Methods, apparatus, devices, media, and products for image-based metrology

Abstract

Methods, apparatuses, devices, media and products of image-based metrology are provided. The method includes: obtaining a reference image and a plurality of to-be-measured images, the reference image and the plurality of to-be-measured images corresponding to different chip units in a same wafer; determining reference information of a target geometric element in the reference image based on the reference image and element information associated with the target geometric element, the reference information including at least one parameter used to extract the target geometric element; and for a to-be-measured image in the plurality of to-be-measured images, migrating the reference information in the reference image to the to-be-measured image to obtain migrated reference information, the migrated reference information representing the at least one parameter in pixel coordinates of the to-be-measured image; extracting a to-be-measured geometric element matching the target geometric element from the to-be-measured image based on the migrated reference information; and determining at least part of a measurement result of the to-be-measured geometric element of the to-be-measured image based on a geometric property of the to-be-measured geometric element.

Description

Technical Field

[0001] The examples in this paper generally relate to the field of semiconductor image processing, and in particular to image-based measurement methods, apparatuses, electronic devices, computer-readable storage media, and program products. Background Technology

[0002] In the semiconductor manufacturing field, the dimensional accuracy of each component directly determines the structural functionality and electrical performance of the product. Therefore, accurately acquiring the dimensional information of semiconductor devices is crucial. Current solutions typically involve imaging key components using a scanning electron microscope to capture their microstructural features, and then extracting specific measurement data using image processing and pattern recognition technologies. However, the structural designs of semiconductor devices are highly diverse, and measurement targets in different scenarios often exhibit varied pattern features. This has placed higher demands on the industry for developing universal image measurement methods that are adaptable to multiple patterns and have a lower barrier to entry. Summary of the Invention

[0003] In a first aspect of this paper, an image-based measurement method is provided. The method includes: acquiring a reference image and multiple images to be measured, the reference image and the multiple images to be measured corresponding to different chip cells in the same wafer; determining reference information for a target geometric element in the reference image based on the reference image and element information associated with a target geometric element, the reference information including at least one parameter for extracting the target geometric element; transferring the reference information from the reference image to the images to be measured for one of the multiple images to be measured, to obtain transferred reference information, the transferred reference information representing at least one parameter in pixel coordinates of the images to be measured; extracting a target geometric element matching the target geometric element from the images to be measured based on the transferred reference information; and determining at least a portion of the measurement results of the target geometric element in the images to be measured based on the geometric properties of the target geometric element.

[0004] In a second aspect of this document, an apparatus for image-based measurement is provided. The apparatus includes: an image acquisition module configured to acquire a reference image and a plurality of test images, the reference image and the plurality of test images corresponding to different chip cells in the same wafer; an information determination module configured to determine reference information of a target geometric element in the reference image based on the reference image and element information associated with a target geometric element, the reference information including at least one parameter for extracting the target geometric element; and a result determination module configured to transfer the reference information from the reference image to the test image among the plurality of test images to obtain transferred reference information, the transferred reference information representing at least one parameter in pixel coordinates of the test image; extract test geometric elements matching the target geometric element from the test image based on the transferred reference information; and determine at least a portion of the measurement result of the test geometric element in the test image based on the geometric properties of the test geometric element.

[0005] In a third aspect of this document, an electronic device is provided. The device includes at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor. When executed by the at least one processor, the instructions cause the device to perform the method of the first aspect.

[0006] In a fourth aspect of this document, a computer-readable storage medium is provided. The computer-readable storage medium stores computer-executable instructions that can be executed by a processor to implement the method of the first aspect.

[0007] In a fifth aspect of this document, a computer program product is provided, which is tangibly stored in a computer storage medium and includes computer-executable instructions that, when executed by a device, cause the device to perform the method of the first aspect.

[0008] In this way, by determining the reference information of the target geometric element based on the reference image and the element information associated with the target geometric element, and transferring the reference information to multiple test images on the same wafer to batch extract and match the test geometric elements, a general image measurement scheme that can cover line segment and region classes can be formed, reducing the self-built threshold of measurement program set and improving the versatility and generalization of measurement of different patterns and multiple images.

[0009] It should be understood that the content described in this section is not intended to limit the key or important features of the examples in this article, nor is it intended to restrict the scope of the solution. Other features will become readily apparent from the following description. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the various examples herein will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. In the accompanying drawings, the same or similar reference numerals denote the same or similar elements, wherein: Figure 1 A schematic diagram of the example environment is shown; Figure 2 A flowchart of a method for image-based measurement according to some embodiments is shown; Figure 3A A flowchart illustrating a method for measuring a first geometric element of a line segment type according to some embodiments is shown; Figure 3B A flowchart is shown of a method for measuring a second geometric element of a region type according to some embodiments; Figure 4 A block diagram of an apparatus for image-based measurement according to some embodiments is shown; and Figure 5 A block diagram of an electronic device capable of implementing multiple illustrative embodiments is shown. Detailed Implementation

[0011] The examples in this document will now be described in more detail with reference to the accompanying drawings. While some examples are shown in the drawings, it should be understood that solutions can be implemented in various forms and should not be construed as limited to the examples presented herein. Rather, these examples are provided to provide a more thorough and complete understanding of the solutions. It should be understood that the drawings and examples in this document are for illustrative purposes only and are not intended to limit the scope of protection of the solutions.

[0012] In the description of the examples in this document, the term "comprising" and similar terms should be understood as open inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "an example" or "the example" should be understood as "at least one example". The term "some examples" should be understood as "at least some examples". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below. It should be noted that, in this document, unless explicitly stated otherwise, performing a step in "response to A" does not mean that the step is performed immediately after "A", but may include one or more intermediate steps.

[0013] The examples in this document may involve user data, data acquisition, and / or use. All of these aspects comply with relevant laws, regulations, and rules. In the examples, all data collection, acquisition, processing, manipulation, forwarding, and use are conducted with the user's knowledge and confirmation. Accordingly, when implementing each example, the type, scope of use, and usage scenarios of any data or information that may be involved should be communicated to the user and their authorization obtained through appropriate means, in accordance with relevant laws and regulations. The specific methods of notification and / or authorization can vary depending on the actual situation and application scenario; the scope of the solution is not limited in this regard.

[0014] In this manual and the sample solutions, any processing of personal information will be conducted only under legal grounds (such as obtaining the consent of the data subject or being necessary for the performance of a contract) and will only be carried out within the scope stipulated or agreed upon. A user's refusal to process personal information beyond what is necessary for basic functions will not affect the user's use of basic functions.

[0015] The term "mask" as used in this article refers to a region label represented as a binary image, where pixels within the target region are labeled as foreground (e.g., white) and pixels outside the target region are labeled as background (e.g., black).

[0016] The term "wafer" as used in this article refers to the circular silicon substrate used to manufacture semiconductor integrated circuits, often simply called a wafer. A single wafer can contain multiple chip cells.

[0017] The term "chip cell" as used in this article refers to a single chip area on a wafer, often also called a die. Different chip cells on the same wafer have the same design layout, but due to differences in manufacturing processes, there may be some differences in the microstructure of each chip cell.

