Optical proximity correction mask verification method, system, program product and terminal based on analysis of curve geometric characteristics

Abstract

The application provides an optical proximity correction mask verification method, system, program product and terminal based on curve geometric feature analysis. The mask pattern data after optical proximity effect correction is preprocessed and converted into a continuously derivable parameterized curve, the geometric features of the curve are quantitatively analyzed, an adaptive sampling strategy based on curvature change is adopted, the geometric feature values are calculated, and a quality evaluation report is outputted through the geometric feature values. Through the quantitative analysis of the geometric features of the mask pattern data and the all-round quality verification, the application solves the problem of incomplete defect identification in the traditional mask verification, balances the mask verification precision and the calculation efficiency, provides an explicit and feasible optimization direction for the optical proximity effect correction, helps the designers to quickly improve the mask pattern quality, significantly shortens the iteration period and reduces the research and development cost.

Description

Technical Field

[0001] This invention relates to the field of integrated circuit lithography, and in particular to an optical proximity correction mask verification method, system, program product, and terminal based on curve geometric feature analysis. Background Technology

[0002] As semiconductor manufacturing processes continue to shrink to the nanoscale, optical proximity correction (OPC) technology has become a crucial step in ensuring the accuracy of photolithography imaging. However, in complex OPC correction processes, especially when dealing with high-curvature patterns, traditional verification methods primarily focus on first-order geometric deviations such as edge placement error (EPE), neglecting the geometric quality characteristics of the curve itself. This limitation leads to the following problems:

[0003] First, traditional Design Rule Check (DRC) methods cannot effectively detect non-physical abrupt changes in curve curvature. During OPC correction, to compensate for optical proximity effects, the algorithm may introduce drastic changes in curvature in local areas. These curvature abrupt changes can produce unpredictable optical effects during photolithography imaging, leading to pattern distortion or even imaging failure. However, existing OPC verification tools lack the ability to quantitatively analyze curvature continuity, meaning these problems are often only discovered after tape-out, resulting in significant time and economic losses.

[0004] Secondly, the problem of excessive local complexity caused by overcorrection is becoming increasingly prominent. When attempting to meet stringent edge placement error (EPE) requirements, OPC correction algorithms may generate mask patterns containing complex structures such as microscopic jagged edges and over-spiral curves. These complex structures not only significantly increase the difficulty and cost of mask manufacturing but may also introduce new lithography imaging problems. Existing verification methods mainly focus on the macroscopic dimensional conformity of the pattern, lacking effective means to evaluate local geometric complexity.

[0005] Furthermore, current OPC verification systems suffer from a lack of dimensionality. Most OPC verification tools only provide analysis based on EPE and process windows, lacking a dedicated module for evaluating curve geometry quality. This deficiency in verification dimensions makes it difficult for designers to comprehensively assess the quality of mask patterns.

[0006] In actual production processes, traditional OPC verification methods have significantly impacted mask manufacturing and photolithography imaging by neglecting two-dimensional geometric defects such as abrupt changes in curve curvature and local high complexity introduced by OPC correction. Therefore, there is an urgent need for a verification method capable of evaluating the geometric quality of mask curves. Summary of the Invention

[0007] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide an optical proximity correction mask verification method, system, program product and terminal based on curve geometric feature analysis, to solve the technical problem that the existing OPC verification method ignores the two-dimensional geometric defects such as curve curvature abrupt change and local high complexity introduced by OPC correction, which ultimately affect mask manufacturing and photolithography imaging.

[0008] To achieve the above and other related objectives, this application provides a method for verifying optical proximity correction masks based on curve geometric feature analysis. The method includes: receiving mask image data after optical proximity effect correction; preprocessing the mask image data and converting the preprocessed mask image data into a parameterized curve for quantitative analysis of the curve's geometric features; adaptively sampling the parameterized curve using an adaptive sampling strategy and calculating geometric feature values; the calculated geometric feature values ​​include curvature change feature values ​​characterizing the smoothness of the curve and local curvature integral feature values ​​characterizing the local geometric complexity of the curve; determining whether the curve is qualified based on the curvature change feature values ​​and the local curvature integral feature values, and outputting a quality assessment report based on the judgment result.

[0009] In some embodiments of the first aspect of this application, the step of adaptively sampling the parameterized curve using an adaptive sampling strategy and calculating geometric feature values ​​includes: pre-setting a correspondence between multiple guiding curvature change intervals and sampling intervals; performing preliminary geometric analysis on the parameterized curve to obtain a guiding curvature change estimate within multiple analysis windows distributed along the curve arc length; determining the sampling interval of the corresponding analysis window based on the correspondence according to the interval where the guiding curvature change estimate is located; sampling the corresponding analysis window according to the determined sampling interval, and calculating the curvature change feature value of each sampling point and the local curvature integral feature value of each analysis window accordingly.

[0010] In some embodiments of the first aspect of this application, the local curvature integral eigenvalue is calculated using the following formula: Wherein, LCI is the local curvature integral eigenvalue, which is also the absolute value of the average curvature within the analysis window; Δs is the length of the analysis window; and κ(s) is the curvature function.

[0011] In some embodiments of the first aspect of this application, the length of the analysis window ranges from 5 nm to 15 nm.

[0012] In some embodiments of the first aspect of this application, the curvature change characteristic value is calculated using the following formula:

[0013] Where C_rate is the curvature change characteristic value; κ is the curvature; and s is the arc length.

[0014] In some embodiments of the first aspect of this application, the determination of whether a curve is qualified based on the curvature change feature value and the local curvature integral feature value includes: the curve is qualified when a first condition and a second condition are met simultaneously; wherein the first condition is configured such that the maximum value of the curvature change feature value of each sampling point in each analysis window is less than or equal to a preset curvature change feature threshold; and the second condition is configured such that the local curvature integral feature value of each analysis window is less than or equal to a preset local curvature integral feature threshold.

[0015] In some embodiments of the first aspect of this application, the quality assessment report includes location information of the problem area, classification information of the problem type, severity rating information of the problem, and specific optimization suggestions.

[0016] To achieve the above and other related objectives, a second aspect of this application provides an optical proximity correction mask verification system based on curve geometric feature analysis, comprising: a data preprocessing module, configured to receive mask image data after optical proximity effect correction, preprocess the mask image data, and convert the preprocessed mask image data into a parameterized curve for quantitative analysis of the geometric features of the curve; a feature extraction module, configured to adaptively sample the parameterized curve using an adaptive sampling strategy and calculate geometric feature values; the calculated geometric feature values ​​include curvature change feature values ​​characterizing the smoothness of the curve and local curvature integral feature values ​​characterizing the local geometric complexity of the curve; and a quality assessment and diagnosis module, configured to determine whether the curve is qualified based on the curvature change feature values ​​and the local curvature integral feature values, and output a quality assessment report based on the judgment result.

