Evaluation methods, systems, media, terminals, and application products for curve mask process correction based on SEM image contour extraction.

Abstract

This application provides an evaluation method for curve mask process correction based on SEM image contour extraction. The method includes: acquiring an initial curve pattern and its corresponding standard SEM image contour; constructing a simulation prediction model using a machine learning model based on the initial curve pattern and the standard SEM image contour; obtaining a corrected curve pattern of the initial curve pattern using a curve mask method, and extracting the corrected SEM image contour corresponding to the corrected curve pattern; inputting the corrected curve pattern into the simulation prediction model to generate a corrected simulation image of the corrected curve pattern; calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image; and analyzing the correction difference value to obtain a correction evaluation result. This application can more comprehensively and objectively reflect the contour accuracy and stability after curve mask process correction.

Description

Technical Field

[0001] This application relates to the field of evaluation technology for curve mask process correction, and in particular to an evaluation method, system, medium, terminal and program product for curve mask process correction based on SEM image contour extraction. Background Technology

[0002] With the continuous development of the semiconductor manufacturing field, the industry is constantly pursuing higher precision and smaller feature sizes. However, traditional Manhattan mask patterns, composed of straight lines and right angles, have limitations in high-precision lithography: on the one hand, as feature sizes continue to shrink, the effects of diffraction and scattering become more pronounced for traditional Manhattan patterns; on the other hand, traditional mask process correction neglects the optimization of the oblique direction, affecting the accuracy of distortion compensation. With the evolution of lithography technology, multi-beam electron beam mask writing technology can manufacture mask patterns of arbitrary shapes, greatly improving the efficiency and precision of mask manufacturing; reverse lithography technology calculates the mask pattern with the best imaging quality under given process conditions, generating curved mask patterns that can better adapt to the optical characteristics of the lithography process, reducing optical distortion and diffraction effects. This makes it possible to manufacture complex curved mask patterns, and curved mask technology shows extremely broad application prospects in the field of advanced process node semiconductor manufacturing. To solve the problems in curved mask manufacturing, curved mask process correction technology plays an irreplaceable role. Curve mask process correction technology uses models to accurately correct and optimize complex curve graphics, thereby ensuring that curve mask patterns are manufactured precisely.

[0003] For traditional Manhattan patterns, the correction target mainly involves adjusting the position of straight line edges. During simulation verification, only the edge position and dimensional changes of the lines need to be considered. Measurement typically only requires measuring key dimensions such as one-dimensional width and spacing. By comparing simulation data before and after mask pattern correction with the measurement data of key dimensions of the mask pattern, the accuracy of the mask process correction model and the final correction can be effectively evaluated. How to evaluate the accuracy of curve mask process correction is a significant challenge currently facing the industry.

[0004] The existing technology has the following drawbacks:

[0005] (1) Curved mask patterns are usually non-Manhattan structures with continuously changing curved shapes at their boundaries. They do not have fixed line width and spacing characteristics, making it difficult to accurately characterize them using traditional measurement methods based on key dimensions. As a result, the evaluation results are difficult to reflect the true contour offset.

[0006] (2) Due to the complex structure and varied boundary morphology of curved patterns, traditional measurement methods based on regular or simple geometric features need to process a large amount of irregular boundary data when performing contour matching and error calculation. This results in a complex calculation process, low computational efficiency, and difficulty in balancing evaluation accuracy and computational cost. In addition, existing technologies usually lack a unified evaluation index that can simultaneously characterize the overall contour offset and local error distribution characteristics, making it difficult to provide effective guidance for the optimization of curve mask processes.

[0007] Therefore, it is necessary to provide an evaluation method for curve mask process correction based on SEM image contour extraction to solve the above-mentioned problems in the prior art. Summary of the Invention

[0008] In view of the shortcomings of the prior art described above, the purpose of this application is to provide an evaluation method, system, medium, terminal and program product for curve mask process correction based on SEM image contour extraction, so as to solve the above-mentioned evaluation problem of curve mask process correction.

[0009] To achieve the above and other related objectives, a first aspect of this application provides a method comprising: acquiring an initial curve pattern and a corresponding standard SEM image contour; constructing a simulation prediction model using a machine learning model based on the initial curve pattern and the standard SEM image contour; obtaining a corrected curve pattern of the initial curve pattern using a curve masking method, and extracting the corrected SEM image contour corresponding to the corrected curve pattern; inputting the corrected curve pattern into the simulation prediction model to generate a corrected simulation image of the corrected curve pattern; calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image; and analyzing the correction difference value to obtain a correction evaluation result.