[0018] The term "scanning electron microscope" as used in this article refers to an instrument that uses a focused electron beam to scan the surface of a sample to obtain high-resolution images of its microstructure, often simply referred to as SEM. In semiconductor manufacturing, SEM images are widely used for critical dimension measurement and defect detection.

[0019] Figure 1 A schematic diagram of example environment 100 is shown. (e.g.) Figure 1 As shown, example environment 100 may include electronic device 110.

[0020] In this example environment 100, electronic device 110 may run an application 115 that supports image measurement. Application 115 may be any suitable type of application for image measurement, including but not limited to semiconductor measurement software, image processing and analysis software, or other suitable applications.

[0021] exist Figure 1 In environment 100, a semiconductor wafer 101 contains multiple chip units (e.g., N chip units, where N is any suitable positive integer). Electronic device 110 can acquire multiple chip images 102 corresponding to these multiple chip units (e.g., chip image 1, chip image 2, up to chip image N). These multiple chip images 102 can be acquired in any way; for example, images of these multiple chip units can be acquired using an imaging device such as a scanning electron microscope to obtain multiple chip images 102, or electronic device 110 can acquire multiple chip images 102 from an imaging device. It should be noted that the design layouts of different chip units (dies) within the same wafer are completely identical, and the target pattern structure is consistent, with only minor physical deviations due to process variations. Electronic device 110 performs measurement processing on the multiple chip images 102 using application 115 to obtain image measurement results 103.

[0022] Electronic device 110 can be a terminal device or a server device. A terminal device can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, handheld computers, portable gaming terminals, virtual reality (VR) / augmented reality (AR) devices, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. In some embodiments, electronic device 110 can also support any type of user-facing interface (such as "wearable" circuitry).

[0023] Server-side equipment can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms. Server-side equipment may include, for example, computing systems / servers, such as mainframes, edge computing nodes, computing devices in cloud environments, and so on.

[0024] It should be understood that the structure and function of the various elements in environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of the scheme.

[0025] As mentioned above, in the semiconductor manufacturing field, imaging key components using scanning electron microscopy and extracting measurement data through image processing technology is the mainstream approach for obtaining dimensional information. However, while the Critical Dimension Scanning Electron Microscope (CDSEM) software commonly used in semiconductor metrology supports methods such as contour extraction and image grayscale curves, it has significant limitations in practical applications: for different patterns, algorithm parameters need to be adjusted one by one, and a dedicated set of measurement recipes needs to be created, resulting in high difficulty and time consumption in using the tool; moreover, its measurement capabilities are limited to single patterns, making it difficult to adapt to overlay (OVL) calculations of multiple overlay patterns. In addition, besides measuring critical dimensions (CD), current pattern measurement needs are more diverse, requiring better basic elements combined with simpler methods for processing.

[0026] In view of the above, there is a need for a general image measurement method that is easy for users to build themselves and can adapt to a variety of patterns, so as to reduce the barrier to building measurement program sets and improve the universality and generalization of different patterns.

[0027] An image-based measurement scheme is proposed. According to the scheme, a reference image and multiple test images are acquired, the reference image and the multiple test images corresponding to different chip cells in the same wafer; based on the reference image and element information associated with the target geometric element, reference information of the target geometric element in the reference image is determined, the reference information including at least one parameter for extracting the target geometric element; and for the test images among the multiple test images, the reference information in the reference image is transferred to the test images to obtain transferred reference information, the transferred reference information representing at least one parameter in the pixel coordinates of the test image; based on the transferred reference information, test geometric elements matching the target geometric element are extracted from the test images; and based on the geometric properties of the test geometric elements, at least a portion of the measurement results of the test geometric elements in the test images are determined.

[0028] The above scheme determines reference information for the target geometric element based on a reference image and element information associated with the target geometric element. This reference information is then transferred to multiple test images on the same wafer. Based on the transferred reference information, the test geometric element matching the target geometric element is determined from the test images. Finally, the measurement result is obtained based on the geometric properties of the test geometric element. Therefore, it eliminates the need to set measurement parameters and program sets for each image separately. A single interaction on a reference image can be used to batch transfer to a large number of test images, lowering the barrier to building measurement program sets.

[0029] Furthermore, since the reference information extraction and transfer mechanism does not rely on prior assumptions about specific patterns, it can adapt to various types of geometric elements such as line segments and regions, thereby improving the versatility and generalization of the measurement scheme for measuring different patterns and multi-image overlays. In addition, compared with the traditional scheme of adjusting parameters one image at a time, the batch transfer mechanism reduces repetitive parameter configuration operations, thereby reducing the computational and storage overhead of electronic devices during parameter adjustment.

[0030] The following describes various examples of this scheme in further detail with reference to the accompanying drawings.

[0031] Figure 2 A flowchart of an example process 200 for image-based measurement according to some embodiments is shown. Process 200 can be implemented at electronic device 110. Reference is made below. Figure 1 To describe process 200.

[0032] like Figure 2 As shown, in block 210, electronic device 110 acquires a reference image and multiple images under test. The reference image and the multiple images under test correspond to different chip cells in the same wafer.

[0033] As mentioned earlier, a single wafer can contain multiple chip units (e.g., dozens or even hundreds of chip units). The electronic device 110 can acquire multiple chip images 102 corresponding to these multiple chip units. The user can select one chip image as a reference image and use the remaining chip images as images to be measured. It should be understood that the design layouts of different chip units (dies) within the same wafer are completely identical, and the target pattern structure is consistent, with only minor physical deviations caused by the manufacturing process. This is also the condition under which the image to be measured can be measured based on the reference image.

[0034] In box 220, electronic device 110 determines reference information for the target geometric element in the reference image based on a reference image and element information associated with the target geometric element. The reference information includes at least one parameter for extracting the target geometric element. The reference information may include, for example, spatial constraint information, extraction parameters, and morphological features of the target geometric element.

[0035] In some examples, electronic device 110 may receive one or more user inputs to a reference image and determine element information based on these user inputs. In some embodiments, electronic device 110 may determine that user input has been received in response to receiving an interactive operation by a user on the reference image. The interactive operation may include any suitable type of operation such as a box selection operation or a point selection operation. In some embodiments, electronic device 110 may present an input control and receive user input via the input control. By way of example only, the input control may include an input box, via which electronic device 110 may receive user input in the form of natural language text. Of course, electronic device 110 may also receive user input in any other suitable manner.

[0036] In some examples, one or more user inputs include at least a first user input indicating the element type of the target geometric element. The element type can include at least a line type and a zone type. For example, a line-type geometric element refers to a line structure extending along a specific direction in an image, while a zone-type geometric element refers to a closed region structure with continuous brightness characteristics in an image. As an example only, the electronic device 110 can determine that it has received the first user input in response to receiving a annotation operation for a region of interest. The annotation operation can, for example, be used to add annotation text, which can include, for example, the text "line" or the text "zone," visually indicating the element type of the target geometric element. The electronic device 110 can, for example, determine reference information for the target geometric element based on its element type, using a method corresponding to that element type.