[0017] To achieve the above and other related objectives, a third aspect of the present invention provides a computer program product comprising computer program code that, when executed on a computer, causes the computer to implement the optical proximity correction mask verification method based on curve geometric feature analysis.

[0018] To achieve the above and other related objectives, a fourth aspect of the present invention 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 optical proximity correction mask verification method based on curve geometric feature analysis.

[0019] As described above, the optical proximity correction mask verification method, system, program product, and terminal based on curve geometric feature analysis of this application have the following beneficial effects: This invention preprocesses the mask image data after optical proximity effect correction and converts it into a continuously differentiable parameterized curve to quantitatively analyze the geometric features of the curve. It then calculates geometric feature values ​​using an adaptive sampling strategy based on curvature changes and outputs a targeted quality assessment report based on these geometric feature values. This invention, through quantitative analysis of the geometric features of mask image data and comprehensive quality verification, solves the problem of incomplete defect identification in traditional mask verification and balances mask verification accuracy with computational efficiency. It provides a clear and feasible optimization direction for optical proximity effect correction, helping designers quickly improve mask image quality, significantly shortening iteration cycles, and reducing R&D costs. Attached Figure Description

[0020] Figure 1 The diagram shown is a flowchart illustrating an optical proximity correction mask verification method based on curve geometric feature analysis according to an embodiment of the present invention.

[0021] Figure 2 The diagram shown is a flowchart of an adaptive sampling strategy according to an embodiment of the present invention.

[0022] Figure 3 The diagram shows a structural schematic of a curve region with a large predicted value of guiding curvature change in one embodiment of the present invention.

[0023] Figure 4 The diagram shows a structural schematic of a curve region with a small predicted value of guiding curvature change in one embodiment of the present invention.

[0024] Figure 5 The diagram shows a flowchart illustrating the implementation of an adaptive sampling strategy in one embodiment of the present invention.

[0025] Figure 6 The diagram shown is a structural schematic of an optical proximity correction mask verification system based on curve geometric feature analysis according to an embodiment of the present invention.

[0026] Figure 7 The diagram shown is an electronic terminal according to an embodiment of the present invention. 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] It should be noted that in the following description, reference is made to the accompanying drawings, which illustrate several embodiments of the present invention. It should be understood that other embodiments may also be used. In the embodiments of the present invention, the terms "first," "second," etc., are used to distinguish identical or similar items with substantially the same function and effect, without limiting their order. Those skilled in the art will understand that the terms "first," "second," etc., do not limit the quantity or execution order, and that "first," "second," etc., are not necessarily different.

[0029] Furthermore, in the embodiments of the present invention, the words "exemplary" or "for example" indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0030] Furthermore, in this embodiment of the invention, "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> Optical Proximity Correction (OPC): OPC is a key computational lithography technique. Its core purpose is to compensate for pattern distortion caused by factors such as optical diffraction, photoresist chemical and physical effects, and subsequent etching processes by intentionally and non-intuitively deforming and modifying the mask pattern in advance, so as to ensure that the final pattern formed on the wafer accurately matches the design requirements.

[0033] <2> Edge Placement Error (EPE): EPE refers to the deviation or error between the actual edge position of an integrated circuit pattern manufactured on a wafer and the expected ideal edge position in the design layout (target pattern). This error is one of the key indicators for measuring the accuracy of photolithography and pattern fidelity, and directly affects the performance, yield, and reliability of the circuit.

[0034] <3> Traditional Design Rule Checking (DRC): In the field of integrated circuit lithography, DRC refers to the automatic verification of the geometry, size, spacing, and topological relationships of the physical layout of an integrated circuit based on a fixed set of geometrically hierarchical constraint rules. This ensures that the layout meets the basic feasibility, yield, and reliability requirements of a specific manufacturing process (especially lithography) at the current technology node.

[0035] <4> Graphic Design System II (GDSII) file format: GDSII is the most widely used standard format in the field of semiconductor mask design and is gradually becoming an industry-wide standard. This format stores data in binary files, supports multi-layer graphic definitions (up to 65,535 layers), and can accurately record key data such as the coordinates of discrete boundary points of the mask pattern, graphic topological relationships, layer attributes (such as metal layers, dielectric layers, and doped layers), text annotations, and process information. It also has strong cross-platform compatibility and is the preferred format for delivering mask pattern data after optical proximity correction.

[0036] <5> Open Artwork System Interchange Standard (OASIS): OASIS format: This format employs a more efficient compression algorithm. Through techniques such as repeated graphic indexing and coordinate difference encoding, it can compress file size by 30%-70% compared to the GDSII format. It is particularly suitable for storing and transmitting mask graphic data containing massive amounts of tiny graphics in advanced processes. This format also supports multi-layer graphic definitions and topology recording, and is compatible with the core data fields of the GDSII format. It also features process rule annotation and graphic attribute extension functions, and can directly embed process constraint information (such as minimum linewidth and spacing requirements) after optical proximity correction.

[0037] <6> Drawing Exchange Format (DXF): The DXF format supports cross-software transmission of 2D / 3D graphic data and is occasionally used in the design of simple mask graphics (such as regular rectangles, straight line contours, etc.). However, due to the lack of semiconductor industry-specific layer attribute definition and topology description capabilities, and the fact that the discrete point accuracy can only reach the micrometer level, it is difficult to meet the high precision requirements of advanced process masks. It is usually only used for early concept design or verification of simple structure masks.

[0038] <7> Critical Dimension (CD): CD refers to the width or length of a specific structure in the semiconductor manufacturing process, especially those tiny dimensions that directly affect device performance. CD is a core indicator for evaluating the accuracy and consistency of manufacturing processes.

[0039] <8> Control points: In the field of parameterized curves for integrated circuits (especially NURBS), control points are core geometric elements used to define and manipulate the shape of parameterized curves. They are not necessarily located on the curve itself, but they influence and shape the curve's direction and form like "magnets" through specific mathematical functions (basis functions).

[0040] <9> Knot Vector: In the field of parametric curves for integrated circuits (especially NURBS), the knot vector is a core mathematical parameter defining the "hybrid rules" and "piecewise structure" of the parametric curve. Working in conjunction with control points, it precisely determines the shape of the basis functions, thus governing how the control points influence the final curve shape.

[0041] The present invention provides an optical proximity correction mask verification method, system, program product, and terminal based on curve geometric feature analysis. It preprocesses the mask image data after optical proximity effect correction and converts it into a continuously differentiable parameterized curve to quantitatively analyze the geometric features of the curve. Furthermore, it calculates geometric feature values ​​using an adaptive sampling strategy based on curvature changes and outputs a targeted quality assessment report based on these geometric feature values. The technical solutions in the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the invention.

[0042] like Figure 1 The diagram illustrates a flowchart of an optical proximity correction mask verification method based on curve geometric feature analysis, as described in an embodiment of the present invention. The optical proximity correction mask verification method based on curve geometric feature analysis in this embodiment includes the following steps:

[0043] Step S11: Receive the mask pattern data after optical proximity effect correction, preprocess the mask pattern data, and convert the preprocessed mask pattern data into a parametric curve for quantitative analysis of the geometric features of the curve.

