​the candidate alignment region can be more than twice the area of ​​the bounding rectangle of the candidate alignment position.

[0082] In the above embodiments, by using the first feature information to pre-screen the candidate alignment positions in the target area map, the candidate alignment positions that match the features of the first contour map can be screened out from the target area map by feature matching method. In this way, the preliminary alignment position in the target area map can be accurately determined.

[0083] Furthermore, compared to existing technologies that slide the contour map frame by frame on the design layout and calculate the similarity at each sliding position, this application can pre-select candidate alignment regions with a higher alignment probability for subsequent similarity calculation, thereby reducing similarity calculation at invalid positions and significantly improving alignment efficiency.

[0084] In some embodiments, prior to S23, the rapid alignment method further includes:

[0085] Feature analysis is performed on the first contour map and the target area map respectively to obtain the first feature information and the second feature information.

[0086] Here, the rapid alignment device can perform feature analysis on the first contour map using a graphic feature analysis method to obtain first feature information. Furthermore, the rapid alignment device can use the same graphic feature analysis method to perform feature analysis on the target area map to obtain second feature information. The feature analysis may include, but is not limited to, the analysis of graphic features such as shape, corner points, centroid, and area.

[0087] The first feature information and the second feature information may include at least one of the following: the corner positions of the polygon, the centroid position of the polygon, the aspect ratio and relative position of the circumscribed rectangle of the polygon, and the area of ​​the polygon.

[0088] In this embodiment, the rapid alignment device performs feature analysis on the first contour map. This can be achieved by the rapid alignment device identifying a polygon from the first contour map, and then performing feature analysis on this polygon to obtain first feature information. Similarly, the rapid alignment device performs feature analysis on the target area map. This can be achieved by the alignment device identifying multiple polygons from the target area map and performing feature analysis on each polygon to obtain second feature information. Thus, through feature analysis, both the first and second feature information can be extracted, facilitating subsequent preliminary alignment.

[0089] In some embodiments, feature matching is performed between the first feature information and the second feature information using a feature matching method to determine candidate alignment positions that match the polygons in the first contour map from the target region map, which may include:

[0090] The third feature information that matches the first feature information in the second feature information is determined by feature matching method, and the position of the third feature information in the target area map is determined. One or more positions corresponding to the third feature information are used as candidate alignment positions.

[0091] Here, the third feature information can be the feature information in the second feature information that matches the first feature information, and the third feature information includes the same types of features as the first feature information. The third feature information can be at least one of the following: the corner position of the polygon, the centroid position of the polygon, the aspect ratio and relative position of the circumscribed rectangle of the polygon, and the area of ​​the polygon.

[0092] In the above embodiments, a preliminary alignment operation is performed by feature matching, thereby initially screening out areas with a higher probability of alignment.

[0093] In some embodiments, the location of the third feature information in the target area map can be part or all of the polygon in the target area map that satisfies at least one of the following conditions: the corner position of the polygon is close to the corner position of the polygon in the first contour map; the mass position of the polygon is close to the mass position of the polygon in the first contour map; the aspect ratio and relative position of the circumscribed rectangle of the polygon and the circumscribed rectangle of the polygon in the first contour map are close; and the area of ​​the polygon is close to the area of ​​the polygon in the first contour map.

[0094] In this embodiment of the application, the corner positions of the polygon are close to the corner positions of the polygon in the first contour map. This can be understood as the error between the corner positions of the polygon in the target area map and the corner positions of the polygon in the first contour map being within a very small first error range.

[0095] The position of the point mass of the polygon is close to that of the point mass of the polygon in the first contour map. This can be understood as the error between the position of the point mass of the polygon in the target area map and the position of the point mass of the polygon in the first contour map being within a very small second error range.

[0096] The aspect ratio and relative position of the bounding rectangle of the polygon in the target area map are close to those of the bounding rectangle of the polygon in the first contour map. This can be understood as the error between the aspect ratio of the bounding rectangle of the polygon in the target area map and the aspect ratio of the bounding rectangle of the polygon in the first contour map being within a very small third error range, and the error between the relative position of the bounding rectangle of the polygon in the target area map and the relative position of the bounding rectangle of the polygon in the first contour map being within a very small fourth error range.

[0097] The area of ​​the polygon is close to the area of ​​the polygon in the first contour map. This can be understood as the error between the area of ​​the polygon in the target area map and the area of ​​the polygon in the first contour map being within a very small fifth error range.