[0037] As an example, the methods for determining reference information can differ depending on the element type. For instance, a target geometric element of the line segment type can be determined based on the anchor points associated with that target geometric element, while a target geometric element of the region type can be determined based on at least one positive point and / or at least one negative point.

[0038] In some examples, in addition to the first user input, one or more user inputs may also include a second user input and a third user input. The second user input may indicate the location of the Region of Interest (ROI) in the reference image, the ROI including the target geometric element, and the third user input may indicate the location of the target geometric element within the ROI. As an example only, the electronic device 110 may determine that the second user input has been received in response to receiving a region selection operation on the reference image. The electronic device 110 may determine the ROI in the reference image based on the region selection operation on the reference image. The ROI can have any suitable shape, typically a rectangle. The electronic device 110 then determines the target geometric element from the ROI based on the third user input. It should be noted that, in this case, the reference information for the target geometric element may include the location of the ROI. The reference information may include, but is not limited to, spatial constraint information (e.g., coordinates, orientation, reference line positions, etc.), extraction parameters (e.g., edge detection thresholds, region segmentation rules, etc.), and morphological features of the target geometric element (e.g., masks, fragment indices, etc.). In this way, the electronic device 110 can measure the region of interest in the image. Compared with measuring the entire image, measuring the region of interest can not only improve the accuracy of the measurement, but also reduce the computational load of the device.

[0039] Of course, in some examples, in addition to the first user input, one or more user inputs may include a fourth user input, which may indicate the position of the target geometric element in the reference image. In this case, the electronic device 110 can directly determine the target geometric element from the reference image based on the user input. In this way, the electronic device 110 can directly perform measurements on the reference image, reducing interaction steps and lowering the amount of localization processing.

[0040] In some examples, the electronic device 110 may also determine element information associated with the target geometric element based on predetermined configuration information. As an example, the configuration information may directly indicate the element type, position, etc., of the target geometric element in the reference image. This configuration information may be manually specified by the user or determined automatically by the electronic device 110 based on predetermined rules. The various embodiments described herein do not limit the specific method of determining the element information. The element information determined by the electronic device 110 may at least indicate the element type of the target geometric element. In some examples, the element information may also indicate the position of the region of interest in the reference image, the position of the target geometric element in the region of interest, etc. In some examples, the element information may also directly indicate the position of the target geometric element in the reference image.

[0041] In some embodiments, the electronic device 110 can also determine the edge features of the reference image based on brightness changes in the reference image. For example, the electronic device 110 can determine the brightness changes of the image based on the grayscale values ​​of individual pixels in the image. Edge features include at least one of abrupt brightness changes and gradual brightness changes. An abrupt brightness change edge is an edge where the pixel grayscale value changes instantaneously and abruptly, while a gradual brightness change edge is an edge where the pixel grayscale value changes smoothly and continuously along the vertical boundary direction, gradually transitioning. The electronic device 110 can determine reference information, for example, based on edge features and user interaction. In this way, both abrupt brightness changes and gradual brightness changes can be adaptively identified, improving the accuracy and robustness of feature location recognition and reducing the overhead of repeated searches caused by misjudgments of edge types.

[0042] In block 230, for a test image among multiple test images, electronic device 110 transfers reference information from a reference image to the test image to obtain transferred reference information, the transferred reference information representing at least one parameter in the pixel coordinates of the test image; based on the transferred reference information, extracts test geometric elements that match the target geometric elements from the test image; and based on the geometric properties of the test geometric elements, determines at least a portion of the measurement results of the test geometric elements in the test image.

[0043] In some embodiments, the electronic device 110 can directly determine the geometric element to be measured from the image under test based on reference information from a reference image. In some examples, due to slight differences between different chip units, the reference image and the image under test may also differ. To improve the accuracy of the measurement, the electronic device 110 can first transfer the reference information from the reference image to the image under test to obtain the transferred reference information. Specifically, the electronic device 110 can first perform an image registration operation on the reference image and the image under test to determine the spatial mapping relationship between the reference image and the image under test. The spatial mapping relationship indicates the correspondence between pixel coordinates in the reference image and pixel coordinates in the image under test. Image registration refers to the process of aligning two or more images in space so that the same physical location corresponds to the same pixel coordinates in different images. Image registration is usually achieved by calculating the spatial mapping relationship between images.

[0044] The electronic device 110 can then perform an information transfer operation on the reference information based on the spatial mapping relationship to represent the at least one parameter in the pixel coordinates of the image under test. Based on the transferred reference information, the electronic device 110 can extract the geometric elements to be tested that match the target geometric elements from the image under test. In this way, the reference information on the reference image is automatically mapped to the coordinate system of the image under test, achieving one-time setup and batch reuse. This eliminates the manual positioning steps for each image, reduces invalid searches caused by coordinate misalignment, and lowers computational overhead.

[0045] The electronic device 110 can then determine at least a portion of the measurement results of the geometric element to be measured in the image based on the geometric properties of the geometric element to be measured. In some embodiments, the geometric properties may include the position coordinates, shape parameters, and size information of the geometric element to be measured. The measurement results may include key dimensions, line width roughness, line edge roughness, geometric center of the region, folded area difference rate, fitted roundness, overlay deviation, etc. In some examples, the electronic device 110 can directly measure a reference image to obtain the measurement results of the reference image.

[0046] By employing the above method, reference information is determined based on the baseline image and element information associated with the target geometric elements. This reference information is then batch-transferred to the images under test to automatically extract and match the geometric elements to be measured, eliminating the need to adjust algorithm parameters and create a dedicated set of measurement programs for each image. This reduces the operational complexity and time cost for users to construct sets of measurement programs. Furthermore, since the reference information transfer and matching mechanism is not dependent on specific pattern types, it can adapt to various geometric elements such as line segments and regions, improving the versatility and generalization of the measurement scheme. In addition, the batch transfer mechanism reduces the computational overhead of electronic devices during repetitive parameter configuration, shortening the processing latency of batch images.

[0047] In some examples, the target geometry may include a first geometry element of type line segment. Figure 3A A flowchart of a measurement method 300A for a first geometric element of a line segment type according to some embodiments is shown. Process 300A can be implemented at electronic device 110. Reference is made below. Figure 1 To describe process 300A.

[0048] In some examples, element information can indicate the positions of at least two anchor points, which are located near the first geometric element in the reference image. For example... Figure 3A As shown in box 301, electronic device 110 determines at least two anchor points for a reference image based on element information. Taking only the example that the element information is received based on user input, in some embodiments, the user can mark at least two anchor points in the region of interest of the reference image, near the line segment to be measured, by a point selection operation.

[0049] In block 302, electronic device 110 verifies at least two anchor points. In some embodiments, electronic device 110 performs a feature location recognition operation for each anchor point based on edge features of a reference image. As mentioned above, the edge features may be located at edges of abrupt changes in brightness or edges of gradual changes in brightness. Electronic device 110 may determine the feature location corresponding to the anchor point on the edge feature along a direction perpendicular to the line direction of the first geometric element. In some examples, the feature location on an edge of abrupt changes in brightness may be a brightness peak location, and the feature location on an edge of gradual changes in brightness may be a gradient extremum location.