[0044] It should be understood that this step, as the starting point of the optical proximity correction mask verification method of the present invention, has the core function of transforming the discrete mask pattern data after optical proximity effect correction into a parameterized curve form that can be accurately quantified and analyzed, thus laying the data foundation for subsequent adaptive sampling and calculation of geometric feature values.

[0045] The mask pattern data is standard mask design data output after optical proximity correction. Its data format supports mainstream mask pattern formats in the semiconductor industry, including but not limited to GDSII, OASIS, and DXF formats. This application embodiment preferentially uses the GDSII format because it has excellent vector graphics description capabilities, can accurately define the geometry, layer properties, and physical dimensions of multilayer masks, and can accommodate full-process accuracy requirements from the micrometer to the nanometer scale.

[0046] In this embodiment, the core content of the mask pattern data is the set of coordinates of all discrete points on the boundary of the mask pattern (i.e., the contour point data of the mask pattern). These discrete points are the discrete representation of the boundary of the mask pattern after optical proximity effect correction, and contain all the geometric details of the mask pattern (such as line edges, corners, arc contours, etc.).

[0047] It should be noted that although the mask pattern data after optical proximity effect correction has undergone geometric compensation, the following potential defects may still exist, which need to be eliminated through preprocessing to ensure the accuracy of subsequent parametric curve transformation and geometric feature calculation. These potential defects include:

[0048] Redundancy and repetition of discrete points: In the process of optical proximity effect correction, in order to characterize the complex contour after compensation (such as the "jagged" edge compensation caused by optical proximity effect), a large number of dense discrete points will be generated. Some areas may have duplicate points with the same coordinates, or redundant points with a spacing smaller than the preset redundancy spacing value (such as a spacing smaller than 0.001nm), which leads to data redundancy, increases the computational load of subsequent parametric curve conversion, and reduces conversion efficiency.

[0049] Discrete point order disorder: After optical proximity effect correction, the generation order of discrete points in complex graphics (such as intersecting lines, ring structures, etc.) in the mask pattern may be logically disordered (such as clockwise and counterclockwise order mixed), which will lead to problems such as self-intersection and contour distortion in the parameterized curves of subsequent transformations, and will not be able to truly reflect the design intent of the mask pattern.

[0050] Minor gaps and fractures: At the turning points of the mask pattern contour or in areas where the optical proximity effect correction compensation amount changes abruptly, minor gaps may appear due to the limitation of algorithm accuracy, such as an abnormally large increase in the spacing between discrete points (e.g., the spacing is greater than 0.1nm), forming a logical contour fracture. If not repaired, it will lead to discontinuity in the parametric curve transformation, affecting the accuracy of subsequent geometric feature value calculation.

[0051] Topological conflicts: For overlapping or nested graphics, the discrete points after optical proximity correction may have topological conflicts (such as the inner graphic contour intersecting or overlapping with the outer graphic contour), which makes it impossible to accurately distinguish the boundaries of different graphics during subsequent geometric feature analysis.

[0052] Noise interference: During the numerical iteration of the optical proximity effect correction algorithm, a small number of noise points that deviate from the theoretical contour may be introduced (such as discrete point coordinates deviating from the expected position by 0.01-0.05nm). These noise points will cause false abrupt changes in the subsequent curvature change feature value calculation, affecting the accuracy of mask quality determination.

[0053] Format compatibility issues: If the output after optical proximity correction is in a non-preferred format such as OASIS or DXF, the difference in format parsing by EDA tools may lead to loss of precision in discrete point coordinates (such as precision deviation during parsing of OASIS format compression encoding) or loss of layer attribute information, affecting the integrity of the data and the effectiveness of subsequent processing.

[0054] In view of this, the embodiments of this application specifically address the above-mentioned problems through a preprocessing stage. The main goal of preprocessing is to eliminate noise and topological defects in the mask image data, ensure the accuracy of subsequent parametric curve transformation, and avoid deviations in parametric curve transformation or errors in geometric feature calculation due to data defects. To facilitate understanding by those skilled in the art, the preprocessing operation is described below:

[0055] 1. Topology Check and Repair: Traverse the set of discrete points, specifically detect and repair various potential problems according to the following process:

[0056] To address the issue of redundant and repetitive discrete points: First, identify redundant points whose discrete point coordinates are completely identical or whose spacing is less than the preset redundancy spacing value (e.g., spacing less than 0.001nm). Then, based on the principle of preserving key geometric information, eliminate meaningless duplicate and redundant points, thereby reducing the amount of data without losing the geometric details of the mask pattern.

[0057] To address the issue of disordered discrete point order: First, based on the natural direction of the mask pattern outline (such as the general rule of the outer outline circling clockwise and the inner outline circling counterclockwise), determine whether the discrete point order is reasonable. Then, reorganize the discrete point order according to the rule of continuous extension along the boundary of the mask pattern to ensure that the mask pattern outline is coherent and not distorted.

[0058] For issues such as small gaps and breaks: First, the location and size of the gap are determined by detecting the distance between adjacent points. Then, the gap is filled by adding interpolation points to ensure a smooth transition of the mask pattern outline without breaks.

[0059] To address the issue of local intersections in graphic outlines: First, locate the conflicting areas where the outlines intersect. Then, based on the design logic of the mask graphic (such as the superposition relationship of different layers of mask graphics and functional zoning), decompose the conflicting areas and reorganize the discrete points to clarify the independent boundaries of each part of the mask graphic outline, thereby eliminating intersection conflicts.

[0060] To address the issue of boundary confusion in overlapping / nested graphics: First, based on the design intent of the mask graphics, distinguish the boundary affiliation of different mask graphics; then, through a process of layered marking and boundary redefinition, clarify the independent outline of each mask graphic to avoid boundary confusion.

[0061] To address format compatibility issues: Verify the data format parsing results, confirm whether layer attributes are complete and coordinate accuracy meets standards, then supplement missing layer attribute information and correct coordinate deviations to restore data integrity and accuracy.

[0062] 2. Data Denoising Processing: Specifically targeting the problem of discrete point jitter (i.e., noise interference), the following process is followed: First, identify noise points that deviate from the theoretical contour position (such as discrete points with a deviation exceeding 0.005nm), and then fine-tune the coordinates of the noise points. Only eliminate the random jitter interference introduced during the optical proximity effect correction process, without changing the core geometric features of the mask pattern (such as corner angles, curvature, key dimensions, etc.), ensuring that the original design intent of the mask pattern is not destroyed.

[0063] It should be noted that this preprocessing process only optimizes the integrity, orderliness, and accuracy of the discrete point data, and does not substantially modify the original design outline of the mask graphic data. All preprocessing operations are based on the principle of preserving the original design information and ensuring the accuracy of subsequent parametric curve conversion.