[0098] It should be noted that the first error range, the second error range, the third error range, the fourth error range, and the fifth error range can represent the minimum error range and can be set according to requirements.

[0099] In the above embodiments, the feature information includes one or more features. This allows the first contour image to be initially aligned with the target region image by considering area and graphic features, ensuring that the graphic area and / or graphic features of the first contour image are close to those of the candidate alignment position. This reduces false alarms about alignment positions and further narrows down the alignment range. Furthermore, adding a matching process for other graphic features besides area features maximizes the reliability of feature matching and reduces false alarms about candidate alignment positions.

[0100] For example, Figure 6This illustration shows a schematic diagram of a first contour map that marks first feature information according to an embodiment of this application, such as... Figure 6 As shown, the first feature information may include the positions of two corner points, marked by circular boxes. The rapid alignment device searches for two similar corner points that match the positions of these two corner points from the second feature information. Figure 7 This diagram illustrates a target area map showing candidate alignment positions and candidate alignment regions, as provided in an embodiment of this application. Figure 7 As shown, the corner features of the region within the circular frame are similar to the corner features of the first contour map. Therefore, the region within the circular frame is selected as a candidate alignment position. Figure 7 The square boxes in the diagram represent candidate alignment regions. Each square box contains a circular box, which can be understood as an area slightly larger than the outer rectangle of the circular box.

[0101] In some embodiments, S24, calculating the first similarity between the plurality of candidate alignment regions and the first contour map may include:

[0102] The preset sliding window is slid in each candidate alignment area with a preset step size, and the similarity between the area covered by the sliding window at each sliding position and the first contour map is calculated to obtain the second similarity between multiple sliding positions and the first contour map.

[0103] Based on the second similarity corresponding to multiple sliding positions in each candidate alignment region, a first similarity between each candidate alignment region and the first contour map is determined.

[0104] Here, the preset size of the sliding window is the same as the size of the first contour map. The preset step size can be set according to requirements. It should be noted that the smaller the preset step size, the more accurate the alignment. The sliding position can be the corresponding coverage position where the sliding window slides across the candidate alignment area with a preset step size.

[0105] Calculating the similarity between the area covered by the sliding window at each sliding position and the first contour map can be understood as calculating the similarity between the area covered by the sliding window at each sliding position within the candidate alignment region and the first contour map based on a similarity calculation method. There are various similarity calculation methods. Optionally, formula (1) can be used to calculate the similarity, i.e., the area intersection-union ratio is used as the similarity.

[0106] The second similarity can refer to the similarity between the sliding window coverage area corresponding to the sliding position and the first contour map. One second similarity corresponds to one sliding position, meaning the number of sliding positions is the same as the number of second similarities.

[0107] It should be noted that a candidate alignment region has multiple sliding positions, and each sliding position corresponds to a second similarity.

[0108] Figure 8 This diagram illustrates a sliding window method within a candidate alignment area according to an embodiment of this application. Figure 8 As shown, the arrows indicate the sliding direction of the sliding window. The sliding window can slide from the top left corner of the candidate area to the bottom right corner in both horizontal and vertical directions with a preset step size. That is, it slides horizontally. After sliding one row, the sliding window slides one step vertically, and then continues to slide horizontally until the sliding window reaches the bottom right corner. At this point, the sliding window has finished sliding in one candidate alignment area.

[0109] It should be noted that there are multiple ways to slide a sliding window, such as sliding from the upper right corner to the lower left corner, from the lower left corner to the upper right corner, and from the lower right corner to the upper left corner. In this embodiment, the sliding method of the sliding window is not limited.

[0110] The first similarity can be one of several second similarities between the candidate alignment region and the first contour map. The sliding position corresponding to the first similarity can be the position in the candidate alignment region that is most similar to the first contour map.

[0111] In the above embodiments, by sliding the sliding window with a preset step size, the similarity between each sliding position in each candidate alignment area and the first contour map is detected, and the first similarity between each candidate alignment area and the first contour map is determined. In this way, more accurate sliding alignment can be performed in each candidate alignment area, so as to select the global best alignment position in the subsequent process, thereby improving the accuracy of alignment.

[0112] In some embodiments, determining a first similarity between each candidate alignment region and the first contour map based on a second similarity corresponding to multiple sliding positions in each candidate alignment region includes:

[0113] From the second similarity scores corresponding to multiple sliding positions in each candidate alignment region, the maximum second similarity score is selected.