[0010] In some embodiments of the first aspect of this application, a simulation prediction model is constructed using a machine learning model based on the initial curve pattern and the standard SEM image contour. The specific process includes: inputting the initial curve pattern into a pre-constructed simulation prediction model to generate a corresponding initial simulation image; calculating the image difference value between the standard SEM image contour and the initial simulation image; when the image difference value meets a preset accuracy requirement, using the pre-constructed simulation prediction model as the final simulation prediction model; when the image difference value does not meet the preset accuracy requirement, adjusting the parameters of the simulation prediction model based on the initial curve pattern and the corresponding standard SEM image contour to obtain an updated simulation prediction model; and inputting the initial curve pattern again into the updated simulation prediction model to generate a new initial simulation image, recalculating the image difference value between the standard SEM image contour and the new initial simulation image, until the image difference value meets the preset accuracy requirement to obtain the final simulation prediction model.

[0011] In some embodiments of the first aspect of this application, the specific process of calculating and obtaining the image difference value between the standard SEM image contour and the initial simulation image includes: aligning the standard SEM image contour with the initial simulation image in coordinates; calculating the contour error value between the edge pixels corresponding to the standard SEM image contour and the edge pixels corresponding to the initial simulation image contour, so as to obtain the image difference value.

[0012] In some embodiments of the first aspect of this application, the specific process of calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image includes: aligning the corrected SEM image contour with the corrected simulation image; marking a plurality of first sampling points on the corrected SEM image contour; obtaining the edge intersection points corresponding to each first sampling point and the corrected simulation image along the normal direction of each first sampling point; connecting each first sampling point with the corresponding edge intersection points and calculating the first vector value corresponding to each first sampling point; and calculating the corresponding average value, absolute value, and average of absolute values ​​based on all the first vector values. The first correction evaluation result is obtained by aligning the initial curve pattern with the corrected simulation image, marking several second sampling points on the initial curve pattern, obtaining the edge intersection points of each second sampling point and the corrected simulation image along the normal direction of each second sampling point, connecting each second sampling point with the corresponding edge intersection point and calculating the second vector value corresponding to each second sampling point, and calculating the corresponding average, absolute value, average of absolute values ​​and standard deviation based on all second vector values, which are used as the second correction evaluation result; the correction difference value is calculated based on the first correction evaluation result and the second correction evaluation result.

[0013] In some embodiments of the first aspect of this application, the specific process of calculating the correction difference value based on the first correction evaluation result and the second correction evaluation result includes: averaging the average of the absolute values ​​in the first correction evaluation result and the average of the absolute values ​​in the second correction evaluation result to obtain the overall contour offset error; averaging the standard deviations in the first correction evaluation result and the standard deviations in the second correction evaluation result to obtain the error dispersion index; and weighted summing the overall contour offset error and the error dispersion index to obtain the correction difference value.

[0014] In some embodiments of the first aspect of this application, the specific process of obtaining the standard SEM image contour includes: acquiring partial SEM image contours with Manhattan structure features in the initial curve pattern several times based on edge detection technology; obtaining acquisition differences by comparing the several partial SEM image contours with the corresponding initial curve pattern; when the acquisition differences meet the preset acquisition rules, acquiring the SEM image contour of the initial curve pattern as a preliminary SEM image contour based on the edge detection technology; acquiring several preliminary SEM image contours of the initial curve pattern; and calculating the average image of the several preliminary SEM image contours as the standard SEM image contour of the initial curve pattern.

[0015] To achieve the above and other related objectives, a second aspect of this application provides an evaluation system for curve mask process correction based on SEM image contour extraction, comprising: a model building module for acquiring an initial curve pattern and a corresponding standard SEM image contour, and constructing a simulation prediction model using a machine learning model based on the initial curve pattern and the standard SEM image contour; and a correction evaluation module for obtaining a corrected curve pattern of the initial curve pattern using a curve mask method, and extracting the corrected SEM image contour corresponding to the corrected curve pattern; inputting the corrected curve pattern into the simulation prediction model to generate a corrected simulation image of the corrected curve pattern, calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image, and analyzing the correction difference value to obtain a correction evaluation result.

[0016] To achieve the above and other related objectives, a third aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the evaluation method for curve mask process correction based on SEM image contour extraction.

[0017] To achieve the above and other related objectives, a fourth aspect of this application provides a computer program product comprising computer program code that, when executed on a computer, causes the computer to implement the evaluation method for curve mask process correction based on SEM image contour extraction.