[0064] Furthermore, although the mask image data after optical proximity correction repairs issues such as topological defects and noise, it is still essentially a set of discrete point coordinates and cannot be directly used for geometric feature analysis (such as curvature, curvature variation features, and local curvature integral features). Therefore, it is necessary to upgrade the data format through parametric transformation, replacing isolated discrete coordinate points with a continuously differentiable mathematical model. This transforms the discrete data, which originally only represented the points of the mask image contour, into a standardized carrier that can accurately describe the overall shape of the mask image contour and support the calculation of geometric quantities at arbitrary locations (such as curvature and curvature variation features). The goal is to fully preserve all geometric details of the mask image (such as corners and curved contours) while meeting the requirements of continuous differentiability, high precision, and local controllability for subsequent analysis, laying a mathematical foundation for adaptive sampling and geometric feature extraction. The specific necessity includes the following three aspects:

[0065] Overcoming the limitations of discrete points: Discrete points can only represent isolated points of the mask shape contour and cannot cover the continuous area between points, resulting in the inability to directly solve geometric features such as curvature and curvature variation features; while parametric curves realize the continuous representation of the mask shape contour through a unified mathematical model, support the accurate calculation of geometric quantities at arbitrary arc length positions, and provide a basic mathematical carrier for subsequent adaptive sampling and geometric feature extraction.

[0066] Eliminating the error caused by uneven density of discrete points: After optical proximity effect correction, there is a distribution difference between densely packed points at corners and sparsely packed points on straight lines. Directly calculating geometric features based on discrete points will lead to local deviations due to uneven density. The parameterized curve eliminates this difference through a global fitting algorithm, ensuring that the calculation accuracy of curvature change feature values ​​and local curvature integral feature values ​​meets the mask verification requirements of advanced processes.

[0067] Improve the efficiency and engineering adaptability of the verification process: The number of discrete points in complex mask graphic data can reach millions. Direct analysis requires traversing all discrete points, resulting in low computational efficiency. Parametric curves represent the complete contour with a small number of parameters such as control points and node vectors, which greatly reduces the data scale. Moreover, its standardized mathematical model can be seamlessly connected to the geometric analysis module of mainstream electronic design automation tools, making it suitable for engineering application scenarios.

[0068] In view of this, the embodiments of this application further perform a parametric curve conversion operation after preprocessing. Specifically, the preprocessed mask pattern data is converted into a parametric curve. In mask pattern contour representation scenarios, selectable parametric curve types include Bezier curves, B-spline curves, rational B-spline curves, and non-uniform rational B-spline (NURBS) curves. For ease of understanding by those skilled in the art, the following detailed explanation uses the non-uniform rational B-spline (NURBS) curve as an example. This curve is the preferred option for representing advanced process mask pattern data, and its advantages include:

[0069] Global High-Precision Fitting: NURBS curves, through the parameterization characteristics of non-uniformly distributed node vectors, can specifically adapt to the non-uniform distribution characteristics of discrete points in the mask pattern after optical proximity effect correction. For areas with dense geometric details, such as corners and line ends, the algorithm automatically densifies the distribution density of node vectors, enhancing the fitting ability for local fine contours; for sparse discrete points, such as straight line segments and gentle arcs, the node vectors maintain a reasonable spacing to balance fitting efficiency, ultimately achieving global accurate fitting of all contour points of the mask pattern. The overall fitting error can be stably controlled within a preset range (e.g., preset fitting error ≤ 0.01nm), meeting the high-precision requirements of advanced semiconductor processes for mask pattern characterization.

[0070] Flexible local controllability: NURBS curves have an independent local contour adjustment mechanism. The curve shape is determined by the coordinate position of the control points and the weight coefficients. The control points in different regions have parameter independence. When it is necessary to correct the contour of a specific region (such as the smoothness of a corner or a local bulge), it is only necessary to adjust the coordinate offset or increase or decrease the weight coefficient of the local control points corresponding to that region to achieve precise correction of the contour of the target region. The control point parameters of the rest of the curve remain unchanged. Therefore, it will not affect the contour shape of non-target regions and can completely reproduce all the core geometric details of the mask graphic data.

[0071] Higher-order continuity and differentiability: The continuity of NURBS curves (such as positional continuity, tangent continuity, curvature continuity, etc.) can be precisely controlled through coordinated adjustment of the node repetition degree of the node vector and the order of the basis functions. By properly setting these parameters, it can be ensured that the generated curve has the required smoothness overall. In particular, achieving curvature continuity is crucial for the geometric feature analysis of this invention. By configuring an appropriate continuity level, it can be ensured that the NURBS curve model is sufficiently smooth and differentiable throughout the entire domain, thereby providing a stable and abrupt mathematical foundation for subsequent calculations of core geometric feature values ​​such as curvature and curvature variation characteristics, avoiding numerical anomalies or definition failures at feature points.

[0072] Full-type contour compatibility: Based on its unified rational polynomial mathematical model, NURBS curves have the ability to represent all types of geometric contours. Whether it is regular geometry (such as straight lines, standard circular arcs, elliptical arcs, etc.) or complex free curves formed after optical proximity effect correction, accurate representation can be achieved by adjusting the coordinates of control points, weight coefficients and node vector distribution. There is no need to switch fitting models for different geometric types. It can adapt to the diverse contour features of mask graphic data after optical proximity effect correction.

[0073] To facilitate understanding by those skilled in the art, the process of parametric transformation of NURBS curves is described below:

[0074] Initial control point layout: Based on the principle of full coverage and uniform distribution, control points are uniformly set on the entire contour of the mask pattern data at a preset fixed interval (e.g., 0.8nm) to ensure that the control points completely cover the distribution range of all discrete contour points, providing basic support for subsequent fitting and avoiding fitting deviation due to incomplete control point coverage.

[0075] Curve Fitting and Optimization: Based on the NURBS curve fitting algorithm, the goal is to minimize the fitting error between the parameterized curve and all discrete contour points. In detailed areas with dense discrete points, the algorithm automatically optimizes the coordinate positions or weight allocation of local control points to improve fitting accuracy. In areas with sparse discrete points, the algorithm maintains the uniform distribution of control points, ultimately achieving high-precision matching with the global fitting error controlled within a preset range (e.g., preset fitting error ≤ 0.01nm).

[0076] Through the above process, the discrete mask pattern contour is transformed into a continuous, complete, and high-precision NURBS parametric curve. This curve can completely preserve all the geometric details of the mask pattern data and supports the solution of geometric quantities at arbitrary positions with arc length as a parameter, providing a reliable and efficient analysis carrier for subsequent adaptive sampling and geometric feature value calculation.

[0077] Step S12: Adaptive sampling strategy is used to adaptively sample the parameterized curve and calculate geometric feature values; the calculated geometric feature values ​​include curvature change feature values ​​used to characterize the smoothness of the curve and local curvature integral feature values ​​used to characterize the local geometric complexity of the curve.