[0114] The maximum second similarity corresponding to each candidate alignment region is determined as the first similarity between each candidate alignment region and the first contour map.

[0115] Here, the first similarity can be the largest second similarity among multiple second similarities corresponding to the candidate alignment region. The largest second similarity indicates that the sliding position has the greatest similarity to the first contour map, that is, the sliding position corresponding to the largest similarity is the position in the candidate alignment region that is most similar to the first contour map.

[0116] In the above embodiments, for each candidate alignment region, the largest second similarity within each candidate alignment region is selected, and the largest second similarity within each candidate alignment region is taken as its corresponding first similarity. In this way, the sliding position most similar to the first contour within each candidate alignment region can be determined.

[0117] In some embodiments, S25, determining the target alignment result based on a first similarity between a plurality of candidate alignment regions and a first contour map includes:

[0118] The maximum first similarity is selected from the first similarity between multiple candidate alignment regions and the first contour map;

[0119] If the maximum first similarity is greater than a preset threshold, the alignment is determined to be successful, and the sliding position in the candidate alignment region corresponding to the maximum first similarity is determined as the target alignment position.

[0120] If the maximum first similarity is not greater than a preset threshold, the alignment is determined to be a failure.

[0121] Here, the maximum first similarity score can be the largest among multiple first similarity scores. The maximum first similarity score can represent the highest degree of similarity between multiple candidate alignment regions and the first contour map. The preset threshold can be determined according to requirements; for example, the preset threshold can be any natural number greater than 90%.

[0122] In this embodiment, to further improve alignment accuracy, the rapid alignment device compares the maximum first similarity with a preset threshold and determines the alignment result based on the comparison result. Specifically, if the maximum first similarity is not greater than the preset threshold, alignment is determined to have failed. If the maximum first similarity is greater than the preset threshold, alignment is determined to have succeeded, and the sliding position corresponding to the candidate alignment region corresponding to the maximum first similarity is determined as the target alignment position. Figure 9 This illustration shows a schematic diagram of an alignment position provided in an embodiment of this application, such as... Figure 9 As shown, the area indicated by the dashed box is the sliding position in the target area map that is aligned with the first contour map.

[0123] In the above embodiments, the maximum first similarity is first selected from multiple first similarity values. This allows for the preliminary determination of the sliding position within multiple candidate regions that best matches the first contour map. A preset threshold is then used to determine whether the sliding position corresponding to the maximum first similarity is the alignment position of the first contour map. Compared to existing technologies that determine whether to prematurely terminate the alignment process based on a preset threshold during sliding, where the returned alignment result is highly susceptible to threshold influence and cannot guarantee the globally optimal alignment position, this application can perform more precise sliding alignment within all candidate alignment regions. This ensures that the most accurate returned alignment position originates from the position with the highest global similarity, further improving alignment accuracy and reducing false positives or false negatives due to unreasonable preset thresholds.

[0124] To further enhance the understanding of rapid alignment methods, this application provides another rapid alignment method. Figure 10 This document illustrates a flowchart of another rapid alignment method provided in an embodiment of this application, as shown below. Figure 10 As shown, this rapid alignment method may include:

[0125] S101: Collect SEM images at a specified location, extract the contours from the SEM images, and convert the SEM image contours into the same format as the design layout.

[0126] S102, determine a target area map of a certain size on the design map according to the map acquisition accuracy;

[0127] S103, the same feature extraction algorithm is used to extract the graphic features of the SEM image contour map and the target area map respectively;

[0128] Here, graphic features can be the location of the corner points and centroid of each polygon in the graphic, the aspect ratio and relative position of the bounding rectangle of the polygon, the area of ​​the polygon, etc.

[0129] S104, Based on the extracted feature information (i.e., the first feature information and the second feature information), the SEM image contour map and the target area map are initially aligned by feature matching, and multiple potential final alignment positions are obtained.

[0130] Here, the preliminary alignment can be determined by at least one of the following methods: the size and relative position of the bounding rectangle of the polygon in the SEM outline graphic and the bounding rectangle of the polygon in the design layout graphic should be similar, and the centroids of each polygon should be distributed in similar positions, etc.

[0131] S105, perform a more precise sliding alignment operation near all potential alignment positions (i.e. candidate alignment regions), and calculate the similarity between the SEM image contour map and the target region map at each sliding position, and record the maximum similarity and the corresponding sliding position.