[0050] Taking anchor point A with coordinates (x1, y1) and the line direction of the first geometric element as the X direction as an example, the feature position corresponding to anchor point A on the edge feature can be (x1, y2). This feature position has the same x-coordinate as anchor point A but a different y-coordinate. This process can be called the snapping process of anchor points to edge features. In some examples, electronic device 110 can determine that the two anchor points have passed the verification in response to the successful snapping of at least two anchor points to the edge feature of the first geometric element. Electronic device 110 can also present a prompt message in response to the determination that there is an anchor point among the at least two anchor points that cannot be snapped to the feature position of the first geometric element, to guide the user to re-mark the anchor points. Thus, anchor points can be verified, allowing electronic device 110 to perform subsequent processing based on the verified anchor points, thereby improving the accuracy of subsequent processing.

[0051] In block 303, electronic device 110 determines the extension direction of the first geometric element as a first direction based on the relative position between at least two anchor points. In some embodiments, if the distribution span of the at least two anchor points in the horizontal direction is greater than the distribution span in the vertical direction, the first direction is determined to be the horizontal direction (also referred to as the X direction); otherwise, the first direction is determined to be the vertical direction (also referred to as the Y direction). Of course, this is just an example.

[0052] In block 304, electronic device 110 determines a first reference line based on the coordinates of at least two anchor points in a second direction, the second direction being perpendicular to the first direction. The first reference line is directed to a first geometric element. In some embodiments, taking the first direction as the X direction and the second direction as the Y direction as an example, electronic device 110 calculates the average of the coordinates of at least two anchor points in the Y direction (i.e., the y-coordinates of each of the at least two anchor points), and determines the line corresponding to the average value as the first reference line. For example, if the at least two anchor points include four anchor points: anchor point A (x1, y1), anchor point B (x2, y2), anchor point C (x3, y3), and anchor point D (x4, y4), the average value of the coordinates of these four anchor points in the second direction is y... 平 = (y1 + y2 + y3 + y4) / 4, the first reference line is y = y 平 This first reference line serves as the starting point for subsequent point-by-point searching of feature locations along the line.

[0053] In block 305, electronic device 110 determines multiple feature positions belonging to a first geometric element based on a first reference line and a predetermined range extending perpendicular to the first reference line. In some embodiments, electronic device 110 traverses each coordinate position within the region of interest along a first direction (X direction), and for each coordinate position, searches for feature positions conforming to edge features within a predetermined range, starting from the position of the first reference line in a second direction (Y direction).

[0054] As an example only, for the coordinate position x=1, electronic device 110 can use the point corresponding to this coordinate position on the first reference line (e.g., point (1, y)). 平 Starting from a predetermined point, the electronic device 110 searches for feature locations within a predefined range. This predefined range refers to feature locations in the Y direction. Taking a predefined unit of 3 as an example, the electronic device 110 searches for feature locations within the range (1, y). 平 +3) to (1, y) 平 -3) Search for feature locations on edge features within this range. For example, electronic device 110 can search for (1, y 平 +1) is the feature location. The electronic device 110 can progressively search for each coordinate position with x = 1, 2, 3... based on this method.

[0055] The electronic device 110 can record each feature location found, and can mark the coordinates of the feature locations that are outside a predetermined range and for which no valid feature location was found as abnormal coordinate points. For example, if no feature location was found at the coordinate x=4, the electronic device 110 can mark this coordinate location as an abnormal coordinate point.

[0056] In block 306, electronic device 110 determines multiple smooth line segments based on multiple feature locations. In some embodiments, for the multiple feature locations, electronic device 110 may traverse the multiple feature locations along a first direction and calculate the positional standard deviation of each feature location and a predetermined number (e.g., K, where K is any suitable positive integer) of other feature locations in a second direction. For example, for the i-th feature location, electronic device 110 may calculate the positional standard deviation (i.e., the standard deviation of the Y coordinates of these 2K+1 feature locations) of the feature location, the K feature locations before the feature location, and the K feature locations after the feature location (a total of 2K+1 feature locations) in the Y direction. Electronic device 110 may determine that these 2K+1 feature locations are in the same smooth line segment in response to determining that the standard deviation is lower than a preset threshold (e.g., a first threshold).

[0057] In some examples, the electronic device 110 can also directly determine the smooth line segment based on the difference between the y-coordinates of different feature locations. For example, if the difference between the y-coordinates of feature location A and feature location B in the Y direction is less than a certain threshold (e.g., a first threshold), it is determined that these two feature locations are on the same smooth line segment. The electronic device 110 can also use other methods to traverse multiple feature locations and obtain multiple smooth line segments based on the multiple feature locations. The electronic device 110 can determine the smooth line segment corresponding to a set of feature locations that are on the same smooth line segment.

[0058] For adjacent smooth line segments among multiple smooth line segments, the electronic device 110 can further determine whether the adjacent smooth line segments meet predetermined conditions. The predetermined conditions may indicate that the spacing between adjacent smooth line segments in a first direction (i.e., the distance between the x-coordinate of the end feature position of the preceding smooth line segment and the x-coordinate of the beginning feature position of the following smooth line segment) is less than a second threshold, and the position difference in a second direction (i.e., the distance between the y-coordinate of the end feature position of the preceding smooth line segment and the y-coordinate of the beginning feature position of the following smooth line segment) is less than a third threshold. In response to determining that two adjacent smooth line segments meet this predetermined condition, the electronic device 110 can merge these two smooth line segments into a single smooth line segment. This merging mechanism can address line segment breaks caused by local noise or crosstalk on the line. The electronic device 110 can thus obtain at least one merged smooth line segment.

[0059] In box 307, electronic device 110 determines index information for multiple smooth line segments. The index information indicates the position of at least one merged smooth line segment in a first direction; for example, the index information may indicate the start and end coordinates of at least one smooth line segment in the first direction. Alternatively, the index information may indicate the coordinates of individual feature positions on each of the at least one smooth line segment. The index information can be used as part of the reference information for subsequent extraction of smooth line segments from the image under test. By selecting reliable smooth line segment fragments as reference information, the occlusion, intersection, and distortion effects caused by other lines or region features on the lines can be avoided.

[0060] In box 308, electronic device 110 determines reference information for a first geometric element in the reference image. The reference information may include the coordinates of at least two anchor points, a first direction, a first reference line, index information of multiple smooth line segments, and edge feature types, etc.

[0061] In box 309, electronic device 110 determines the geometric element to be measured from the image to be measured based on a reference image and transferred reference information. In some examples, the reference information may indicate a first direction and a first reference line. It is understood that the reference information may directly include the first reference line, or it may directly include at least two anchor points, and electronic device 110 may calculate the first reference line based on at least two anchor points. The transferred reference information may indicate a transferred first direction and a transferred first reference line.