[0078] In one embodiment of this application, an adaptive sampling strategy is used to adaptively sample the parameterized curve and calculate geometric feature values. The execution flow is as follows: Figure 2 As shown, it includes the following steps:

[0079] Step S121: Pre-set the correspondence between multiple guiding curvature change intervals and sampling intervals.

[0080] In one embodiment of this application, the configuration principle of the correspondence is: the larger the estimated value of the guiding curvature change, the smaller the sampling interval determined according to the correspondence. This configuration principle ensures that in potentially high-risk curve regions with drastic curvature changes and complex geometric features, dense sampling points are used to capture possible micro-mutations; while in regions with gentle changes, relatively sparse sampling points are used to save computational resources.

[0081] To clearly illustrate the relationship between different guiding curvature change estimates and sampling intervals, we will now combine... Figure 3 and Figure 4 For example. Figure 3 This is an example of the distribution of sampling points 2 on curve region 1 (corresponding to densely geometrically detailed areas such as corners and local bulges in the mask pattern) where the indicative curvature change prediction is large. Figure 4 This is an example of the distribution of sampling points 4 on the curve region 3 (corresponding to the straight line segment, gentle arc, or other geometrically flat region of the mask pattern) where the indicative curvature change prediction is small.

[0082] from Figure 3 and Figure 4 As can be seen, the sampling interval d1 (exemplary value 1.0 nm) of sampling point 2 on curve region 1 with a large predicted value of guiding curvature change is significantly smaller than the sampling interval d2 (exemplary value 2.0 nm) of sampling point 4 on curve region 3 with a small predicted value of guiding curvature change.

[0083] The advantages of this distribution method are: Curve region 1 can accurately capture the geometric features of curvature abrupt changes through dense sampling with small spacing, avoiding the loss of details; Curve region 3 can reduce the amount of data and improve the verification efficiency while ensuring the accuracy of geometric feature calculation through sparse sampling with large spacing.

[0084] In some optional implementations, for ease of understanding by those skilled in the art, a specific example of a guiding curvature variation interval hierarchy is used below. It should be specifically noted that this example is merely a feasible way to illustrate the correspondence and does not constitute a limitation of the present invention. Those skilled in the art can adaptively adjust the hierarchy distribution and corresponding sampling intervals according to the accuracy requirements of actual process nodes (such as 5nm, 3nm, etc.). Table 1 below exemplarily presents the correspondence between the guiding curvature variation interval hierarchy and the sampling interval, as follows:

[0085] Table 1. Correspondence between guiding curvature variation ranges and sampling intervals (example)

[0086] Guiding curvature variation range hierarchy Instructional range of curvature change estimates Corresponding sampling interval range Low curvature variation range ≤0.01 nm⁻² 2.0-2.5nm Medium curvature variation range 0.01nm⁻² < estimated value ≤ 0.02nm⁻² 1.2-1.8nm High curvature variation range >0.02nm⁻² 0.5-1nm

[0087] For a guiding curvature change estimate falling within a certain curvature change range, a specific sampling interval value is determined within its corresponding sampling interval range. This specific sampling interval value can be achieved in various ways, such as directly taking the median of the corresponding sampling interval range, or, more preferably, using methods such as linear interpolation, so that the specific sampling interval value can be continuously adjusted as the guiding curvature change estimate changes.

[0088] In some preferred implementations, in order to achieve the technical effect of continuously and smoothly adjusting the specific sampling interval value according to the predicted value of the guiding curvature change, a linear interpolation method is used to determine the specific sampling interval value within the corresponding sampling interval range.

[0089] Specifically, for any range of curvature changes within which the guiding curvature change estimate falls:

[0090] When the estimated value of the guiding curvature change is equal to the lower limit of the curvature change range, the maximum sampling interval (i.e. the sparsest sampling density) within the corresponding sampling interval range is adopted.

[0091] When the estimated value of the guiding curvature change is equal to the upper limit of the curvature change range, the minimum sampling interval (i.e. the densest sampling density) within the corresponding sampling interval range is adopted.

[0092] When the predicted curvature change falls between the upper and lower limits of the curvature change interval, the corresponding sampling interval value is calculated proportionally based on the relative position of the predicted curvature change within the interval. The closer the predicted curvature change is to the upper limit of the curvature change interval, the closer the calculated sampling interval value is to the minimum sampling interval within the corresponding range; conversely, the closer the predicted curvature change is to the lower limit of the curvature change interval, the closer the calculated sampling interval value is to the maximum sampling interval within the corresponding range.

[0093] To facilitate understanding by those skilled in the art, the curvature variation range in Table 1 is used as an example: If the estimated value of the guiding curvature variation of a certain analysis window on the curve is 0.015 nm⁻², which is exactly in the middle of the range of curvature variation, then the sampling interval value calculated by linear interpolation will also be in the middle of the corresponding sampling interval range, i.e., 1.5 nm; if the estimated value of the guiding curvature variation is closer to 0.02 nm⁻², then the calculated sampling interval value will be closer to 1.2 nm; if the estimated value of the guiding curvature variation is closer to 0.01 nm⁻², then the calculated sampling interval will be closer to 1.8 nm.

[0094] Step S122: Perform preliminary geometric analysis on the parameterized curve to obtain a guiding curvature change estimate within multiple analysis windows distributed along the curve arc length.

[0095] This step aims to predict the degree of curvature change in each local region (analysis window) through a rapid preliminary geometric analysis, thereby providing a basis for subsequent adaptive sampling interval allocation and avoiding the efficiency bottleneck caused by direct high-precision calculation across the entire domain.

[0096] Specifically, firstly, along the arc length direction of the parameterized curve, a series of continuous and non-overlapping analysis windows are divided with a preset window length (e.g., 5nm to 15nm). The analysis windows extend sequentially from the starting point of the curve. If the length of the last analysis window is less than the preset window length value, the analysis window range is supplemented with the end point of the curve as the boundary, ensuring that the entire curve is covered.

[0097] Subsequently, a low-precision, rapid preliminary geometric analysis is performed on the parametric curves (such as NURBS curves) of each analysis window. One feasible approach is to select a few representative points within the analysis window (such as the start, midpoint, and end point of the analysis window), quickly estimate the curvature of these points using a continuous differentiable model of the parametric curve, and represent the curvature change magnitude of the analysis window by comparing the differences between these curvature values ​​(such as the difference between the maximum and minimum values, or the variance). Finally, this curvature change magnitude is used as a guiding estimate of the curvature change for the analysis window.

[0098] It should be noted that this process focuses on computational speed, and the resulting indicative curvature change estimates are intended to reliably indicate the relative strength of curvature changes between different analysis windows, rather than precisely calculating their absolute values.

