A simulation image generation method of a simulation scanning electron microscope and a related device

Abstract

The application provides a simulation image generation method of a simulation scanning electron microscope and related equipment, and the method comprises the following steps: based on scanning electron microscope image data of a real wafer, a first imaging feature mapping table and a second imaging feature mapping table are respectively constructed; based on the two imaging feature mapping tables, a standard imaging script and a defect imaging script are respectively made; a wafer three-dimensional model is acquired and associated process attribute information in each structure unit is analyzed; based on the process attribute information, a corresponding standard imaging script is determined, surface characteristics of the structure unit are adjusted, and a defect-free simulation model is generated; according to a preset defect introduction condition, a corresponding defect imaging script is determined, and defects are added to specified structure units to generate a simulation model with defects; and finally, a simulation image of a simulation scanning electron microscope is generated. Through virtual simulation means, a high-simulation scanning electron microscope image is batch-manufactured, and the diversity and coverage of the simulation image data set are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor technology, and in particular to a simulation image generation method of simulating a scanning electron microscope and related equipment. BACKGROUND

[0002] As a key detection device, a scanning electron microscope (SEM) is widely used in the process of semiconductor manufacturing. By focusing an electron beam to scan the surface of a sample and collecting secondary electrons or backscattered electrons, the SEM can generate high-resolution images to provide information about the morphology, structure, and composition of the sample. Traditionally, engineers observe SEM photos by naked eye to detect whether there are defects on the wafer surface. However, with the continuous increase in wafer size and the continuous improvement in device integration, the manual detection method gradually exposes problems such as low efficiency, strong subjectivity, and high missed detection rate.

[0003] To improve detection efficiency, image recognition programs based on artificial intelligence (AI) are currently widely used to detect defects in SEM photos. The performance and accuracy of AI recognition models depend largely on the richness and diversity of training data. In actual production, due to the limited types and distribution of naturally formed defects, and the high cost of collecting high-quality and labeled defect samples, the scale of training data is limited, which affects the accuracy and generalization ability of AI defect recognition models.

[0004] Therefore, how to generate a large number of rich and diverse simulation images of scanning electron microscopes at a low cost and in a controllable manner, especially training samples containing different types of defects, to support the efficient training and performance improvement of AI defect recognition models, has become a technical problem to be solved. SUMMARY

[0005] To address the deficiencies in the prior art, the present application provides a simulation image generation method of simulating a scanning electron microscope and related equipment, at least to solve the problem that there is a lack of high-quality and diverse simulation image data of scanning electron microscopes to support artificial intelligence defect detection training in the prior art.

[0006] To achieve the above-mentioned purposes and other advantages, some embodiments of the present application provide the following aspects:

[0007] In a first aspect, some embodiments of the present application provide a simulation image generation method of simulating a scanning electron microscope, comprising:

[0008] Based on real wafer-based scanning electron microscope image data, a first imaging feature mapping table and a second imaging feature mapping table are respectively constructed, the first imaging feature mapping table is used to establish a mapping relationship between different material types and process recipes and corresponding standard imaging features in a defect-free state; the second imaging feature mapping table is used to establish a mapping relationship between defect types and corresponding defect imaging features, the defect imaging features are used to describe the imaging performance of defects in the scanning electron microscope image;

[0009] Based on each standard imaging feature of the first imaging feature mapping table and each defect imaging feature of the defect type and defect imaging feature mapping table, corresponding standard imaging scripts and defect imaging scripts are respectively made;

[0010] A wafer three-dimensional model containing a plurality of structure units is obtained, and the associated process attribute information of each structure unit is analyzed, the process attribute information including: material type, process recipe information and mask pattern information;

[0011] Based on the process attribute information, the corresponding standard imaging script is determined by querying the first imaging feature mapping table, and the surface characteristics of the structure unit are adjusted based on the standard imaging script to generate a defect-free simulation model;

[0012] Based on the generation of the defect-free simulation model, according to the preset defect introduction condition, the corresponding defect imaging script is determined by querying the second imaging feature mapping table, and defects are added on the specified structure unit based on the defect imaging script to generate a simulation model with defects;

[0013] Based on the defect-free simulation model and the simulation model with defects, a simulation image of a simulated scanning electron microscope is generated.

