Electron Microscopic Imaging Measurement Method, Device, Equipment, Medium and Program Product

Abstract

This application relates to the field of semiconductor measurement technologies, and particularly to an electron microscopy imaging measurement method, apparatus, device, medium, and program product. The method includes: obtaining a detection image of a target object to be processed; processing the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image; determining target information of a target region in the template image based on a preset image analysis algorithm, where the target region includes a measurement target and the edge of the measurement target; generating a measurement target box corresponding to the target region according to the target information, applying the measurement target box to the detection image to determine a local region containing the target object in the detection image, and imaging the local region to achieve measurement of the target object. Using this method can improve the overall efficiency and accuracy of the measurement process.

Description

Technical Field

[0001] The present application relates to the field of semiconductor measurement technologies, and particularly to an electron microscopy imaging measurement method, apparatus, device, medium, and program product. Background Art

[0002] An electron microscope is an important instrument used for measuring critical dimensions in the semiconductor manufacturing process. It can measure the pattern dimensions formed during the lithography process with high precision, which is crucial for ensuring the performance and stability of semiconductor chips. The main functions and applications of an electron microscope include: high-resolution imaging, critical dimension measurement, defect detection, process control, and three-dimensional measurement. An electron microscope is used as a front-end inspection device in the field of semiconductor integrated circuits, and can achieve the fast and accurate autofocus function of the device, providing image quality guarantee for subsequent functions such as visual positioning, critical dimension measurement, and defect detection.

[0003] In related technologies, measurement algorithms in electron microscopy imaging detection are a series of algorithms used for measuring and analyzing various features and parameters on a semiconductor wafer. These algorithms are crucial in the semiconductor manufacturing process, especially in wafer quality control, defect detection, and process optimization. Among them, common algorithms and technologies mainly include image processing algorithms, geometric measurement, statistical analysis, model-based algorithms, measurement algorithms of optical and other types of sensing technologies, etc. Currently, most measurement algorithms are based on traditional algorithms, and technicians need to manually set recipes according to the set imaging conditions to locate the measurement positions and measurement parameters.

[0004] However, the current measurement methods have the following technical problems: the existing measurement methods are inefficient and have poor accuracy, and need to be optimized. Summary of the Invention

[0005] Based on this, it is necessary to provide an electron microscopy imaging measurement method, apparatus, device, medium, and program product that can improve the efficiency and accuracy of measuring electron microscopy images in view of the above technical problems.

[0006] In a first aspect, the present application provides an electron microscopy imaging measurement method. The method includes:

[0007] Obtain a detection image of a target object to be processed;

[0008] Process the detection image based on a pre-constructed image conversion model to obtain a template image that matches the detection image;

[0009] Determine target information of a target region in the template image based on a preset image analysis algorithm, where the target region includes a measurement target and the edge of the measurement target;

[0010] Generate a measurement target box corresponding to the target area according to the target information, apply the measurement target box to the detection image to determine a local area containing the target object in the detection image, and image the local area to measure the target object.

[0011] In one embodiment, determining the target information of the target area in the template image based on a preset image analysis algorithm, the target area includes a measurement target and a measurement target edge, including:

[0012] Traverse the template image, and determine several connected domains in the template image based on the pixel value differences in the template image;

[0013] Assign a region identifier to the connected domain, and the region identifier is a unique identifier.

[0014] In one embodiment, determining the target information of the target area in the template image based on a preset image analysis algorithm, the target area includes a measurement target and a measurement target edge, including:

[0015] Identify the boundary of the connected domain based on a preset contour extraction algorithm to obtain initial boundary information;

[0016] Perform geometric approximation processing on the boundary of the connected domain according to the initial boundary information to obtain the boundary information of the connected domain that satisfies the preset geometric constraint conditions.

[0017] In one embodiment, determining the target information of the target area in the template image based on a preset image analysis algorithm, the target area includes a measurement target and a measurement target edge, including:

[0018] Obtain the measurement target edge information of the target area, and obtain the pixel change trend curve of the measurement target edge;

[0019] Determine the edge type of the measurement target edge according to the pixel change trend curve.