[0099] In some preferred implementations, the length of the analysis window is set to a range of 5 nm to 15 nm. The selection of this range takes into account the physical characteristics of advanced semiconductor manufacturing processes, the typical scale of mask pattern defects, and algorithm efficiency. The specific basis includes the following:

[0100] Adapting to the geometric feature size of advanced process masks: In the verification scenarios of optical proximity correction masks for 3nm and 5nm advanced semiconductor processes, the geometric anomalies introduced by optical proximity correction (such as non-physical curvature abrupt changes, microscopic jagged edges, etc.) typically have feature sizes on the order of several nanometers to tens of nanometers. For example, the effective arc length of a sharp corner or local bulge that needs to be detected is often in the range of several nanometers. Setting the analysis window length to 5-15nm ensures that a single analysis window is sufficient to completely cover a typical local defect region, thereby enabling the calculated guiding curvature change estimate to effectively characterize the severity of the local defect. If the analysis window length is less than 5nm, the narrow coverage area may not be able to fully encompass the entire region of a single local defect, leading to deviations in the predicted curvature change and failing to accurately reflect the curvature change trend of the region. If the analysis window length is greater than 15nm, the wide coverage area may include multiple different local defect regions in the same analysis window, resulting in distorted averaging of the predicted curvature change and an inability to accurately distinguish the differences in curvature changes among different local defect regions.

[0101] Ensuring Statistical Stability and Noise Suppression: The analysis window needs to be based on a sufficient number of sampling points for statistical calculations to ensure the robustness of the results to random noise. If the analysis window length is less than 5 nm, the number of sampling points included within the window will be below the minimum threshold required for effective statistical calculations. Statistical quantities calculated based on this finite sample set (such as the difference in curvature values) are significantly affected by the random bias of individual sampling points, leading to drastic fluctuations in the indicative curvature change estimate unrelated to geometric features, thus reducing reliability. If the analysis window length is greater than 15 nm, the number of sampling points within the window increases, but the arc length of the curve covered by the window exceeds the typical scale of a single local feature. This results in different local defect regions being included in the same statistical calculation, producing an averaging effect. This makes the estimated indicative curvature change estimate unable to accurately reflect the severity of curvature changes in specific local defect regions within the analysis window, weakening its guiding value as a basis for subsequent differentiated sampling. If the analysis window length is within the range of 5-15 nm, it can provide a stable data foundation for the rapid evaluation of the indicative curvature change estimate, effectively suppressing the impact of noise while ensuring speed.

[0102] Balancing computational efficiency of the prediction: If the analysis window length is less than 5nm, the entire parameterized curve will be divided into too many analysis windows (for example, a curve with a length of 1μm will be divided into more than 200 analysis windows), which will significantly increase the computational load of the preliminary geometric analysis and reduce the efficiency advantage of rapid prediction; if the analysis window length is greater than 15nm, although the number of analysis windows can be reduced, the prediction will be distorted due to the change in guiding curvature, forcing the subsequent sampling stage to add an additional verification step to correct the deviation, which will reduce the overall process efficiency.

[0103] Therefore, setting the analysis window length within the range of 5nm to 15nm can not only accurately adapt to the core geometric feature scale of advanced process masks, ensuring the accuracy and discriminativeness of the guiding curvature change prediction, but also take into account the computational efficiency of the preliminary geometric analysis, providing a key scale benchmark for the subsequent implementation of adaptive sampling strategies.

[0104] Step S123: Based on the interval where the guiding curvature change estimate is located, determine the sampling interval of the corresponding analysis window according to the correspondence; sample the corresponding analysis window according to the determined sampling interval, and calculate the curvature change feature value of each sampling point and the local curvature integral feature value of each analysis window accordingly.

[0105] It should be understood that this step is the implementation phase of the adaptive sampling strategy, such as... Figure 5 As shown, it specifically includes:

[0106] Step S1231: Based on the interval where the guiding curvature change estimate is located, determine the sampling interval of the corresponding analysis window according to the correspondence, and sample the corresponding analysis window according to the determined sampling interval.

[0107] First, for each analysis window, the corresponding estimated value of the guiding curvature change is compared with the pre-defined correspondence between multiple guiding curvature change intervals and sampling intervals to determine the sampling interval for each analysis window (for example, if the estimated value of the guiding curvature change for a certain analysis window is 0.015nm⁻², it falls into the medium curvature change interval, and the corresponding sampling interval is 1.5nm). Then, sampling is performed in the corresponding analysis window according to the corresponding sampling interval for each analysis window.

[0108] Step S1232: Calculate the curvature change characteristic value of each sampling point within each analysis window.

[0109] For each sampling point within each analysis window, based on a continuously differentiable mathematical model of a parametric curve (such as a NURBS curve), the curvature κ of each sampling point is first solved (e.g., calculated using the second-order differential equation of the NURBS curve). Then, a high-precision numerical differentiation algorithm (preferably the central difference method, which has higher computational accuracy and lower sensitivity to random noise compared to forward and backward difference methods, effectively avoiding computational bias caused by single-direction difference) is used to calculate its curvature change characteristic value. Each sampling point corresponds to a curvature change characteristic value, thus fully capturing the local smoothness of the curve corresponding to the analysis window.

[0110] The curvature variation characteristic value is calculated using the following formula:

[0111] ;(Formula 1)

[0112] In Formula 1, C_rate is the curvature change feature value, the magnitude of which directly characterizes the local smoothness of the curve (the larger the value (e.g., greater than the preset curvature change feature threshold), the more significant the local curvature abrupt change of the curve, and the corresponding mask pattern may have potential photolithographic defects such as burrs and sharp corners; the smaller the value (e.g., less than the preset curvature change feature threshold), the smoother the local shape of the curve, and the stronger the photolithographic adaptability of the mask pattern); κ is the curvature of the sample points, which is obtained by solving the second-order differential equation of the parameterized curve; s is the arc length parameter of the parameterized curve.

[0113] Step S1233: Calculate the local curvature integral eigenvalues ​​for each analysis window.

[0114] For each analysis window, the continuous curvature function k(s) (such as the curvature equation of the parameterized curve) is first obtained based on the parameterized curve (such as the NURBS curve) within the analysis window. Then, the average value of the absolute curvature within the analysis window is calculated using a numerical integration algorithm (preferably the trapezoidal integral method, which has strong adaptability to discrete sampling points and low computational complexity, and can ensure integration accuracy while taking into account the efficiency of the overall verification process, making it suitable for rapid analysis of large-scale mask patterns). This average value is the local curvature integral characteristic value of the analysis window, used to characterize the local geometric complexity of the curve corresponding to the analysis window.

[0115] The eigenvalues ​​of the local curvature integral are calculated using the following formula:

[0116] ;(Formula 2)

[0117] In Formula 2, LCI is the local curvature integral eigenvalue, which is essentially the absolute value of the average curvature within the analysis window; Δs is the length of the analysis window; κ(s) is the continuous curvature function of the parameterized curve within the analysis window, which can be obtained by solving the second-order differential equation of the NURBS curve or by sampling point curvature interpolation.