[0062] In some embodiments, electronic device 110 can determine at least one index point from the image under test based on migrated index information and a migrated first direction. Electronic device 110 then determines a second reference line for the image under test based on a migrated first reference line. The second reference line can be considered as a corresponding reference line at the corresponding position of the first reference line on the image under test. Electronic device 110 can determine at least one target point corresponding to each of the at least one index point on the second reference line, where each index point and its corresponding target point have the same coordinates in the first direction, and can determine multiple target feature positions of the geometric element under test based on at least one target point and a predetermined range extending perpendicular to the second reference line. This step can refer to step 305 described above. The multiple target feature positions can be located at the edge features of the geometric element under test. Electronic device 110 can obtain the geometric element under test based on the multiple target feature positions; for example, electronic device 110 can connect the multiple target feature positions and determine the resulting line segment as the geometric element under test. Since the search is performed only on the available segment locations indicated by the index information in the image under test, invalid searches are avoided in areas that have been identified as unreliable in the reference image, thus improving the accuracy and efficiency of target line segment extraction in the image under test.

[0063] In block 310, electronic device 110 determines the measurement results of the geometric elements to be measured in the image based on their geometric properties. Electronic device 110 can determine the geometric properties of the geometric elements to be measured in any suitable manner and can determine the measurement results based on these properties. In some embodiments, the measurement results may include the average value of the position coordinates of the geometric elements to be measured (i.e., line position), line width roughness (LWR), line edge roughness (LER), etc.

[0064] By employing the above method, for target geometric elements of the line segment type, reference lines are determined through anchor point snapping. Then, feature positions are searched point-by-point along the line direction, and stationarity assessment and segment merging are performed. This allows for reliable extraction of the target line segment's positional information from SEM images containing noise and cross-interference. This improves the accuracy and robustness of line segment extraction. Furthermore, since the search is performed only at the index positions of available segments, the computational overhead of invalid searches is reduced, shortening the processing latency of a single image.

[0065] In some examples, the target geometry may include a second geometry of region type. Figure 3B A flowchart of a measurement method 300B for a second geometric element of a region type according to some embodiments is shown. Process 300B can be implemented at electronic device 110. Reference is made below. Figure 1 To describe process 300B.

[0066] like Figure 3B As shown in box 311, electronic device 110 segments the reference image to obtain multiple first candidate geometric elements. In some embodiments, region segmentation can employ a seed-point-based region growing method or a trained image segmentation model (e.g., an image segmentation base model). When using a region growing method, electronic device 110 can determine the brightness gradient distribution within the region of interest of the reference image and select the location of the local minimum brightness gradient as the seed location (these locations are typically located at the center of the region to be extracted). Electronic device 110 can build a queue starting from the seed location and, during the growing process, continuously calculate the mean gradient μ and standard deviation σ within the currently grown region. Electronic device 110 can merge locations in the neighborhood where the brightness change is less than the growth threshold into the region where the seed location is located, based on a predetermined growth threshold T, until a preset growth stopping condition is met. The growth threshold T can, for example, be μ+k. σ (where k is an empirical coefficient that decreases with the number of growth rounds).

[0067] In some embodiments, during region growth, if a location in the neighborhood has a brightness gradient greater than or equal to a growth threshold, the electronic device 110 can trigger a boundary verification mechanism. Specifically, the electronic device 110 can calculate the consistency coefficient between the principal gradient direction at that location and the contour normal direction of the current first candidate geometric element; if the consistency coefficient is greater than or equal to a preset direction threshold, the location is determined to be a true boundary break point, and branch growth in that direction is stopped; if the consistency coefficient is less than the direction threshold, the location is determined to be isolated noise or a false edge, allowing growth to continue across that location and updating local statistical parameters. The electronic device 110 can, for example, calculate the consistency coefficient C based on the following formula:

[0068] Where g is a pixel in a certain neighborhood of pixel p, N is the number of these pixels, and the subscripts x and y represent the brightness gradients in the X and Y directions, respectively.

[0069] Electronic device 110 can obtain multiple regions in this manner. Electronic device 110 can also filter these multiple regions. For example, electronic device 110 can discard regions with areas below a minimum effective threshold or with abnormal aspect ratios. Electronic device 110 can obtain multiple regions and identify these multiple regions as multiple first candidate geometric elements. In this way, it is possible to distinguish between true boundaries and noise or false edges, improving the accuracy and robustness of region segmentation. Electronic device 110 can also obtain multiple masks corresponding to each of the multiple first candidate geometric elements. Simultaneously, it avoids premature termination of growth due to noise, reduces missed segmentation, and lowers the computational cost of subsequent correction.

[0070] When using a trained image segmentation model, the electronic device 110 can use the trained image segmentation model to segment a reference image to obtain multiple first candidate geometric elements and multiple masks corresponding to the multiple first candidate geometric elements.

[0071] In some examples, element information may indicate at least one of at least a positive point and at least one negative point, with the at least positive point located inside the second geometric element and the at least one negative point located outside the second geometric element. In block 312, electronic device 110 determines a second geometric element from a plurality of first candidate geometric elements based on the element information. By way of example only, electronic device 110 may determine the element information based on a region marking operation on a reference image. The region marking operation may, for example, include at least one of a positive point marking operation and a negative point marking operation. The positive point marking operation is used to mark at least one positive point inside the second geometric element, and the negative point marking operation is used to mark at least one negative point outside the second geometric element. Electronic device 110 displays the first candidate geometric elements obtained from region segmentation to a user, who defines the extent of the target region by clicking on a positive point inside the target region and a negative point outside the target region. Electronic device 110 may then determine the second geometric element based on at least one positive point and at least one negative point. For example, electronic device 110 may determine an edge feature located between at least one positive point and at least one negative point and identify that edge feature as an edge feature of the second geometric element.

[0072] In frame 313, electronic device 110 acquires at least a portion of the first mask of the second geometric element as reference information. The first mask records the shape and position information of the target region determined by the user on the reference image in the form of a binary image.

[0073] In box 314, the electronic device 110 performs region segmentation on the image to be tested, obtaining multiple second candidate geometric elements and corresponding multiple second masks. In some embodiments, the electronic device 110 first performs region segmentation on the image to be tested to obtain multiple third candidate geometric elements. Similar to the reference image, the electronic device 110 may employ a seed point-based region growing method or a trained image segmentation model to perform image segmentation. The electronic device 110 may, for example, perform processing on the multiple third candidate geometric elements to obtain multiple second candidate geometric elements, and the processing may include at least one of region merging, overlap extraction, region culling, and common part removal. In some embodiments, the electronic device 110 constructs a spatial adjacency graph using the bounding rectangles of the bounding boxes of each basic mask as nodes; calculates the overlap rate and centroid Euclidean distance between any two node masks, and establishes a connected edge if a preset adjacency condition is met (overlap rate greater than a spatial overlap threshold or centroid distance less than an adjacency tolerance).