[0018] To achieve the above and other related objectives, a fifth aspect of this application provides an electronic terminal, including a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the evaluation method for curve mask process correction based on SEM image contour extraction.

[0019] As described above, the evaluation method for curve mask process correction based on SEM image contour extraction provided in this application has the following beneficial effects:

[0020] By introducing a machine learning-based simulation prediction model, a mapping relationship is established between the curve pattern and its corresponding SEM image contour, achieving high-precision simulation prediction of the pattern contour after curve mask correction. This avoids the high cost and low efficiency problems caused by relying entirely on actual lithography experiments for verification. This application constructs sampling points on the corrected SEM image contour and the initial curve pattern respectively, and calculates the contour offset vector based on the normal direction. It quantitatively analyzes the correction effect from two dimensions: overall offset error and error dispersion. Compared with the evaluation method that only uses a single error index, it can more comprehensively and objectively reflect the contour accuracy and stability after curve mask process correction. By weighted fusion of the overall contour offset error and error dispersion index, a unified correction difference value is constructed, realizing the comparability evaluation of the effects of different correction schemes and improving the engineering guidance significance of the evaluation results. Attached Figure Description

[0021] Figure 1 The diagram shown is a flowchart illustrating an evaluation method for curve mask process correction based on SEM image contour extraction in one embodiment of this application.

[0022] Figure 2 The diagram shown is a schematic diagram of contour extraction based on SEM images in one embodiment of this application.

[0023] Figure 3 The diagram shown is a schematic representation of the extraction of a standard SEM image contour in one embodiment of this application.

[0024] Figure 4 The diagram shown is a schematic diagram of the calculation of the first vector value in one embodiment of this application.

[0025] Figure 5The diagram shown is a structural schematic of an evaluation system for curve mask process correction based on SEM image contour extraction in one embodiment of this application.

[0026] Figure 6 The diagram shown is a structural schematic of an electronic terminal according to an embodiment of this application. Detailed Implementation

[0027] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0028] In the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" do not necessarily imply that they are different.

[0029] It should be noted that, in the embodiments of this application, the words "exemplary" or "for example" refer to examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0030] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0031] Before providing a further detailed description of the present invention, the nouns and terms used in the embodiments of the present invention are explained, and the nouns and terms used in the embodiments of the present invention are subject to the following interpretations:

[0032] <1> SEM: Scanning Electron Microscopy, which uses a focused high-energy electron beam to scan the surface of a sample and observe its microstructure and composition by detecting the various signals emitted by the sample.

[0033] To facilitate understanding of the embodiments of this application, firstly, in conjunction with Figure 1 Detailed explanation. Figure 1 This document illustrates a flowchart of an evaluation method for curve mask process correction based on SEM image contour extraction, as described in an embodiment of the present invention. The evaluation method for curve mask process correction based on SEM image contour extraction in this embodiment mainly includes the following steps:

[0034] Step S11: Obtain the initial curve pattern and the corresponding standard SEM image contour. Based on the initial curve pattern and the standard SEM image contour, construct a simulation prediction model using a machine learning model.

[0035] It should be noted that the initial curve pattern refers to the original layout curve graphic used for photolithography simulation or mask design before the curve mask process is modified. It is the initial curve structure obtained by the target graphic in the layout design stage, used as the input sample of the simulation prediction model, and compared with the corresponding standard SEM image contour to establish the simulation prediction model.

[0036] In one embodiment of this application, the specific process of obtaining the standard SEM image contour includes: acquiring partial SEM image contours with Manhattan structure features in the initial curve pattern several times based on edge detection technology; obtaining acquisition differences by comparing the several partial SEM image contours with the corresponding initial curve pattern; when the acquisition differences meet the preset acquisition rules, acquiring the SEM image contour of the initial curve pattern as a preliminary SEM image contour based on the edge detection technology, and obtaining several preliminary SEM image contours of the initial curve pattern; calculating the average image of the several preliminary SEM image contours as the standard SEM image contour of the initial curve pattern.

[0037] It should be noted that the Manhattan structure feature refers to a partial contour formed by horizontal and vertical line segments connected at right angles in layout design. Since the boundary direction of this type of structure is basically consistent with the X and Y axes of the layout coordinate system, it is widely used in integrated circuit layout design and photolithography. Compared to non-Manhattan structures, Manhattan structures have clear geometric features, stable edge directions, and smaller contour extraction errors, making it easier to obtain stable and reliable contour information through edge detection technology during SEM image contour extraction. Therefore, in this invention, by selecting a portion of the contour with Manhattan structure features in the initial curve pattern as a reference area, the accuracy of SEM contour extraction can be improved, and more stable reference data can be provided for the construction of subsequent simulation prediction models and curve mask correction evaluation.