[0132] S106, determine whether the maximum similarity is greater than the preset threshold. If it is greater, proceed to S107; otherwise, proceed to S108.

[0133] S107, Confirm alignment successful, return to the corresponding sliding position as the final alignment position;

[0134] S108, Alignment failed.

[0135] In the above embodiments, since the SEM image outline and the design layout are analyzed before alignment, the potential alignment positions in the design layout are screened in advance by using the feature information of the SEM image. Compared with the prior art, which slides the SEM image frame by frame on the design layout and calculates the similarity at each sliding position, the present invention can avoid similarity calculations at most positions, thereby significantly improving alignment efficiency.

[0136] Meanwhile, when calculating similarity, since graphic feature information is referenced in advance, it can be ensured that at the calculated alignment position, not only are the areas of the SEM image graphic and the design layout graphic similar, but the features of the SEM image outline graphic and the design layout graphic are also similar, thus avoiding false alarms about alignment position to a certain extent.

[0137] Furthermore, existing technologies determine whether to prematurely terminate the alignment process based on a preset threshold during the sliding process. The returned alignment result is highly susceptible to the threshold and cannot guarantee that it is the globally optimal alignment position. In contrast, the embodiments of this application can perform more precise sliding alignment near all potential alignment positions, and the final alignment position will be generated from the position with the highest similarity, thus improving the accuracy of the alignment result.

[0138] Another embodiment of this application provides a rapid alignment device. Figure 11 The figure shows a schematic diagram of a rapid alignment device provided in an embodiment of this application. As shown, the rapid alignment device may include:

[0139] The region map determination module 111 is used to determine the target region map from the design layout based on the scanning electron microscope image;

[0140] The contour map acquisition module 112 is used to process the scanning electron microscope image to obtain a first contour map in the target format;

[0141] The candidate region determination module 113 is used to determine multiple candidate alignment regions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map;

[0142] The first similarity calculation module 114 is used to calculate the first similarity between the multiple candidate alignment regions and the first contour map respectively;

[0143] The target result determination module 115 is used to determine the target alignment result based on the first similarity between the multiple candidate alignment regions and the first contour map.

[0144] In the above embodiments, a target region map corresponding to the design layout can be determined from the scanning electron microscope image, thereby initially determining the alignment range of the layout. By processing the scanning electron microscope image, a first contour map of the target format is obtained, thus converting the scanning electron microscope image into a first contour image, facilitating subsequent alignment operations. Furthermore, based on the first feature information of the first contour map and the second feature information of the target region map, multiple candidate alignment regions are determined from the target region map. In this way, preliminary alignment operations can be performed using feature information, allowing multiple candidate alignment regions with high alignment probabilities to be selected from the target region map, while invalid regions are eliminated. This reduces the need for similarity calculations on invalid positions in subsequent fine alignment operations, significantly improving alignment efficiency. Additionally, a first similarity between each of the multiple candidate alignment regions and the first contour map is calculated. Based on the first similarity between the multiple candidate alignment regions and the first contour map, the target alignment result is determined. This allows for similarity calculations on each candidate region through a sliding operation to achieve fine alignment operations. Furthermore, evaluating the target alignment result based on multiple first similarities maximizes the accuracy of the alignment result. Thus, by performing two alignment operations—initial alignment and precise alignment—the alignment accuracy and efficiency between the SEM image and the design layout are improved, achieving a precise and fast alignment effect.

[0145] In some embodiments, the contour map obtaining module 112 may include:

[0146] The second contour map acquisition submodule is used to extract contours from scanning electron microscope images to obtain the second contour map;

[0147] The first contour map is used to obtain a sub-module, which is used to convert the format of the second contour map to obtain the first contour map in the target format.

[0148] In some embodiments, the first similarity calculation module 114 may include:

[0149] The second similarity submodule is used to slide a preset sliding window in each candidate alignment area with a preset step size, and calculate the similarity between the area covered by the sliding window at each sliding position and the first contour map to obtain the second similarity between multiple sliding positions and the first contour map.

[0150] The first similarity determination submodule is used to determine the first similarity between each candidate alignment region and the first contour map based on the second similarity corresponding to multiple sliding positions in each candidate alignment region.