[0020] In one embodiment, before processing the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image, it further includes:

[0021] Obtain a sample data set, and the sample data set includes historical detection images and corresponding graphic data system templates;

[0022] Perform pattern matching on the historical detection image and the graphic data system template, and generate several types of sample template images based on the graphic data system template;

[0023] Construct a label dataset based on the historical detection images and the sample template images at a preset ratio, and train a pre-constructed cyclic generative adversarial network model through the label dataset to obtain the image conversion model.

[0024] In one embodiment, the constructing a label dataset based on the historical detection images and the sample template images at a preset ratio, and training a pre-constructed cyclic generative adversarial network model through the label dataset to obtain the image conversion model includes:

[0025] The cyclic generative adversarial network model includes two generative networks and corresponding discriminative networks. The generative network is used to realize the conversion and generation between the detection image and the template image, and the discriminative network is used to realize the authenticity discrimination of the generated image.

[0026] Construct an optimization objective function, and constrain the training process of the cyclic generative adversarial network model based on the optimization objective function. The optimization objective function includes the optimization objective function between the corresponding generative network and the discriminative network, and also includes the cyclic loss function between the generative networks.

[0027] In a second aspect, the present application also provides an electron microscopy imaging measurement device. The device includes:

[0028] A detection image module, configured to obtain a detection image of a target object to be processed;

[0029] A template image module, configured to process the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image;

[0030] A target detection module, configured to determine target information of a target region in the template image based on a preset image analysis algorithm, where the target region includes a measurement target and a measurement target edge;

[0031] A measurement parameter module, configured to generate a measurement target box corresponding to the target region according to the target information, apply the measurement target box to the detection image to determine a local region including the target object in the detection image, and image the local region to realize the measurement of the target object.

[0032] In one embodiment, the target detection module includes:

[0033] A connected component module, configured to traverse the template image and determine several connected components in the template image based on the pixel value difference in the template image;

[0034] A region identification module, configured to assign a region identifier to the connected component, and the region identifier is a unique identifier.

[0035] In one embodiment, the target detection module includes:

[0036] An initial boundary module, configured to identify the boundary of the connected domain based on a preset contour extraction algorithm, and obtain initial boundary information;

[0037] A geometric approximation module, configured to perform geometric approximation processing on the boundary of the connected domain according to the initial boundary information, and obtain the boundary information of the connected domain that satisfies preset geometric constraint conditions.

[0038] In one embodiment, the target detection module includes:

[0039] A pixel trend module, configured to obtain the measured target edge information of the target area, and obtain the pixel change trend curve of the measured target edge;

[0040] An edge type module, configured to determine the edge type of the measured target edge according to the pixel change trend curve.

[0041] In one embodiment, before the template image module, there is further included:

[0042] A sample data module, configured to obtain a sample data set, where the sample data set includes historical detection images and corresponding graphic data system templates;

[0043] A pattern matching module, configured to perform pattern matching between the historical detection image and the graphic data system template, and generate several types of sample template images based on the graphic data system template;

[0044] A conversion model module, configured to construct a label data set based on a preset proportion of the historical detection images and the sample template images, and train a pre-constructed cyclic generative adversarial network model through the label data set to obtain the image conversion model.

[0045] In one embodiment, the conversion model module includes:

[0046] The cyclic generative adversarial network model includes two generative networks and corresponding discriminative networks. The generative network is used to realize the conversion generation between the detection image and the template image, and the discriminative network is used to realize the authenticity discrimination of the generated image;

[0047] An optimization function module, configured to construct an optimization objective function, and constrain the training process of the cyclic generative adversarial network model based on the optimization objective function. The optimization objective function includes the optimization objective function between the corresponding generative network and the discriminative network, and also includes the cyclic loss function between the generative networks.

[0048] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the steps in an electron microscopy imaging measurement method described in any one of the embodiments in the first aspect are implemented.

[0049] In a fourth aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps in an electron microscopy imaging measurement method described in any one of the embodiments in the first aspect are implemented.

[0050] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program. When the computer program is executed by a processor, the steps in an electron microscopy imaging measurement method described in any one of the embodiments in the first aspect are implemented.