[0014] In a second aspect, some embodiments of the present application also provide an electronic device, which comprises:

[0015] One or more processors; and a memory storing computer program instructions, which when executed cause the processor to perform the simulation image generation method of the simulated scanning electron microscope as described above.

[0016] In a third aspect, some embodiments of the present application also provide a computer readable storage medium, which stores computer programs and / or instructions, which when executed by a processor, implement the simulation image generation method of the simulated scanning electron microscope as described above.

[0017] In a fourth aspect, some embodiments of the present application further provide a computer program product comprising computer programs and / or instructions, which, when executed by a processor, implement the simulation image generation method of the simulation scanning electron microscope as described above.

[0018] Compared with the related art, in the scheme provided by the embodiments of the present application, the imaging feature mapping table is established, the three-dimensional model process attributes of the wafer are automatically analyzed, the simulation script is intelligently called to construct the defect-free and defective models, and the high-simulation scanning electron microscope images are batch-manufactured through virtual simulation means. Not only the automation and standardization degree of the simulation image generation is improved, but also the imaging details under the normal and abnormal states of the wafer structure are effectively restored, the actual defect image acquisition cost is significantly reduced, the diversity and coverage of the simulation image data set are improved, and then the efficiency and accuracy of the defect detection model training and verification are improved. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other embodiments can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0020] Figure 1 is a flowchart of a simulation image generation method of a simulation scanning electron microscope provided by the embodiments of the present application;

[0021] Figures 2a to 2d is a process control effect diagram of defect insertion at a structure unit level provided by the embodiments of the present application;

[0022] Figures 3a to 3c is a scanning electron microscope simulation image effect diagram based on multi-view acquisition provided by the embodiments of the present application;

[0023] Figure 4 is a structural diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0025] First embodiment

[0026] The first embodiment of the present application relates to a simulation image generation method of a simulated scanning electron microscope, referring to Figure 1 The method can include the following steps:

[0027] Step S1: Based on the scanning electron microscope image data of the real wafer, a first imaging feature mapping table and a second imaging feature mapping table are respectively constructed, the first imaging feature mapping table is used to establish the mapping relationship between different material types and process recipes in their defect-free state and the corresponding standard imaging features; the second imaging feature mapping table is used to establish the mapping relationship between the defect type and the corresponding defect imaging feature, and the defect imaging feature is used to describe the imaging performance of the defect in the scanning electron microscope image.

[0028] For step S1, specifically, based on a large amount of scanning electron microscope image data of real wafers, the visual characteristics (such as brightness distribution, surface texture, contrast change, etc.) of different material types, process recipes and defect types under SEM imaging are analyzed. Through region segmentation, process attribute analysis and visual feature extraction of real scanning electron microscope images, data correlation between materials, processes and imaging features is formed, for example, the imaging characteristics of aluminum material using chemical vapor deposition process are smooth surface, high brightness, etc. The mapping relationship between defect types (such as cracks, particles, aperture too small) and their defect imaging features in SEM images is analyzed, for example, the crack defects caused by mechanical stress, thermal mismatch or processing residues are represented as linear dark bands or local gray scale steep drop areas in the image, with clear edges and high contrast imaging features. Thus, the first imaging feature mapping table and the second imaging feature mapping table are respectively constructed.

[0029] Step S2: Based on each standard imaging feature of the first imaging feature mapping table and each defect imaging feature of the defect type and defect imaging feature mapping table, corresponding standard imaging scripts and defect imaging scripts are respectively made.

[0030] For step S2, specifically, for each standard imaging feature, a corresponding standard imaging script file is made based on the material surface imaging brightness, surface texture characteristics and contrast information represented by the feature. For example, for a standard imaging feature with high imaging brightness, smooth surface texture and high image contrast, a high-brightness-smooth-high-contrast standard imaging simulation script can be made. The simulation script configures surface lighting parameter instructions (set the reflection characteristics parameters of the structure unit surface under the simulation light source, such as diffuse reflectivity, specular reflectivity, incident angle related brightness adjustment coefficient, etc.), material reflectivity instructions (define the unit area reflectivity or absorbance of the surface, according to the optical properties of the material itself, to simulate the brightness difference when imaging different materials), texture disturbance intensity instructions (set the degree of micro-texture fluctuation, such as simulating the surface micro-roughness by Perlin Noise disturbance or normal disturbance function), thereby simulating the real imaging effect of structure units such as metal interconnection, electrode area, polysilicon gate, etc. in the defect-free state.