[0038] Furthermore, contours conforming to the Manhattan structure are extracted from the initial curve pattern based on edge detection technology. All contours conforming to the Manhattan structure in the initial curve pattern constitute a partial SEM image contour of the initial curve pattern. Since the partial SEM image contour contains Manhattan structure contours with simple structures, a Manhattan structure region corresponding to the partial SEM image contour is selected in the initial curve pattern. The average distance between the partial SEM image contour and the partial SEM image contour is measured along a direction perpendicular to the boundary of the Manhattan structure region, obtaining the measured value of the partial SEM image contour and the initial curve pattern. When the difference between the measured value of the partial SEM image contour and the initial curve pattern is within ±1 nanometer, we consider the SEM image contour to have been accurately extracted, and the same scheme is applied to the extraction of curve-based SEM image contours. When the difference in the measured value exceeds ±1 nanometer, the acquisition conditions of the edge detection technology are adjusted, and the new measured value of the partial SEM image contour and the initial curve pattern is recalculated until the difference between the new measured value of the partial SEM image contour and the initial curve pattern is within ±1 nanometer.

[0039] Furthermore, such as Figure 2 As shown, the initial SEM image of the initial curve pattern is first obtained using a scanning electron microscope. Figure 2 In (a) of the above, edge detection technology based on the same acquisition conditions is used to acquire the entire contour of the initial curve pattern to obtain the preliminary SEM image contour. Figure 2 (b) in the middle.

[0040] It should be noted that in SEM images, the boundaries of target graphics are usually represented by areas with significant grayscale value changes. Therefore, edge detection technology can accurately extract the corresponding image contour information from SEM images. In this invention, by performing edge detection processing on the acquired SEM images, the contour of the target graphics can be obtained, which can then be used for subsequent contour alignment, error calculation, and simulation model evaluation. The edge detection technology can be any one or a combination of the Sobel operator edge detection method, the Canny edge detection method, the Laplacian operator edge detection method, or the LoG edge detection method. The specific method can be selected according to actual needs, and no specific limitation is made here.

[0041] Furthermore, such as Figure 3 As shown, three preliminary SEM image contours (preliminary SEM image contour 1, preliminary SEM image contour 2, and preliminary SEM image contour 3) corresponding to the same initial curve pattern are obtained by repeatedly acquiring data three times. First, spatial alignment is performed on each preliminary SEM image contour to ensure that different preliminary SEM image contours have a consistent reference position in the same coordinate system, thus eliminating the influence of scanning position offset or image acquisition errors. After spatial alignment, several marker points are selected at preset intervals on any preliminary SEM image contour, and the contour point coordinates corresponding to the marker points are determined in the remaining preliminary SEM image contours. Then, the contour point coordinates at corresponding positions in each preliminary SEM image contour are averaged, with both the horizontal and vertical coordinates being averaged to obtain the average contour point at the corresponding position. Finally, all average contour points are connected or fitted according to the contour order to construct an average image, which is then used as the standard SEM image contour corresponding to the initial curve pattern. It should be noted that connecting or fitting all average contour points according to the contour order and averaging the contours of SEM images acquired multiple times can effectively reduce the impact of SEM imaging noise and random measurement errors on the contour extraction results, thereby improving the stability and accuracy of contour evaluation.

[0042] In one embodiment of this application, a simulation prediction model is constructed using a machine learning model based on the initial curve pattern and the standard SEM image contour. The specific process includes: inputting the initial curve pattern into a pre-constructed simulation prediction model to generate a corresponding initial simulation image; calculating the image difference value between the standard SEM image contour and the initial simulation image; when the image difference value meets a preset accuracy requirement, using the pre-constructed simulation prediction model as the final simulation prediction model; when the image difference value does not meet the preset accuracy requirement, adjusting the parameters of the simulation prediction model based on the initial curve pattern and the corresponding standard SEM image contour to obtain an updated simulation prediction model; and inputting the initial curve pattern again into the updated simulation prediction model to generate a new initial simulation image, recalculating the image difference value between the standard SEM image contour and the new initial simulation image, until the image difference value meets the preset accuracy requirement to obtain the final simulation prediction model.