[0074] Using connected edges as a reference, electronic device 110 can perform Boolean algebra operations on adjacent masks, including union, intersection, difference, and symmetric difference operations. The union operation merges two regions into one; the intersection operation extracts the common portion of two regions into a single region; the difference operation removes the portion of the subtrahend region from the minuend region, leaving only one region; and the symmetric difference operation removes the common portion from the merged region, leaving only one region. Electronic device 110 can verify the validity of the operation results, removing invalid masks with zero area or degraded shape, to obtain multiple second candidate geometric elements and their corresponding second masks.

[0075] The transferred reference information may include, for example, a transferred first mask. In block 315, the electronic device 110 selects the geometric element to be tested from a plurality of second candidate geometric elements based on the similarity between the transferred first mask and each of the plurality of second masks. In some embodiments, the electronic device 110 first eliminates the second masks that have a large relative difference from the first mask in terms of area and perimeter, leaving the remaining masks as candidate masks. The electronic device 110 can calculate the geometric shape similarity (first similarity) between the candidate masks and the first mask, for example, by calculating the difference in Hu invariant moments between different masks. The electronic device 110 can eliminate candidate masks with a proportion lower than the highest similarity threshold. Further, the electronic device 110 can extract the pixel grayscale histograms of the remaining candidate mask region and the first mask region respectively, calculate the grayscale distribution similarity (second similarity), i.e., the overlap between the two, and eliminate candidate masks with a proportion lower than the highest overlap threshold. Electronic device 110 can normalize the geometric and photometric features of each candidate mask and then concatenate them into a comprehensive feature vector. Cosine similarity or Bhattacharyya distance is used to measure the similarity between the comprehensive feature vectors, and the candidate geometric element corresponding to the optimal one is selected as the geometric element to be tested. This improves the accuracy and robustness of region matching and reduces false matches caused by insufficient single-dimensional similarity to distinguish candidate regions.

[0076] In box 316, electronic device 110 determines the measurement result of the geometric element to be measured in the image based on the geometric attributes of the geometric element to be measured. Electronic device 110 can determine the geometric attributes of the geometric element to be measured in any appropriate manner, and can determine the measurement result based on the geometric attributes. In some embodiments, the geometric attributes of the region type may include key points (top, bottom, left, right), geometric center, folded area difference rate, fitted roundness, and fitted ellipticity, etc.

[0077] By employing the above method, for target geometric elements of a region type, candidate regions are extracted through seed point region growth and boundary verification mechanisms. The target region is then confirmed by marking positive and negative points. Boolean operations are used to expand the candidate set, and a step-by-step selection process using geometric and photometric dual-dimensional similarity is employed. This allows for accurate matching of the geometric elements in the test image that are most similar in shape to the target region in the reference image. This improves the accuracy and recall of region matching. Furthermore, the step-by-step selection strategy eliminates mismatched candidate regions in order of increasing computational complexity, avoiding high-complexity similarity calculations for all candidate regions and reducing overall computational overhead and processing latency.

[0078] In some embodiments, after obtaining the target line segment or region-type geometric elements to be measured, the user can obtain the secondary calculation results by setting a calculation formula. For example, based on the positional relationship between two line segment-type geometric elements to be measured, the critical dimension (CD) or overlay deviation (OVL) value can be calculated. In this way, the calculation of the relationship between multiple elements can be supported, meeting various measurement needs of users. As a result, there is no need to develop separate algorithm modules for different measurement indicators, reducing the system's storage overhead and maintenance costs.

[0079] A corresponding apparatus for implementing the above methods or processes is also provided.

[0080] Figure 4 Block diagrams of an apparatus 400 for image-based measurement according to some embodiments are shown. Apparatus 400 may be implemented as or included in electronic device 110. Various modules / components in apparatus 400 may be implemented by hardware, software, firmware, or any combination thereof.

[0081] like Figure 4 As shown, the apparatus 400 includes an image acquisition module 410 configured to acquire a reference image and multiple test images, the reference image and the multiple test images corresponding to different chip cells in the same wafer; an information determination module 420 configured to determine reference information of a target geometric element in the reference image based on the reference image and element information associated with the target geometric element, the reference information including at least one parameter for extracting the target geometric element; a result determination module 430 configured to transfer the reference information in the reference image to the test image among the multiple test images to obtain transferred reference information, the transferred reference information representing at least one parameter in the pixel coordinates of the test image; extract test geometric elements matching the target geometric element from the test image based on the transferred reference information; and determine at least a portion of the measurement results of the test geometric elements in the test image based on the geometric properties of the test geometric elements.

[0082] In some embodiments, the element information at least indicates the element type of the target geometric element, including a line segment type or a region type.

[0083] In some embodiments, the element information also indicates the following: the location of the region of interest in the reference image, the region of interest including target geometric elements; and the location of the target geometric elements in the region of interest.

[0084] In some embodiments, the element information also indicates the location of the target geometric element in the reference image.

[0085] In some embodiments, the information determination module 420 may also be configured to present a reference image; determine a region of interest in the reference image based on a region selection operation on the reference image, the region of interest including a target geometric element; receive a labeling operation on the region of interest, the labeling operation indicating the element type of the target geometric element; and determine reference information of the target geometric element based on user interaction on the region of interest and the element type of the target geometric element.

[0086] In some embodiments, the information determination module 420 may also be configured to determine edge features of the reference image based on brightness changes in the reference image, the edge features including at least one of the following: abrupt brightness change edges in the reference image, a gradual brightness change edges in the reference image; and determine reference information based on the edge features and element information.

[0087] In some embodiments, the result determination module 430 may also be configured to perform an image registration operation on a reference image and a test image to determine a spatial mapping relationship between the reference image and the test image; and to perform an information migration operation on reference information based on the spatial mapping relationship to represent at least one parameter in the pixel coordinates of the test image.

[0088] In some embodiments, the target geometric element includes a first geometric element of the line segment type, and the element information indicates the positions of at least two anchor points, which are located near the first geometric element in the reference image. The information determination module 420 may also be configured to: determine at least two anchor points based on the element information; and determine description information of at least one line segment based on the at least two anchor points, as at least a part of the reference information of the first geometric element.

[0089] In some embodiments, the information determination module 420 may also be configured to determine the extension direction of the first geometric element as a first direction based on the relative position between at least two anchor points; determine a first reference line based on the coordinates of the at least two anchor points in a second direction; determine multiple feature positions belonging to the first geometric element based on the first reference line and a predetermined range extending perpendicular to the first reference line; and determine descriptive information of at least one line segment based on the multiple feature positions.

[0090] In some embodiments, multiple feature locations may be determined based on edge features of a reference image, and the multiple feature locations include at least one of the following: the brightness peak location of a brightness abrupt change edge in the reference image; and the gradient extremum location of a brightness gradient edge in the reference image.

[0091] In some embodiments, the information determination module 420 may also be configured to determine, for each of at least two anchor points, whether a corresponding feature location is identified based on that anchor point; and to present a prompt message in response to the existence of an anchor point among the at least two anchor points where a feature location cannot be identified, the prompt message being used to guide the user to remark the anchor point.