[0051] For the above electron microscopy imaging measurement method, device, equipment, medium, and computer program product, by deriving through the technical features in the independent claim, the beneficial effects corresponding to the technical problems in the background art can be achieved:

[0052] The present application provides an electron microscopy imaging measurement method, including obtaining a detection image of a target object to be processed; processing the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image; determining target information of a target region in the template image based on a preset image analysis algorithm, where the target region includes a measurement target and the edge of the measurement target; generating a measurement target frame corresponding to the target region according to the target information, and applying the measurement target frame to the detection image to implement the measurement of the target object. In implementation, after obtaining the detection image of the target object to be processed, the pre-constructed image conversion model is applied to process the detection image, so that the detection image can be converted into a template image. At this time, the obtained template image has stronger standardization and consistency, enabling the imaging measurement method to adapt to different imaging conditions, improving the stability of target recognition and measurement. On the other hand, the template image can weaken the influence of noise information in the detection image on the measurement, improve the efficiency and accuracy of measuring the target object, enhance the identification of the target region and the edge, and contribute to the subsequent step of extracting target information, further improving the speed and accuracy of the measurement. Finally, a measurement target frame can be generated based on the obtained target information and applied to the detection image to implement the measurement process of the target object, improving the overall efficiency and accuracy of the measurement process. Description of the Drawings

[0053] To more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the accompanying drawings required for the description of the embodiments of the present application or the related art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related accompanying drawings can also be obtained based on these drawings.

[0054] Figure 1 It is a schematic diagram of the first process of an electron microscopy imaging measurement method in an embodiment.

[0055] Figure 2 It is a schematic diagram of the process of image acquisition in the measurement process.

[0056] Figure 3 It is a schematic diagram of the second process of an electron microscopy imaging measurement method in another embodiment.

[0057] Figure 4 It is a schematic diagram of the third process of an electron microscopy imaging measurement method in another embodiment.

[0058] Figure 5 It is a schematic diagram of the fourth process of an electron microscopy imaging measurement method in another embodiment.

[0059] Figure 6 It is a schematic diagram of the classification of edge types in an embodiment.

[0060] Figure 7 It is a schematic diagram showing the correspondence between the pixel change trend curve and the edge type in an embodiment.

[0061] Figure 8 It is a schematic diagram of the fifth process of an electron microscopy imaging measurement method in another embodiment.

[0062] Figure 9 It is a schematic diagram of the sixth process of an electron microscopy imaging measurement method in another embodiment.

[0063] Figure 10 It is a schematic diagram of the process of the cyclic generation discriminant network in an embodiment.

[0064] Figure 11 It is a schematic diagram of the architecture of the deep learning generation network.

[0065] Figure 12 It is a schematic diagram of the architecture of the deep learning discriminant network.

[0066] Figure 13 It is a block diagram of the structure of an electron microscopy imaging measurement device in an embodiment.

[0067] Figure 14Internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0068] To make the objectives, technical solutions and advantages of this application clearer, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0069] In related technologies, measurement algorithms in electron microscopy imaging detection are a series of algorithms used to measure and analyze various features and parameters on semiconductor wafers. These algorithms are crucial in the semiconductor manufacturing process, especially in wafer quality control, defect detection, and process optimization. Among them, common algorithms and technologies mainly include image processing algorithms, geometric measurement, statistical analysis, model-based algorithms, measurement algorithms for optical and other types of sensing technologies, etc. Currently, most measurement algorithms are based on traditional algorithms, and technicians need to manually set recipes according to the set imaging conditions to locate the measurement positions and measurement parameters.

[0070] However, current measurement methods have the following technical problems: existing measurement methods are inefficient and have poor accuracy, and need to be optimized.

[0071] Based on this, the embodiments of this application provide an electron microscopy imaging measurement method, device, equipment, medium, and program product.

[0072] In one embodiment, as Figure 1 and Figure 2 shown, an electron microscopy imaging measurement method is provided. In this embodiment, taking the application of this method to a terminal as an example, it can be understood that this method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0073] Step 102: Obtain a detection image of a target object to be processed.

[0074] Step 104: Process the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image.

[0075] Step 106: Determine the target information of the target region in the template image based on a preset image analysis algorithm, where the target region includes a measurement target and the edge of the measurement target.

[0076] Step 108: Generate a measurement target box corresponding to the target area according to the target information, apply the measurement target box to the detection image to determine a local area containing the target object in the detection image, and image the local area to measure the target object.