[0118] It should be noted that the magnitude of the LCI value is directly related to the lithographic adaptability of the mask pattern: the larger the LCI value (e.g., greater than the preset local curvature integral feature threshold), the more discrete the curvature distribution of the curve within the analysis window and the more complex the geometric shape (e.g., the presence of dense corners, microscopic jagged edges, etc.). These structures may not be clearly imaged during lithography, or may increase the difficulty and cost of mask manufacturing; the smaller the LCI value (e.g., less than the preset local curvature integral feature threshold), the more regular and smooth the curve shape within the analysis window, and the better the consistency and accuracy of the lithographic transfer of the mask pattern.

[0119] In one embodiment of this application, the criterion for determining whether a curve is qualified based on the curvature change characteristic value and the local curvature integral characteristic value includes:

[0120] The parameterized curve is deemed a qualified curve when it simultaneously meets the first and second conditions below; if either condition is not met, it is deemed an unqualified curve.

[0121] First condition: Within each analysis window, the maximum value of the curvature change characteristic value of all sampling points is less than or equal to the preset curvature change characteristic threshold. If the maximum value of the curvature change characteristic value of all sampling points within a certain analysis window exceeds the preset curvature change characteristic threshold, it indicates that there are local defects such as sharp burrs or non-physical curvature abrupt changes in the curve region corresponding to that analysis window. Such local traps can easily lead to edge jaggedness or linewidth deviation in the wafer pattern during the photolithography process.

[0122] The second condition is that the local curvature integral characteristic value of each analysis window is less than or equal to the preset local curvature integral characteristic threshold. If the local curvature integral characteristic value of a certain analysis window exceeds the preset local curvature integral characteristic threshold, it indicates that the geometric shape of the curve region corresponding to the analysis window is complex (such as the presence of dense corners, microscopic jagged edges, etc.), which can easily increase the difficulty of mask manufacturing and affect the clarity of photolithography imaging.

[0123] It should be noted that the two preset thresholds mentioned above are not fixed values ​​and can be flexibly adjusted according to the actual application scenario: for more advanced processes such as 3nm and 2nm, the preset thresholds can be appropriately reduced to improve the curve quality requirements; for large-size, low-complexity mask patterns (such as peripheral wiring areas), the preset thresholds can be appropriately relaxed to balance verification efficiency and process requirements.

[0124] In one embodiment of this application, a mask pattern quality assessment report is generated based on the judgment result of whether the above-mentioned curve is qualified. The report includes the following information:

[0125] 1. Precise location information of the problem area

[0126] This section needs to clearly label the mask graphic areas corresponding to all non-compliant curves, clarifying both the physical and geometric location of the problem areas and linking them to the specific abnormal feature values ​​that caused the non-compliance. This provides accurate location data and technical attribution for subsequent defect correction. Examples include: geometric coordinate location information, curve segment location information, analysis window and feature value location information, etc.

[0127] 2. Detailed classification information of problem types

[0128] Based on the type of the excessive feature value, the defect type of the problem area is classified. The classification criteria directly correspond to the technical meaning of the geometric features mentioned above, and include, for example:

[0129] If the curvature change characteristic value exceeds the standard and is deemed unqualified, it is classified as "micro-geometric defect", which is further subdivided into subtypes such as sharp burrs, non-physical curvature abrupt changes, and edge jaggedness.

[0130] If the local curvature integral characteristic value exceeds the standard and is deemed unqualified, it is classified as "macromorphic defect", which is further subdivided into subtypes such as dense corner clusters, micro-serrated patches, and uneven curvature distribution.

[0131] 3. Severity rating information of the problem

[0132] By combining the precision requirements of advanced manufacturing processes (such as 3nm / 5nm processes) with quantitative indicators and regional importance, a tiered evaluation of mask pattern data problems can be achieved, providing a basis for subsequent differentiated optimization decisions. For example, based on the extent of exceeding geometric eigenvalue limits and the range of defect impact, it can be divided into three levels: mild, moderate, and severe, including:

[0133] Mild: The eigenvalue exceeds the limit by ≤20% and the problem area only covers a single analysis window. The impact on lithography can be eliminated by optical proximity correction.

[0134] Moderate: The eigenvalue exceeds the standard by 20%-50% or the problem area covers 2-3 analysis windows, requiring targeted adjustment of the parametric curve fitting strategy;

[0135] Severe: If the feature value exceeds the standard by more than 50% or the problem area covers more than 3 analysis windows, it will directly cause the lithography pattern to fail, and the local design of the mask pattern data needs to be redesigned.

[0136] 4. Targeted optimization suggestions

[0137] For different types and severity of problems, we provide actionable optimization suggestions, such as:

[0138] For micro-geometric defects (excessive curvature change characteristic values), it is recommended to backtrack to the corresponding control points of the NURBS curve and achieve smooth correction of the curve profile by fine-tuning the weights or coordinates of local control points (such as adjusting the weight of the control point at the corner from 1.2 to 0.8).

[0139] To address the macroscopic morphological defects (excessive local curvature integral eigenvalues), it is recommended to re-divide the analysis window of the region (e.g., adjust the 15nm window to 10nm) and optimize the sampling interval, or adjust the local algorithm parameters for optical proximity correction to reduce the geometric complexity of the region.

[0140] By integrating the above information, the quality assessment report can provide a comprehensive and accurate evaluation of the geometric quality of mask pattern data, offering clear technical guidance for subsequent mask correction and ensuring that the final mask pattern data meets the lithography application requirements of advanced semiconductor processes.

[0141] like Figure 6 The diagram illustrates a structural schematic of an optical proximity correction mask verification system 600 based on curve geometric feature analysis according to an embodiment of the present invention. The optical proximity correction mask verification system in this embodiment includes: a data preprocessing module 601, a feature extraction module 602, and a quality assessment and diagnosis module 603.

[0142] The data preprocessing module 601 is used to receive the mask pattern data after optical proximity effect correction, preprocess the mask pattern data, and convert the preprocessed mask pattern data into a parametric curve for quantitative analysis of the geometric characteristics of the curve.

[0143] The feature extraction module 602 is used to adaptively sample the parameterized curve using an adaptive sampling strategy and calculate geometric feature values; the calculated geometric feature values ​​include curvature change feature values ​​used to characterize the smoothness of the curve and local curvature integral feature values ​​used to characterize the local geometric complexity of the curve.

[0144] The quality assessment and diagnosis module 603 is used to determine whether the curve is qualified based on the curvature change characteristic value and the local curvature integral characteristic value, and output a quality assessment report based on the judgment result.

[0145] It should be noted that the optical proximity correction mask verification system based on curve geometric feature analysis provided in this embodiment of the invention is similar in principle and process to the optical proximity correction mask verification method based on curve geometric feature analysis described above, and will not be repeated here. 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.