[0151] In some embodiments, the first similarity determination submodule may include:

[0152] The maximum similarity filtering unit is used to filter out the maximum second similarity from the second similarities corresponding to multiple sliding positions in each candidate alignment region;

[0153] The first similarity determination unit is used to determine the maximum second similarity corresponding to each candidate alignment region as the first similarity between each candidate alignment region and the first contour map.

[0154] In some embodiments, the target result determination module 115 may include:

[0155] The maximum similarity filtering submodule is used to filter out the maximum first similarity from the first similarity between multiple candidate aligned regions and the first contour map;

[0156] The successful determination submodule is used to determine that the alignment is successful when the maximum first similarity is greater than a preset threshold, and to determine the sliding position in the candidate alignment region corresponding to the maximum first similarity as the target alignment position;

[0157] The failure determination submodule is used to determine alignment failure if the maximum first similarity is not greater than a preset threshold.

[0158] In some embodiments, the candidate region determination module 113 may include:

[0159] The candidate position submodule is used to determine multiple candidate alignment positions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map;

[0160] The region acquisition submodule is used to determine the candidate alignment region corresponding to each candidate alignment position based on the position of each candidate alignment position in the target region map, so as to obtain multiple candidate alignment regions.

[0161] In some embodiments, the candidate position submodule can be specifically used to determine the third feature information that matches the first feature information in the second feature information by feature matching method, determine the position of the third feature information in the target area map, and take one or more positions corresponding to the third feature information as candidate alignment positions.

[0162] In some embodiments, the rapid alignment device may further include:

[0163] The feature analysis module is used to perform feature analysis on the first contour map and the target region map respectively before determining multiple candidate alignment positions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map, to obtain first feature information and second feature information. The first feature information and second feature information include at least one of the following: the corner position of the polygon, the centroid position of the polygon, the aspect ratio and relative position of the circumscribed rectangle of the polygon, and the area of ​​the polygon.

[0164] It should be understood that the specific features, operations, and details described herein with respect to the methods of this application can also be similarly applied to the apparatus and system of this application, or vice versa. Furthermore, each step of the methods of this application described above can be performed by a corresponding component or unit of the apparatus or system of this application.

[0165] It should be understood that the various modules / units of the device of this application can be implemented wholly or partially through software, hardware, firmware, or a combination thereof. Each module / unit can be embedded in the processor of the electronic device in hardware or firmware form or independent of the processor, or it can be stored in the memory of the electronic device in software form for the processor to call to execute the operation of each module / unit. Each module / unit can be implemented as an independent component or module, or two or more modules / units can be implemented as a single component or module.

[0166] In another aspect, this application provides an electronic device. Figure 12 This illustration shows a structural diagram of an electronic device provided in an embodiment of this application, such as... Figure 12 As shown, the electronic device 120 includes a processor 121 and a memory 122 storing computer program instructions. The processor 121 executes the computer program instructions to implement the steps of the aforementioned fast alignment method. This electronic device can be broadly categorized as a server, terminal, or any other electronic device with the necessary computing and / or processing capabilities.

[0167] In one embodiment, the electronic device 120 may include a processor, memory, network interface, communication interface, etc., connected via a system bus. The processor of the electronic device can be used to provide necessary computing, processing, and / or control capabilities. The memory of the electronic device 120 may include non-volatile storage media and internal memory. The non-volatile storage media may store an operating system, computer programs, etc. The internal memory can provide an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface and communication interface of the electronic device 120 can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, it performs the steps of the fast alignment method of this application.

[0168] This application provides a computer-readable storage medium storing a fast alignment program, which, when executed by a processor, implements the steps of the fast alignment method provided in any of the above embodiments.

[0169] Those skilled in the art will understand that the method steps of this application can be performed by a computer program instructing related hardware, such as electronic devices or processors. The computer program can be stored in a non-transitory computer-readable storage medium, and its execution causes the steps of this application to be performed. Depending on the context, any reference herein to memory, storage, or other media may include non-volatile or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0170] The technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification, provided that such combination does not contain contradictions.