[0043] In one embodiment of this application, the specific process of calculating and obtaining the image difference value between the standard SEM image contour and the initial simulation image includes: aligning the standard SEM image contour with the initial simulation image in coordinates; calculating the contour error value between the edge pixels corresponding to the standard SEM image contour and the edge pixels corresponding to the initial simulation image contour, so as to obtain the image difference value.

[0044] Specifically, to evaluate the prediction accuracy of the simulation prediction model, it is necessary to calculate the difference between the initial curve pattern and the initial simulation image output by the simulation prediction model. Specifically, firstly, the initial curve pattern and the initial simulation image are aligned to be in the same coordinate system. Then, several contour error values ​​are obtained by uniformly sampling points on corresponding edges of the initial curve pattern and the initial simulation image and calculating the corresponding Euclidean distance.

[0045] Furthermore, statistical analysis is performed on the aforementioned contour error values. Here, the statistical analysis can be selected from statistical standards that characterize the differences, such as calculating the average error, the absolute average error, or the standard deviation of the error, so as to obtain image difference values ​​that represent the degree of difference between the initial curve pattern and the initial simulation image.

[0046] Furthermore, when the image difference value is less than or equal to the error threshold, the prediction accuracy of the simulation prediction model is considered to meet the preset accuracy requirements; when the image difference value is greater than the error threshold, the current simulation prediction model is considered to still have a large prediction deviation, requiring further optimization or adjustment of the model parameters. The error threshold refers to a maximum permissible error determined based on the actual lithography process accuracy requirements or historical experimental data. The error threshold can be set according to specific process requirements to ensure that the simulation prediction model can accurately reflect the actual contour features.

[0047] When the image difference value does not meet the preset accuracy requirement, the simulation prediction model is adjusted or optimized using the initial curve pattern and its corresponding standard SEM image contour to obtain an updated simulation prediction model. Subsequently, the initial curve pattern is input again into the updated simulation prediction model to generate a new initial simulation image, and the image difference value between the standard SEM image contour and the new initial simulation image is recalculated. The simulation prediction model is iteratively updated and corrected in this manner until the calculated image difference value meets the preset accuracy requirement, thus obtaining the final simulation prediction model for subsequent curve mask process correction and evaluation.

[0048] In some embodiments, the machine learning model may employ a convolutional neural network model, a deep neural network model, etc., without specific limitations here. It learns the mapping relationship between the initial curve pattern and the corresponding standard SEM image contour to predict the SEM imaging contour of the curve pattern. The simulation prediction model constructed in this way can accurately reflect the contour changes of the curve pattern under actual process conditions, providing a reliable simulation basis for evaluating the subsequent curve mask process correction effect.

[0049] Step S12: Obtain the corrected curve pattern of the initial curve pattern using the curve mask method, and extract the corrected SEM image contour corresponding to the corrected curve pattern; input the corrected curve pattern into the simulation prediction model to generate the corrected simulation image of the corrected curve pattern, calculate the correction difference value between the corrected SEM image contour, the initial curve pattern and the corrected simulation image, and analyze the correction difference value to obtain the correction evaluation result.

[0050] Specifically, the curve mask method is a mask optimization method for compensating for lithographic proximity effects in non-Manhattan structures or curved structures. It continuously adjusts the local geometry of the mask pattern, making the pattern closer to the design target contour during lithographic imaging and subsequent etching. In practice, based on the geometric contour features of the initial curve pattern, the curve boundary is discretely sampled, and combined with a lithographic imaging model or empirical correction rules, the positions of each sampling point are appropriately offset along the contour normal direction, thereby generating the compensated curve mask pattern. After the corrected curve pattern obtained in the above manner undergoes actual lithography, the resulting pattern contour can, to a certain extent, offset the shape errors caused by lithographic imaging blur, optical proximity effects, and process deviations, thereby improving the consistency between the final pattern and the target design pattern. This application does not limit the specific curve mask correction algorithm; it can be implemented using various techniques such as existing curve OPC (Optical Proximity Correction) methods, model-based curve mask optimization methods, or rule-based curve mask correction methods. The key point of this invention is to quantitatively evaluate the correction effect obtained by the curve mask method by combining SEM image contour extraction with simulation prediction model.

[0051] Furthermore, the modified SEM image contour corresponding to the modified curve pattern is obtained in the same way as the standard SEM image contour obtained above, which will not be elaborated here.

[0052] Furthermore, the corrected curve pattern is input into the final simulation prediction model after training to generate a corrected simulation image of the corrected curve pattern.