[0146] It should also be understood that the module division in the embodiments of the present invention is illustrative and only represents one logical functional division; in actual implementation, there may be other division methods. Furthermore, the functional modules in the various embodiments of the present invention can be integrated into a single processor, exist as separate physical entities, or two or more modules can be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0147] like Figure 7 The diagram shown is a schematic of an electronic terminal provided in an embodiment of this application. The electronic terminal includes at least one processor 701, a memory 702, at least one network interface 703, and a user interface 705. The various components in the device are coupled together via a bus system 704. It is understood that the bus system 704 is used to implement communication between these components. In addition to a data bus, the bus system 704 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 7 The general will label all buses as bus systems.

[0148] The user interface 705 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touch screen.

[0149] It is understood that memory 702 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.

[0150] In this embodiment of the invention, the memory 702 is used to store various types of data to support the operation of the electronic terminal 700. Examples of this data include: any executable program for operation on the electronic terminal 700, such as the operating system 7021 and application program 7022; the operating system 7021 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 7022 may contain various applications, such as a media player, browser, etc., for implementing various application services. The method for testing the lateral positioning accuracy of agricultural machinery provided in this embodiment of the invention can be included in the application program 7022.

[0151] The methods disclosed in the above embodiments of the present invention can be applied to processor 701, or implemented by processor 701. Processor 701 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 701 or by instructions in software form. The processor 701 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 701 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. General-purpose processor 701 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 memory. The processor reads the information in the memory and combines it with its hardware to complete the steps of the aforementioned method.

[0152] In an exemplary embodiment, the electronic terminal 700 may be used to execute the aforementioned method by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs).

[0153] According to the method provided in the embodiments of the present invention, the present invention also provides a computer program product, the computer program product comprising: computer program code, which, when executed on a computer, causes the computer to execute... Figure 1The optical proximity correction mask verification method based on curve geometric feature analysis in any of the embodiments shown.

[0154] 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).

[0155] 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.

[0156] 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.

[0157] 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.

[0158] 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.

[0159] 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.

[0160] 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. The 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, etc.).

[0161] 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.

[0162] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0163] In summary, the optical proximity correction mask verification method, system, program product, and terminal based on curve geometric feature analysis provided by this invention preprocesses the mask image data after optical proximity effect correction and converts it into a continuously differentiable parameterized curve to quantitatively analyze the geometric features of the curve. It then calculates geometric feature values ​​using an adaptive sampling strategy based on curvature changes and outputs a targeted quality assessment report based on these geometric feature values. This invention, through quantitative analysis of the geometric features of mask image data and comprehensive quality verification, solves the problem of incomplete defect identification in traditional mask verification and balances mask verification accuracy and computational efficiency. It provides a clear and feasible optimization direction for optical proximity effect correction, helping designers quickly improve mask image quality, significantly shortening iteration cycles, and reducing R&D costs.

[0164] Therefore, this application effectively overcomes the various shortcomings of the prior art and has high industrial application value.

[0165] 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. A method of optical proximity correction mask verification based on analysis of geometric features of curves, characterized in that, include: The system receives mask pattern data after optical proximity effect correction, preprocesses the mask pattern data, and converts the preprocessed mask pattern data into a parametric curve for quantitative analysis of the geometric characteristics of the curve. An adaptive sampling strategy is used to adaptively sample the parameterized curve and calculate geometric feature values. The calculated geometric feature values ​​include curvature change feature values ​​that characterize the smoothness of the curve and local curvature integral feature values ​​that characterize the local geometric complexity of the curve. The local curvature integral eigenvalue is calculated using the following formula: ； Wherein, LCI is the local curvature integral eigenvalue, which is also the absolute value of the average curvature within the analysis window; Δs is the length of the analysis window; κ(s) is the curvature function; The curvature change characteristic value is calculated using the following formula: ； Where C_rate is the curvature change characteristic value; κ is the curvature; and s is the arc length. The curve is judged to be qualified based on the curvature change characteristic value and the local curvature integral characteristic value, and a quality assessment report is output based on the judgment result; The curve is qualified when both the first and second conditions are met simultaneously. The first condition is configured such that the maximum value of the curvature change feature value of the sampling point in each analysis window is less than or equal to a preset curvature change feature threshold; the second condition is configured such that the local curvature integral feature value of each analysis window is less than or equal to a preset local curvature integral feature threshold.

2. The method of claim 1, wherein the method is based on analysis of geometric features of curves. The step of adaptively sampling the parameterized curve using an adaptive sampling strategy and calculating geometric feature values ​​includes: Multiple guiding curvature variation intervals and their corresponding sampling intervals are pre-defined; A preliminary geometric analysis is performed on the parameterized curve to obtain a guiding estimate of curvature change within multiple analysis windows distributed along the curve's arc length; Based on the interval in which the guiding curvature change estimate is located, the sampling interval of the corresponding analysis window is determined according to the correspondence; the corresponding analysis window is sampled according to the determined sampling interval, and the curvature change characteristic value of each sampling point and the local curvature integral characteristic value of each analysis window are calculated accordingly.

3. The method of optical proximity correction mask verification based on analysis of geometrical features of curves as claimed in claim 1, wherein, The length of the analysis window ranges from 5 nm to 15 nm.

4. The method of claim 1, wherein the method is performed by a computer system. The quality assessment report includes information on the location of the problem area, the classification of the problem type, the severity rating of the problem, and specific optimization suggestions.

5. An optical proximity correction mask verification system based on analysis of geometric features of curves, characterized in that, include: The data preprocessing module is used to receive mask pattern data after optical proximity effect correction, preprocess the mask pattern data, and convert the preprocessed mask pattern data into a parametric curve for quantitative analysis of the geometric characteristics of the curve. The feature extraction module is used to adaptively sample the parameterized curve using an adaptive sampling strategy and calculate geometric feature values. The calculated geometric feature values ​​include curvature change feature values ​​that characterize the smoothness of the curve and local curvature integral feature values ​​that characterize the local geometric complexity of the curve. The local curvature integral eigenvalue is calculated using the following formula: ； Wherein, LCI is the local curvature integral eigenvalue, which is also the absolute value of the average curvature within the analysis window; Δs is the length of the analysis window; κ(s) is the curvature function; The curvature change characteristic value is calculated using the following formula: ； Where C_rate is the curvature change characteristic value; κ is the curvature; and s is the arc length. The quality assessment and diagnosis module is used to determine whether the curve is qualified based on the curvature change characteristic value and the local curvature integral characteristic value, and output a quality assessment report based on the judgment result. The curve is qualified when both the first and second conditions are met simultaneously. The first condition is configured such that the maximum value of the curvature change feature value of the sampling point in each analysis window is less than or equal to a preset curvature change feature threshold; the second condition is configured such that the local curvature integral feature value of each analysis window is less than or equal to a preset local curvature integral feature threshold.

6. A computer program product, characterised in that, The computer program product includes computer program code, which, when run on a computer, causes the computer to implement the optical proximity correction mask verification method based on curve geometric feature analysis as described in any one of claims 1 to 4.

7. An electronic terminal comprising a memory, a processor and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the optical proximity correction mask verification method based on curve geometric feature analysis as described in any one of claims 1 to 4.

Images