[0171] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A rapid alignment method, characterized in that, include: Based on scanning electron microscope images, the target area map is determined from the design layout; The scanning electron microscope image is processed to obtain a first contour map in the target format; Based on the first feature information of the first contour map and the second feature information of the target region map, multiple candidate alignment regions are determined from the target region map, including: determining multiple candidate alignment positions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map, wherein the first feature information and the second feature information include at least one of the following: the corner position of the polygon, the centroid position of the polygon, the aspect ratio and relative position of the circumscribed rectangle of the polygon; and determining the candidate alignment region corresponding to each candidate alignment position based on the position of each candidate alignment position in the target region map, so as to obtain multiple candidate alignment regions. Calculate the first similarity between the multiple candidate alignment regions and the first contour map; A target alignment result is determined based on a first similarity between multiple candidate alignment regions and the first contour map, wherein the target alignment result includes alignment success and alignment failure; in the case of alignment success, the target alignment result also includes a target alignment position, wherein the target alignment position is the region in the target region map that best matches the first contour map; The step of determining multiple candidate alignment positions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map includes: The third feature information that matches the first feature information in the second feature information is determined by feature matching method, and the position of the third feature information in the target area map is determined. One or more positions corresponding to the third feature information are used as candidate alignment positions. The third feature information includes the same feature types as the first feature information.

2. The rapid alignment method according to claim 1, characterized in that, The process of processing the scanning electron microscope image to obtain a first contour map in the target format includes: Contour extraction is performed on the scanning electron microscope image to obtain a second contour map; The second contour image is converted to a new format to obtain the first contour image in the target format.

3. The rapid alignment method according to claim 1, characterized in that, The step of calculating the first similarity between the plurality of candidate alignment regions and the first contour map includes: A preset sliding window is slid in each candidate alignment area with a preset step size, and the similarity between the area covered by the sliding window at each sliding position and the first contour map is calculated to obtain a second similarity between multiple sliding positions and the first contour map. Based on the second similarity corresponding to multiple sliding positions in each candidate alignment region, a first similarity between each candidate alignment region and the first contour map is determined.

4. The rapid alignment method according to claim 3, characterized in that, The step of determining the first similarity between each candidate alignment region and the first contour map based on the second similarity corresponding to multiple sliding positions in each candidate alignment region includes: From the second similarities corresponding to multiple sliding positions in each candidate alignment region, the maximum second similarity is selected; The maximum second similarity corresponding to each candidate alignment region is determined as the first similarity between each candidate alignment region and the first contour map.

5. The rapid alignment method according to claim 3, characterized in that, The step of determining the target alignment result based on the first similarity between the multiple candidate alignment regions and the first contour map includes: The maximum first similarity is selected from the first similarity between the multiple candidate alignment regions and the first contour map; If the maximum first similarity is greater than a preset threshold, the alignment is determined to be successful, and the sliding position corresponding to the candidate alignment region corresponding to the maximum first similarity is determined as the target alignment position; If the maximum first similarity is not greater than the preset threshold, the alignment is determined to be a failure.

6. The rapid alignment method according to any one of claims 1 to 5, characterized in that, Before determining multiple candidate alignment positions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map, the method further includes: Feature analysis is performed on the first contour map and the target region map respectively to obtain first feature information and second feature information. The first feature information and the second feature information also include the area of ​​the polygon.

7. A rapid alignment device, characterized in that, include: The area map determination module is used to determine the target area map from the design layout based on scanning electron microscope images; The contour map acquisition module is used to process the scanning electron microscope image to obtain a first contour map in the target format; The candidate region determination module is used to determine multiple candidate alignment regions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map; The first similarity calculation module is used to calculate the first similarity between the multiple candidate alignment regions and the first contour map respectively; The target result determination module is used to determine the target alignment result based on the first similarity between the multiple candidate alignment regions and the first contour map; The candidate region determination module includes a candidate position submodule, used to determine multiple candidate alignment positions from the target region map based on the first feature information of the first contour map and the second feature information of the target region map. The first feature information and the second feature information include at least one of the following: the corner position of the polygon, the centroid position of the polygon, and the aspect ratio and relative position of the circumscribed rectangle of the polygon. The candidate position submodule is specifically used to determine the third feature information that matches the first feature information in the second feature information by using a feature matching method, and to determine the position of the third feature information in the target area map, and to take one or more positions corresponding to the third feature information as candidate alignment positions. The third feature information includes the same type of features as the first feature information. The region acquisition submodule is used to determine the candidate alignment region corresponding to each candidate alignment position based on the position of each candidate alignment position in the target region map, so as to obtain multiple candidate alignment regions. The target alignment result includes alignment success and alignment failure; in the case of alignment success, the target alignment result also includes the target alignment position, which is the area in the target area map that best matches the first contour map.

8. An electronic device, characterized in that, The electronic device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the fast alignment method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the fast alignment method as described in any one of claims 1-6.

Images