[0053] In one embodiment of this application, the specific process of calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image includes: aligning the corrected SEM image contour with the corrected simulation image; marking a plurality of first sampling points on the corrected SEM image contour; obtaining the edge intersection points of each first sampling point and the corrected simulation image along the normal direction of each first sampling point; connecting each first sampling point with the corresponding edge intersection point and calculating the first vector value corresponding to each first sampling point; calculating the corresponding average, absolute value, average of absolute values, and standard deviation based on all first vector values, as a first correction evaluation result; aligning the initial curve pattern with the corrected simulation image; marking a plurality of second sampling points on the initial curve pattern; obtaining the edge intersection points of each second sampling point and the corrected simulation image along the normal direction of each second sampling point; connecting each second sampling point with the corresponding edge intersection point and calculating the second vector value corresponding to each second sampling point; calculating the corresponding average, absolute value, average of absolute values, and standard deviation based on all second vector values, as a second correction evaluation result; and calculating the correction difference value based on the first correction evaluation result and the second correction evaluation result.

[0054] Specifically, such as Figure 4 As shown, several first sampling points are uniformly marked on the boundary line of the corrected SEM image contour. The spacing between adjacent first sampling points can be adaptively set according to the structural density of the corresponding curve pattern to balance calculation accuracy and calculation efficiency. This application does not make specific limitations on this.

[0055] Furthermore, taking each first sampling point as a reference, along its corresponding contour normal direction, sampling line segments of the same length are respectively cut on the inner and outer sides of the modified SEM image contour according to a preset step size. The length of the sampling line segments can be set according to the local structural features or structural density of the curve pattern, and this application does not limit it in this regard.

[0056] Furthermore, taking the first sampling point ( Figure 4 Taking the hollow points on the contour of the corrected SEM image as the starting point, and the intersection of the sampled line segment and the boundary of the corrected simulation image (…), the starting point is the hollow points on the contour of the corrected SEM image. Figure 4 The hollow point on the boundary of the corrected simulation image is taken as the endpoint, and a corresponding first vector value is constructed in the direction from the starting point to the endpoint. Figure 4 (Vectors between hollow points). By performing statistical analysis on all first vector values, the mean, absolute value, mean of absolute values, and standard deviation are calculated to characterize the difference in contour position and dispersion between the corrected SEM image contour and the corrected simulation image, and this is used as the first correction evaluation result.

[0057] Similarly, the second correction evaluation result of the initial curve pattern and the corrected simulation image is calculated using the same method. The specific calculation steps are as shown above and will not be elaborated here.

[0058] In one embodiment of this application, the specific process of calculating the correction difference value based on the first correction evaluation result and the second correction evaluation result includes: averaging the average of the absolute values ​​in the first correction evaluation result and the average of the absolute values ​​in the second correction evaluation result to obtain the overall contour offset error; averaging the standard deviations in the first correction evaluation result and the standard deviations in the second correction evaluation result to obtain the error dispersion index; and weighted summing the overall contour offset error and the error dispersion index to obtain the correction difference value.

[0059] Specifically, the average of the absolute values ​​in the first correction evaluation result and the average of the absolute values ​​in the second correction evaluation result are calculated to obtain the overall contour offset error. The overall contour offset error is used to represent the degree of overall positional offset between the corrected SEM image contour, the corrected simulation image, and the initial curve pattern.

[0060] Furthermore, the standard deviations of the first and second corrected evaluation results are extracted respectively, and the two are averaged to obtain an error dispersion index. This error dispersion index represents the dispersion and stability of the error distribution at each sampling point. Based on this, a weighted summation is performed on the overall contour offset error and the error dispersion index to obtain the corrected difference value. The weights of different indices can be set according to actual process requirements or evaluation priorities; this application does not impose specific limitations on this. Through the above method, the corrected difference value not only reflects the overall contour offset but also the uniformity of the error distribution, thereby achieving a comprehensive quantitative evaluation of the curve mask process correction effect.

[0061] Figure 5 This is a schematic block diagram of an evaluation system for curve mask process correction based on SEM image contour extraction, provided in an embodiment of this application. Figure 5 As shown, the system includes a model building module and a correction and evaluation module. It includes:

[0062] The model building module is used to obtain the initial curve pattern and the corresponding standard SEM image contour, and to build a simulation prediction model based on the initial curve pattern and the standard SEM image contour using a machine learning model.