[0092] In some embodiments, the information determination module 420 may also be configured to determine multiple smooth line segments based on multiple feature locations; merge adjacent smooth line segments into one smooth line segment in response to adjacent smooth line segments meeting predetermined conditions; and determine index information of at least one smooth line segment as descriptive information.

[0093] In some embodiments, the result determination module 430 may also be configured to determine at least one index point and a second reference line for the image under test from the image under test based on the transferred reference information; determine at least one target point corresponding to each of the at least one index point on the second reference line; determine multiple target feature positions of the geometric element under test based on the at least one target point and a predetermined range extending perpendicular to the second reference line; and determine the geometric element under test based on the multiple target feature positions.

[0094] In some embodiments, the information determination module 420 may also be configured to segment the reference image to obtain a plurality of first candidate geometric elements; determine a second geometric element from the plurality of first candidate geometric elements based on element information; and obtain a first mask of the second geometric element as at least a part of the reference information of the second geometric element.

[0095] In some embodiments, the element information indicates at least one of at least a positive point and at least one negative point, wherein at least one positive point is located inside the second geometric element and at least one negative point is located outside the second geometric element. The information determination module 420 may also be configured to determine the second geometric element from a plurality of first candidate geometric elements based on at least one of at least a positive point and at least one negative point.

[0096] In some embodiments, the information determination module 420 may also be configured to determine the brightness distribution of a reference image; select an image location whose brightness change meets a predetermined requirement as a seed location based on the brightness distribution; merge locations in the neighborhood of the seed location whose brightness change is less than a threshold into the region where the seed location is located, starting from the seed location, until a preset growth stopping condition is met; and obtain one of a plurality of first candidate geometric elements based on the region corresponding to the seed location.

[0097] In some embodiments, the information determination module 420 may also be configured to segment a reference image using a trained image segmentation model to obtain a plurality of first candidate geometric elements.

[0098] In some embodiments, the migrated reference information includes a migrated first mask, and the result determination module 430 can also be configured to perform region segmentation on the image to be tested to obtain a plurality of second candidate geometric elements and a plurality of second masks corresponding to the plurality of second candidate geometric elements; and to select the geometric element to be tested from the plurality of second candidate geometric elements based on the similarity between the migrated first mask and each of the plurality of second masks.

[0099] In some embodiments, the information determination module 420 may also be configured to determine descriptive information based on the at least two feature locations in response to the identification of at least two corresponding feature locations at at least two anchor points.

[0100] In some embodiments, the information determination module 420 may also be configured to, in response to the presence of a position in the neighborhood with a brightness gradient greater than or equal to a threshold, determine the consistency coefficient between the gradient direction of the position and the contour normal direction of the first candidate geometric element; stop merging positions into the first candidate geometric element in response to the consistency coefficient being greater than or equal to a direction threshold; and merge the position into the first candidate geometric element in response to the consistency coefficient being less than a direction threshold.

[0101] In some embodiments, the result determination module 430 may also be configured to perform region segmentation on the image to be tested to obtain a plurality of third candidate geometric elements; perform processing on the plurality of third candidate geometric elements to obtain a plurality of second candidate geometric elements, the processing including at least one of region merging, overlap extraction, region removal, and common part removal; and obtain a plurality of second masks corresponding to the plurality of second candidate geometric elements respectively.

[0102] In some embodiments, the similarity includes a first similarity and a second similarity, wherein the first similarity is the geometric shape similarity between the plurality of second masks and the first mask, and the second similarity is the gray-scale distribution similarity between the plurality of second masks and the first mask.

[0103] The modules included in device 400 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more modules may be implemented using software and / or firmware, such as machine-executable instructions stored on a storage medium. In addition to or as an alternative to machine-executable instructions, some or all of the units in device 400 may be implemented at least partially by one or more hardware logic components. By way of example and not limitation, exemplary types of hardware logic components that may be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), and so on.

[0104] Figure 5 A block diagram of an electronic device 500 in which one or more examples may be implemented is shown. It should be understood that... Figure 5 The electronic device 500 shown is merely exemplary and should not be construed as limiting the functionality and scope of the examples described herein. Figure 5 The electronic device 500 shown can be used to implement the electronic device 110 discussed above.

[0105] like Figure 5 As shown, electronic device 500 is in the form of a general-purpose electronic device. Components of electronic device 500 may include, but are not limited to, one or more processing units or processors 510, memory 520, storage devices 530, one or more communication units 540, one or more input devices 550, and one or more output devices 560. Processor 510 may be a physical or virtual processor and is capable of performing various processes according to programs stored in memory 520. In a multiprocessor system, multiple processors execute computer-executable instructions in parallel to improve the parallel processing capability of electronic device 500.

[0106] Electronic device 500 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 500, including but not limited to volatile and non-volatile media, removable and non-removable media. Memory 520 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof). Storage device 530 can be removable or non-removable media and can include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and / or data and can be accessed within electronic device 500.

[0107] Electronic device 500 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not explicitly stated... Figure 5 As shown, disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks can be provided. In these cases, each drive can be connected to a bus (not shown) via one or more data media interfaces. Memory 520 may include computer program product 525 having one or more program modules configured to perform various methods or actions of various examples.

[0108] The communication unit 540 enables communication with other electronic devices via a communication medium. Additionally, the functionality of the components of the electronic device 500 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the electronic device 500 can operate in a networked environment using logical connections to one or more other servers, networked personal computers, or another network node.

[0109] Input device 550 can be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 560 can be one or more output devices, such as a monitor, speaker, printer, etc. Electronic device 500 can also communicate with one or more external devices (not shown) via communication unit 540 as needed. These external devices include storage devices, display devices, etc., and can communicate with one or more devices that enable user interaction with electronic device 500, or with any device that enables electronic device 500 to communicate with one or more other electronic devices (e.g., network card, modem, etc.). Such communication can be performed via an input / output (I / O) interface (not shown).

[0110] A computer-readable storage medium is provided that stores computer-executable instructions thereon, wherein the computer-executable instructions are executed by a processor to implement the methods described above. A computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which are executed by a processor to implement the methods described above.

[0111] The flowcharts and / or block diagrams of the methods, apparatus, devices, and computer program products referred to herein describe various aspects. It should be understood that each block of the flowcharts and / or block diagrams, as well as combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0112] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0113] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0114] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to some embodiments. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0115] Various examples have been described above. The foregoing descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. An image-based measurement method, comprising: Acquire a reference image and multiple images to be tested, wherein the reference image and the multiple images to be tested correspond to different chip cells in the same wafer; Based on the reference image and element information associated with the target geometric element, reference information of the target geometric element in the reference image is determined, the reference information including at least one parameter for extracting the target geometric element; as well as For the image to be tested among the multiple images to be tested, The reference information in the reference image is transferred to the image under test to obtain transferred reference information, wherein the transferred reference information represents the at least one parameter in the pixel coordinates of the image under test; Based on the transferred reference information, the geometric elements to be tested that match the target geometric elements are extracted from the image to be tested; as well as Based on the geometric properties of the geometric element to be measured, at least a portion of the measurement results of the geometric element to be measured in the image to be measured is determined.