[0077] Among them, the measurement target box can refer to a bounding box used to identify and locate the area to be measured in electron microscopy imaging. The measurement target box is usually a conventional geometric shape box, such as a rectangular box, used to accurately frame the measurement target and its edges in the image. The main functions of the measurement target box are: to clarify the position of the measurement target in the image; to define the boundary of the measurement target, determine the edges of the measurement target, and facilitate subsequent geometric measurements and analyses; to assist the algorithm to achieve measurement, reduce manual intervention, and improve the measurement efficiency and accuracy.

[0078] Exemplarily, the terminal can extract key information of the target area from the template image through an image analysis algorithm, such as the target position, size, shape, and edge features. Subsequently. The terminal can generate a measurement target box, such as a rectangular box, according to the target information, so as to frame the measurement target and its edges through the measurement target box. At this time, the size and position of the measurement target are determined by the geometric features of the target. In this way, the terminal can map the measurement target box to the original detection image based on the mapping relationship between the algorithm space coordinate system and the real image coordinate system to ensure that the position and size of the measurement target box are consistent with the target in the detection image. Subsequently, in the processing of the detection image, the terminal performs more accurate imaging and measurement of the target within the local area framed by the measurement target box, obtains the measurement result of the target and outputs it to implement subsequent quality control, defect detection, or process optimization and other steps.

[0079] Specifically, taking the application of this solution to the microscopic measurement of the circuit structure of semiconductor devices as an example: First, through an image analysis algorithm, determine the area where the position of the circuit structure to be inspected is located and the image edge of the circuit structure; then generate a measurement target box according to the size and shape of the circuit structure; map the measurement target box to the detection image of the circuit, perform local focusing imaging measurement on the framed area of the measurement target box, and obtain information such as the size and connection of the local circuit structure; finally, output the measurement result of the circuit structure.

[0080] In the above electron microscopy imaging measurement method, through reasonable derivation combined with the technical features in the embodiments, the following beneficial effects can be achieved to solve the technical problems proposed in the background technology:

[0081] The present application provides an electron microscopy imaging measurement method, including obtaining a detection image of a target object to be processed; processing the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image; determining target information of a target region in the template image based on a preset image analysis algorithm, where the target region includes a measurement target and the edge of the measurement target; generating a measurement target box corresponding to the target region according to the target information, applying the measurement target box to the detection image to determine a local region containing the target object in the detection image, and imaging the local region to achieve measurement of the target object. In implementation, after obtaining the detection image of the target object to be processed, the pre-constructed image conversion model is applied to process the detection image, so as to convert the detection image into a template image. At this time, the obtained template image has stronger standardization and consistency, enabling this imaging measurement method to adapt to different imaging conditions, improving the stability of target recognition and measurement. On the other hand, the template image can weaken the influence of noise information in the detection image on measurement, improve the efficiency and accuracy of measuring the target object, enhance the identification of the target region and the edge, contribute to the processing of extracting target information in subsequent steps, and further improve the speed and accuracy of measurement. Finally, a measurement target box can be generated based on the obtained target information and applied to the detection image to achieve the measurement processing of the target object, improving the overall efficiency and accuracy of the measurement process.

[0082] In one embodiment, as Figure 3 shown, step 106 includes:

[0083] Step 302: Traverse the template image and determine several connected components in the template image based on the pixel value differences in the template image.

[0084] Step 304: Assign a region identifier to the connected component, and the region identifier is a unique identifier.

[0085] In this embodiment, in the process of detecting and identifying the target in the template image, a region identifier is assigned to each detected connected component, and the unique identifier helps to improve the efficiency and accuracy of target recognition in complex images.

[0086] In one embodiment, as Figure 4 shown, step 106 includes:

[0087] Step 402: Identify the boundary of the connected component based on a preset contour extraction algorithm to obtain initial boundary information.

[0088] Step 404: Perform geometric approximation processing on the boundary of the connected component according to the initial boundary information to obtain the boundary information of the connected component that satisfies the preset geometric constraint conditions.

[0089] In this embodiment, in the process of boundary recognition of the target, geometric approximation processing is used to process the boundary contour, which helps to improve the detection efficiency of boundary information.

[0090] In one embodiment, as Figure 5 shown, step 106 includes:

[0091] Step 502: Obtain the measured target edge information of the target area, and obtain the pixel change trend curve of the measured target edge.