[0063] The correction evaluation module is used to obtain a corrected curve pattern of the initial curve pattern using a curve mask method, and extract the corrected SEM image contour corresponding to the corrected curve pattern; input the corrected curve pattern into the simulation prediction model to generate a corrected simulation image of the corrected curve pattern, calculate the correction difference value between the corrected SEM image contour, the initial curve pattern and the corrected simulation image, and analyze the correction difference value to obtain the correction evaluation result.

[0064] It should be understood that the specific process of each module performing the above-mentioned corresponding steps has been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0065] It should also be understood that the module division in the embodiments of this application is illustrative and only represents a logical functional division; in actual implementation, there may be other division methods. Furthermore, the functional modules in the various embodiments of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0066] Figure 6 This is a schematic block diagram of the electronic terminal provided in an embodiment of this application. Figure 6 As shown, the electronic terminal 600 includes at least one processor 601, a memory 602, at least one network interface 603, and a user interface 605. The various components in the electronic terminal 600 are coupled together via a bus system 604. It is understood that the bus system 604 is used to implement communication between these components. In addition to a data bus, the bus system 604 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 6 The general will label all buses as bus systems.

[0067] The user interface 605 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touch screen.

[0068] It is understood that memory 602 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable categories of memory.

[0069] In this embodiment of the invention, the memory 602 is used to store various types of data to support the operation of the electronic terminal 600. Examples of this data include: any executable program for operation on the electronic terminal 600, such as the operating system 6021 and application program 6022; the operating system 6021 contains various system programs, such as the framework layer, core library layer, driver layer, etc., for implementing various basic services and handling hardware-based tasks. The application program 6022 may contain various applications, such as a media player, browser, etc., for implementing various application services. The evaluation method for curve mask process correction based on SEM image contour extraction provided in this embodiment of the invention can be included in the application program 6022.

[0070] The methods disclosed in the above embodiments of the present invention can be applied to processor 601, or implemented by processor 601. Processor 601 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 601 or by instructions in the form of software. The processor 601 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 401 can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present invention. General-purpose processor 401 may be a microprocessor or any conventional processor, etc. The steps of the accessory optimization method provided in the embodiments of the present invention can be directly reflected as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, which is located in a memory. The processor reads the information in the memory and combines it with its hardware to complete the steps of the aforementioned method.

[0071] In an exemplary embodiment, the electronic terminal 600 may be used by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) to perform the aforementioned method.

[0072] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to perform the method of any of the embodiments described above.

[0073] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code, which, when run on a computer, causes the computer to perform the method of any of the embodiments described above.

[0074] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0075] Those skilled in the art will recognize that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0076] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0078] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0079] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0080] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. A computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0081] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0082] In summary, this application provides an evaluation method, system, terminal, and medium for curve mask process correction based on SEM image contour extraction. The evaluation method for curve mask process correction based on SEM image contour extraction is characterized by: acquiring an initial curve pattern and its corresponding standard SEM image contour; constructing a simulation prediction model using a machine learning model based on the initial curve pattern and the standard SEM image contour; obtaining a corrected curve pattern of the initial curve pattern using a curve mask method, and extracting the corrected SEM image contour corresponding to the corrected curve pattern; inputting the corrected curve pattern into the simulation prediction model to generate a corrected simulation image of the corrected curve pattern; calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image; and analyzing the correction difference value to obtain a correction evaluation result. This application can more comprehensively and objectively reflect the contour accuracy and stability after curve mask process correction. Therefore, this application effectively overcomes various shortcomings in the prior art and has high industrial application value.

[0083] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. An evaluation method for curve mask process correction based on SEM image contour extraction, characterized in that, include: Obtain the initial curve pattern and the corresponding standard SEM image contour. Based on the initial curve pattern and the standard SEM image contour, construct a simulation prediction model using a machine learning model. A modified curve pattern of the initial curve pattern is obtained using a curve mask method, and the modified SEM image contour corresponding to the modified curve pattern is extracted; the modified curve pattern is input into the simulation prediction model to generate a modified simulation image of the modified curve pattern, the modification difference value between the modified SEM image contour, the initial curve pattern and the modified simulation image is calculated, and the modification difference value is analyzed to obtain the modification evaluation result.