2. The method according to claim 1, wherein the element information at least indicates the element type of the target geometric element, the element type including line segment type or region type.

3. The method of claim 2, wherein the element information further indicates the following: The region of interest is located in the reference image, and the region of interest includes the target geometric element; The location of the target geometric element in the region of interest.

4. The method of claim 2, wherein the element information further indicates the position of the target geometric element in the reference image.

5. The method of claim 1, wherein determining the reference information comprises: Based on the brightness changes in the reference image, the edge features of the reference image are determined, and the edge features include at least one of the following: abrupt brightness change edges in the reference image, and gradual brightness change edges in the reference image; as well as The reference information is determined based on the edge features and the element information.

6. The method of claim 1, wherein transferring the reference information from the reference image to the image to be tested comprises: An image registration operation is performed on the reference image and the image to be tested to determine the spatial mapping relationship between the reference image and the image to be tested. The spatial mapping relationship indicates the correspondence between pixel coordinates in the reference image and pixel coordinates in the image to be tested. as well as An information migration operation is performed on the reference information based on the spatial mapping relationship to represent the at least one parameter in the pixel coordinates of the image under test.

7. The method of claim 1, wherein the target geometric element comprises a first geometric element of line segment type, the element information indicating the positions of at least two anchor points located near the first geometric element in the reference image. The reference information for determining the target geometric element includes: Based on the element information, at least two anchor points are determined; as well as Based on the at least two anchor points, descriptive information of at least one line segment is determined as at least a part of the reference information of the first geometric element.

8. The method of claim 7, wherein determining the description information of at least one line segment based on the at least two anchor points includes: Based on the relative position between the at least two anchor points, the extension direction of the first geometric element is determined as the first direction; Based on the coordinates of the at least two anchor points in the second direction, a first reference line is determined for the first geometric element, wherein the second direction is perpendicular to the first direction; Based on the first reference line and a predetermined range extending perpendicular to the first reference line, multiple feature positions belonging to the first geometric element are determined. as well as The descriptive information of the at least one line segment is determined based on the multiple feature locations.

9. The method of claim 8, wherein the plurality of feature locations are determined based on edge features of the reference image, and the plurality of feature locations include at least one of the following: The location of the brightness peak at the brightness abrupt change edge in the reference image; and The gradient extremum location of the brightness gradient edge in the reference image.

10. The method of claim 7, further comprising: For each of the at least two anchor points, determine whether a corresponding feature location has been identified based on that anchor point; as well as In response to the presence of an anchor point whose feature location cannot be identified among the at least two anchor points, a prompt message is displayed to guide the user to remark the anchor point.

11. The method of claim 8, wherein determining the descriptive information based on the plurality of feature locations comprises: Multiple smooth line segments are determined based on the multiple feature positions, and the positional difference of each feature position included in each of the multiple smooth line segments in the second direction is less than a first threshold. In response to adjacent smooth line segments among the plurality of smooth line segments satisfying a predetermined condition, the adjacent smooth line segments are merged into one smooth line segment to obtain at least one merged smooth line segment, wherein the predetermined condition indicates that the spacing between the adjacent smooth line segments in the first direction is less than a second threshold and the position difference between the adjacent smooth line segments in the second direction is less than a third threshold. as well as The index information of the at least one smooth line segment is determined as the description information, and the index information is used to indicate the position of the at least one smooth line segment in the first direction.

12. The method of claim 11, wherein the reference information further indicates the first direction and the first reference line for the reference image, and determining the test geometry element matching the target geometry element from the test image comprises: Based on the migrated reference information, at least one index point and a second reference line for the image under test are determined from the image under test. Determine at least one target point corresponding to each of the at least one index point on the second reference line, wherein the coordinates of each index point and the corresponding target point are the same in the first direction; Based on the at least one target point and a predetermined range extending perpendicular to the second reference line, the positions of multiple target features of the geometric element to be measured are determined; as well as The geometric element to be measured is determined based on the locations of the multiple target features.

13. The method of claim 1, wherein the target geometric element comprises a second geometric element of a region type, and the reference information for determining the target geometric element comprises: The reference image is segmented to obtain multiple first candidate geometric elements; Based on the element information, the second geometric element is determined from the plurality of first candidate geometric elements; as well as Obtain the first mask of the second geometric element as at least a part of the reference information of the second geometric element.

14. The method of claim 13, wherein the element information indicates at least one of at least a positive point and at least one negative point, the at least one positive point being located inside the second geometric element and the at least one negative point being located outside the second geometric element; Determining the second geometric element from the plurality of first candidate geometric elements includes: The second geometric element is determined from the plurality of first candidate geometric elements based on at least one of the at least one positive point and the at least one negative point.

15. The method of claim 13, wherein region segmentation of the reference image comprises: Determine the brightness distribution of the reference image, the brightness distribution indicating the variation of brightness in the reference image; Based on the brightness distribution, image locations whose brightness changes meet predetermined requirements are selected as seed locations; Starting from the seed position, merge positions in the neighborhood of the seed position whose brightness changes less than a threshold into the region where the seed position is located, until a preset growth cessation condition is met; as well as Based on the region corresponding to the seed position, one of the plurality of first candidate geometric elements is obtained.

16. The method of claim 13, wherein region segmentation of the reference image comprises: The reference image is segmented using a trained image segmentation model to obtain the plurality of first candidate geometric elements.

17. The method of claim 13, wherein the transferred reference information includes a transferred first mask, and determining a test geometry element from the test image that matches the target geometry element comprises: The image to be tested is segmented into regions to obtain multiple second candidate geometric elements and multiple second masks corresponding to the multiple second candidate geometric elements; as well as Based on the similarity between the transferred first mask and each of the plurality of second masks, the geometric element to be tested is selected from the plurality of second candidate geometric elements.

18. An image-based measurement device, comprising: The image acquisition module is configured to acquire a reference image and multiple images to be tested, wherein the reference image and the multiple images to be tested correspond to different chip units in the same wafer; An information determination module is configured to determine reference information of the target geometric element in the reference image based on the reference image and element information associated with the target geometric element, the reference information including at least one parameter for extracting the target geometric element; as well as The result determination module is configured to transfer the reference information in the reference image to the image under test among the plurality of images under test, so as to obtain transferred reference information, wherein the transferred reference information represents the at least one parameter in the pixel coordinates of the image under test; Based on the transferred reference information, the geometric elements to be tested that match the target geometric elements are extracted from the image to be tested; And based on the geometric properties of the geometric element to be measured, determine at least a portion of the measurement results of the geometric element to be measured in the image to be measured.

19. An electronic device comprising: At least one processor; as well as At least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions causing the electronic device to perform the method according to any one of claims 1 to 17 when executed by the at least one processor.

20. A computer-readable storage medium having stored thereon computer-executable instructions that can be executed by a processor to implement the method according to any one of claims 1 to 17.

21. A computer program product comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 17.

Images