[0092] Step 504: Determine the edge type of the measured target edge according to the pixel change trend curve.

[0093] Specifically, for example, as Figure 2 , Figure 6 and Figure 7 shown, the terminal can traverse the image, and respectively mark all measured targets, measured target edges and background regions according to different pixel values, and three different regions are respectively assigned unique labels. Subsequently, a contour extraction algorithm can be used to identify the boundaries of each connected domain, and the extracted contours are approximated by polygons to simplify the contour shapes. In this way, the terminal can check whether the approximated polygon meets the conditions of a rectangle, that is, all four corners are right angles, and the opposite sides are parallel and of equal length, and calculate the center and the x-direction width of the rectangle that meets the conditions.

[0094] Exemplarily, as Figure 6 shown, the template image can be divided into a dark image with bright edges, a dark image, a bright image and a bright image with bright edges. The terminal can obtain the pixel change information of the measured target in the x direction according to the position of the measured target. As Figure 7 shown, the upper half of the figure shows 4 edge types, and the lower half shows the pixel change trend curves in the x direction corresponding to the edges. In the two-color bright pattern, the gray levels are 128 and 255, where 128 is the background gray level and 255 is the pattern gray level; in the two-color dark pattern, the gray levels are 0 and 128, where 128 is the background gray level and 0 is the pattern gray level; in the three-color bright-edge dark pattern, the gray levels are 0, 128 and 255, where 128 is the background gray level, 0 is the pattern gray level, and 255 is the bright edge; in the three-color bright-edge bright pattern, the gray levels are 128, 200, 255, where 128 is the background gray level, 200 is the pattern gray level, and 255 is the bright edge. The edge type can be quickly determined according to the pixel change trend curve.

[0095] In this embodiment, in the process of identifying the edge type of the target, the edge type can be determined through the pixel change trend curve, which helps to improve the detection efficiency of the edge type.

[0096] In one embodiment, as Figure 8As shown, before step 104, it further includes:

[0097] Step 802: Obtain a sample data set, where the sample data set includes historical detection images and corresponding graphic data system templates.

[0098] Step 804: Perform pattern matching between the historical detection images and the graphic data system templates, and generate several types of sample template images based on the graphic data system templates.

[0099] Step 806: Construct a label data set based on a preset proportion of the historical detection images and the sample template images, and train a pre-constructed cyclic generative adversarial network model through the label data set to obtain the image conversion model.

[0100] In this embodiment, during the training of the image conversion model, the label data set is constructed by matching the historical detection images and the graphic data system templates, which helps improve the conversion accuracy of the finally trained image conversion model, making the obtained template images meet the processing requirements of the graphic data system.

[0101] In one embodiment, as Figure 9 shown, step 806 includes:

[0102] The cyclic generative adversarial network model includes two generative networks and a corresponding discriminative network. The generative network is used to realize the conversion and generation between the detection images and the template images, and the discriminative network is used to realize the authenticity discrimination of the generated images.

[0103] Step 902: Construct an optimization objective function, and constrain the training process of the cyclic generative adversarial network model based on the optimization objective function. The optimization objective function includes the optimization objective function between the corresponding generative network and the discriminative network, and also includes the cyclic loss function between the generative networks.

[0104] Specifically, the terminal can perform pattern matching between the SEM images and the GDS templates, and generate template images corresponding to 4 types of edge SEM images according to the information in the GDS templates. As Figure 6 shown, they are respectively dark image with bright edge, dark image, bright image, and bright image with bright edge, with 500 images for each type of edge image, serving as the label data set for model training.

[0105] Subsequently, the terminal can use the cyclic generative adversarial network. There are 4 models in the method, which are respectively 、 、 and , where and Denote the generative models, which have the same structure. The model composition is as Figure 11 shown, including 3 convolutional modules, 11 residual modules, a 3x3 convolutional layer and an activation layer. And Denote two discriminative models, which are respectively used to discriminate , the authenticity of the generated images. They have the same structure. The model composition is as Figure 12 shown, including 4 convolutional modules, a pad layer and a 3x3 convolutional layer.