2. The evaluation method for curve mask process correction based on SEM image contour extraction according to claim 1, characterized in that, Based on the initial curve pattern and the standard SEM image contour, a simulation prediction model is constructed using a machine learning model. The specific process includes: The initial curve pattern is input into a pre-built simulation prediction model to generate a corresponding initial simulation image, and the image difference value between the standard SEM image contour and the initial simulation image is calculated. When the image difference value meets the preset accuracy requirements, the pre-built simulation prediction model is used as the final simulation prediction model. When the image difference value does not meet the preset accuracy requirement, the parameters of the simulation prediction model are adjusted based on the initial curve pattern and the corresponding standard SEM image contour to obtain an updated simulation prediction model; the initial curve pattern is then input into the updated simulation prediction model to generate a new initial simulation image, and the image difference value between the standard SEM image contour and the new initial simulation image is recalculated until the image difference value meets the preset accuracy requirement to obtain the final simulation prediction model.

3. The evaluation method for curve mask process correction based on SEM image contour extraction according to claim 2, characterized in that, The specific process of calculating the image difference value between the standard SEM image contour and the initial simulation image includes: Align the standard SEM image contour with the initial simulation image in coordinates; The contour error value between the edge pixels corresponding to the contour of the standard SEM image and the edge pixels corresponding to the contour of the initial simulation image is calculated to obtain the image difference value.

4. The evaluation method for curve mask process correction based on SEM image contour extraction according to claim 1, characterized in that, The specific process for calculating the correction difference value between the corrected SEM image contour, the initial curve pattern, and the corrected simulation image includes: Align the modified SEM image contour with the modified simulation image, mark several first sampling points on the modified SEM image contour, obtain the edge intersection points of each first sampling point and the modified simulation image along the normal direction of each first sampling point, connect each first sampling point with the corresponding edge intersection point and calculate the first vector value corresponding to each first sampling point, and calculate the corresponding average value, absolute value, average value of absolute value and standard deviation based on all first vector values, as the first correction evaluation result; Align the initial curve pattern with the corrected simulation image, mark several second sampling points on the initial curve pattern, obtain the edge intersection points of each second sampling point and the corrected simulation image along the normal direction of each second sampling point, connect each second sampling point with the corresponding edge intersection points and calculate the second vector value corresponding to each second sampling point, and calculate the corresponding average value, absolute value, average value of absolute values ​​and standard deviation based on all second vector values, as the second correction evaluation result; The correction difference value is calculated based on the first and second correction evaluation results.

5. The evaluation method for curve mask process correction based on SEM image contour extraction according to claim 4, characterized in that, The specific process of calculating the corrected difference value based on the first corrected evaluation result and the second corrected evaluation result includes: The overall contour offset error is calculated by averaging the average of the absolute values ​​in the first corrected evaluation result and the average of the absolute values ​​in the second corrected evaluation result. The error dispersion index is obtained by averaging the standard deviations of the first and second corrected evaluation results. The corrected difference value is obtained by weighted summation of the overall contour offset error and the error discrete index.

6. The evaluation method for curve mask process correction based on SEM image contour extraction according to claim 1, characterized in that, The specific process for obtaining the standard SEM image contour includes: Based on edge detection technology, a number of partial SEM image contours with Manhattan structure features in the initial curve pattern are acquired; the acquisition differences are obtained by comparing the number of partial SEM image contours with the corresponding initial curve pattern; when the acquisition differences meet the preset acquisition rules, the SEM image contour of the initial curve pattern is acquired based on the edge detection technology as the preliminary SEM image contour. Obtain several preliminary SEM image contours of the initial curve pattern; The average image of the several preliminary SEM image contours is calculated as the standard SEM image contour of the initial curve pattern.

7. An evaluation system for curve mask process correction based on SEM image contour extraction, characterized in that, include: The model building module is used to obtain the initial curve pattern and the corresponding standard SEM image contour, and to build a simulation prediction model based on the initial curve pattern and the standard SEM image contour using a machine learning model. The correction evaluation module is used to obtain a corrected curve pattern of the initial curve pattern using a curve mask method, and extract the corrected SEM image contour corresponding to the corrected curve pattern; input the corrected curve pattern into the simulation prediction model to generate a corrected simulation image of the corrected curve pattern, calculate the correction difference value between the corrected SEM image contour, the initial curve pattern and the corrected simulation image, and analyze the correction difference value to obtain the correction evaluation result.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the evaluation method for curve mask process correction based on SEM image contour extraction as described in any one of claims 1 to 6.

9. A computer program product, characterized in that, The computer program product includes computer program code, which, when run on a computer, causes the computer to implement the evaluation method for curve mask process correction based on SEM image contour extraction as described in any one of claims 1 to 6.

10. An electronic terminal, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the evaluation method for curve mask process correction based on SEM image contour extraction as described in any one of claims 1 to 6.

Images