[0106] Specifically, the training process is as Figure 10 shown. X and Y respectively represent the original SEM image and the template image. The network contains 4 models G, F, Dx and Dy. The G network completes the mapping from X to Y, the F network completes the mapping from Y to X, the Dx network discriminates the authenticity of the X generated by the F network, and the Dy network discriminates the authenticity of the Y generated by the G network. The complete optimization objective function is as follows:

[0107] ;

[0108] Among them:

[0109] ;

[0110] ;

[0111] ;

[0112] Denote the optimization objective function of the generative network and the discriminative network ;

[0113] Denote the optimization objective function of the generative network and the discriminative network ;

[0114] Denote the cyclic loss function between the two generative networks;

[0115] Denote the input of the SEM image into the network;

[0116] Denote the input of the generated template image into the discriminative network;

[0117] Denote the input of the template image into the network;

[0118] Indicates SEM image input Discriminant network;

[0119] In this way, with L(G, F, Dx, Dy) as the optimization objective, two generator networks F and G are obtained after training. G is used to generate a template image from the SEM image, and F is used to generate an SEM image from the template image.

[0120] In this embodiment, in the training process of the image conversion model, a cyclic loss function between the generator networks is set in the optimization objective function, which helps to support that the input image can be converted by the model to obtain an acceptable reconstruction, enhancing the robustness and reliability of the image conversion model.

[0121] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps in other steps.

[0122] Based on the same inventive concept, an embodiment of the present application also provides an electron microscopy imaging measurement device for implementing an electron microscopy imaging measurement method involved above. The implementation solution provided by this device to solve the problem is similar to the implementation solution described in the above method. Therefore, the specific limitations in one or more embodiments of the electron microscopy imaging measurement device provided below can refer to the limitations on the electron microscopy imaging measurement method in the above text, and will not be repeated here.

[0123] In one embodiment, as Figure 13 shown, an electron microscopy imaging measurement device is provided, including: a detection image module, a template image module, a target detection module, and a measurement parameter module, where:

[0124] The detection image module is used to obtain a detection image of a target object to be processed;

[0125] The template image module is used to process the detection image based on a pre-constructed image conversion model to obtain a template image matching the detection image;

[0126] A target detection module, configured to determine target information of a target area in the template image based on a preset image analysis algorithm, where the target area includes a measurement target and the edge of the measurement target;

[0127] A measurement parameter module, configured to generate a measurement target box corresponding to the target area according to the target information, apply the measurement target box to the detection image to determine a local area containing the target object in the detection image, and image the local area to implement measurement of the target object.

[0128] In one embodiment, the target detection module includes:

[0129] A connected component module, configured to traverse the template image and determine several connected components in the template image based on pixel value differences in the template image;

[0130] A region identification module, configured to assign a region identifier to the connected component, where the region identifier is a unique identifier.

[0131] In one embodiment, the target detection module includes:

[0132] An initial boundary module, configured to identify the boundary of the connected component based on a preset contour extraction algorithm and obtain initial boundary information;

[0133] A geometric approximation module, configured to perform geometric approximation processing on the boundary of the connected component according to the initial boundary information to obtain boundary information of the connected component that satisfies preset geometric constraint conditions.

[0134] In one embodiment, the target detection module includes:

[0135] A pixel trend module, configured to obtain measurement target edge information of the target area and obtain a pixel change trend curve of the measurement target edge;

[0136] An edge type module, configured to determine the edge type of the measurement target edge according to the pixel change trend curve.

[0137] In one embodiment, before the template image module, there is also:

[0138] A sample data module, configured to obtain a sample data set, where the sample data set includes historical detection images and corresponding graphic data system templates;

[0139] A pattern matching module, configured to perform pattern matching between the historical detection image and the graphic data system template and generate several types of sample template images based on the graphic data system template;

[0140] A conversion model module for constructing a label data set based on the historical detection images and the sample template images at a preset ratio, and training a pre-constructed cyclic generative adversarial network model through the label data set to obtain the image conversion model.

[0141] In one embodiment, the conversion model module includes:

[0142] The cyclic generative adversarial network model includes two generative networks and corresponding discriminative networks. The generative networks are used to realize the conversion and generation between the detection images and the template images, and the discriminative networks are used to realize the authenticity discrimination of the generated images.

[0143] An optimization function module for constructing an optimization objective function and constraining the training process of the cyclic generative adversarial network model based on the optimization objective function. The optimization objective function includes the optimization objective functions corresponding to the generative network and the discriminative network, and also includes the cyclic loss function between the generative networks.

[0144] Each module in the above electronic microscopy imaging measurement device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0145] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 14As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. The computer program, when executed by the processor, implements an electron microscopy imaging measurement method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0146] Those skilled in the art can understand that Figure 14 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0147] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0148] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0149] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0150] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions.

[0151] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0152] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0153] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. An electron microscopic imaging measurement method, characterized in that: The method comprises: Acquire a detection image of a target object to be processed; Processing the detection image based on a pre-built image conversion model to obtain a template image matching the detection image; Determine target information of a target area in the template image based on a preset image analysis algorithm, wherein the target area includes a measurement target and a measurement target edge; Generating a measurement target frame corresponding to the target area according to the target information, applying the measurement target frame to the detection image to determine a local area containing the target object in the detection image, and imaging the local area to achieve measurement of the target object; Before the detection image is processed based on the pre-built image conversion model to obtain a template image matching the detection image, the method further includes: Acquire a sample data set, wherein the sample data set includes historical detection images and corresponding graphic data system templates; Performing pattern matching between the historical detection image and the graphic data system template, and generating several types of sample template images based on the graphic data system template; A label data set is constructed based on the historical detection images and the sample template images in a preset ratio, and a pre-constructed loop generation discriminant network model is trained by using the label data set to obtain the image conversion model.

2. The method according to claim 1, characterized in that The target information of the target area in the template image is determined based on a preset image analysis algorithm, wherein the target area includes a measurement target and a measurement target edge, including: Traversing the template image, and determining a plurality of connected domains in the template image based on pixel value differences in the template image; A region identifier is assigned to the connected domain, where the region identifier is a unique identifier.

3. The method according to claim 2, characterized in that The target information of the target area in the template image is determined based on a preset image analysis algorithm, wherein the target area includes a measurement target and a measurement target edge, including: Identify the boundary of the connected domain based on a preset contour extraction algorithm to obtain initial boundary information; A geometric approximation process is performed on the boundary of the connected domain according to the initial boundary information to obtain boundary information of the connected domain that meets preset geometric constraints.

4. The method according to claim 1, characterized in that: The target information of the target area in the template image is determined based on a preset image analysis algorithm, wherein the target area includes a measurement target and a measurement target edge, including: Acquire the measurement target edge information of the target area, and acquire the pixel change trend curve of the measurement target edge; The edge type of the edge of the measurement target is determined according to the pixel change trend curve.

5. The method according to claim 1, characterized in that The step of constructing a label data set based on the historical detection images and the sample template images of a preset ratio, and training a pre-constructed cyclic generation discriminant network model through the label data set to obtain the image conversion model includes: The cyclic generation discriminant network model includes two generation networks and a corresponding discriminant network, wherein the generation network is used to realize the conversion generation between the detection image and the template image, and the discriminant network is used to realize the authenticity discrimination of the generated image; An optimization objective function is constructed, and the training process of the cyclic generation and discrimination network model is constrained based on the optimization objective function. The optimization objective function includes the corresponding optimization objective function between the generation network and the discrimination network, and also includes a cyclic loss function between the generation networks.

6. An electron microscopic imaging measurement device, characterized in that: The device comprises: A detection image module, used to obtain a detection image of a target object to be processed; A template image module, used for processing the detection image based on a pre-built image conversion model to obtain a template image matching the detection image; A target detection module, used to determine target information of a target area in the template image based on a preset image analysis algorithm, wherein the target area includes a measurement target and a measurement target edge; A measurement parameter module, configured to generate a measurement target frame corresponding to the target area according to the target information, apply the measurement target frame to the detection image to determine a local area containing the target object in the detection image, and image the local area to achieve measurement of the target object; Before the template image module, it also includes: A sample data module, used to obtain a sample data set, wherein the sample data set includes a historical detection image and a corresponding graphic data system template; A pattern matching module, used for performing pattern matching between the historical detection image and the graphic data system template, and generating several types of sample template images based on the graphic data system template; The conversion model module is used to construct a label data set based on the historical detection images and the sample template images of a preset ratio, and train the pre-constructed loop generation discriminant network model through the label data set to obtain the image conversion model.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images