[0031] Similarly, for each defect imaging feature, a corresponding defect imaging script file is made based on the defect morphology, size variation and imaging brightness anomaly described by the feature. For example, a crack defect imaging simulation script can be made for a crack defect imaging feature. The simulation script can be configured to include simulation control instructions such as defect topography instructions (set the geometric shape of the crack to be irregular line segments, such as single crack, multiple cracks, bifurcated crack, etc. morphology and length, width within a predetermined range of random disturbance), spatial positioning instructions (determine the generation position of the crack according to the edge of the structure unit, stress concentration area or artificially set area), brightness variation instructions (set the gray value to be reduced in the crack path area to form a linear dark band, simulating the visual effect of reduced electron emission or surface charge discharge), texture disturbance superposition instructions (introduce local surface disturbance along the crack area to simulate secondary texture effects such as surface rupture and material peeling), etc. to simulate the generation of local structure changes conforming to the crack defect imaging feature in the specified area of the wafer three-dimensional model.

[0032] Step S3: Obtain a wafer three-dimensional model containing a plurality of structure units and analyze the associated process attribute information in each structure unit, including material type, process recipe information and mask pattern information.

[0033] For step S3, specifically, the wafer three-dimensional model is usually constructed with structural units as the basic granularity, each of which represents an independent or regularly repeated microstructure entity (such as metal traces, vias, gates, etc.) on the wafer. Each structural unit has its structural geometric information and metadata entries bound thereto defined in the modeling stage, and the metadata is the process attribute information, which specifically includes: material type information, indicating the material type used in the process flow of the structural unit, such as aluminum, tungsten, polysilicon, low dielectric material, etc., which is used to determine the basic brightness and texture characteristics of the structural unit in the scanning electron microscope image; process recipe information, indicating the key processing technology experienced by the structural unit in the formation process, including physical vapor deposition (PVD), chemical vapor deposition (CVD), etching (Etching), ion implantation (Ion Implantation), chemical mechanical polishing (CMP), etc.; the process recipe determines the surface topography, structural integrity, and imaging roughness or reflectivity characteristics after material deposition or etching; mask layout information, indicating the mask definition pattern used by the structural unit in the photolithography process, including pattern contour, line width, alignment deviation, etc., which can be used to determine whether the geometric boundary of the structural unit has potential abnormal characteristics such as edge blur, size undersize or overexposure, affecting its standard imaging performance or defect sensitive area identification.

[0034] Step S4: Based on the process attribute information, the corresponding standard imaging script is determined by querying the first imaging feature mapping table, and the surface characteristics of the structural unit are adjusted based on the standard imaging script to generate a defect-free simulation model.

[0035] For step S4, specifically, the system queries the first imaging feature mapping table according to the process attribute information of the structural unit, including material type, process recipe, and mask pattern features, to determine the standard imaging feature entry corresponding to the attribute combination. The imaging feature entry contains the expected performance of the structural unit in the scanning electron microscope image, such as brightness level, surface smoothness level, boundary contrast level, etc. Once the standard imaging feature is matched, the system will automatically call the corresponding standard imaging script according to the associated script path or identifier specified in the mapping table. The script contains a series of parameter settings and execution instructions for controlling the simulation surface performance. The system adjusts the simulation surface characteristics of the current structural unit according to the parameters set in the called standard imaging script, such as material assignment, lighting attribute setting, surface disturbance superposition, etc., and further generates a defect-free simulation model that conforms to the real imaging effect.

[0036] Step S5: Based on the generation of the defect-free simulation model, according to the preset defect introduction condition, the corresponding defect imaging script is determined by querying the second imaging feature mapping table, and the defect is added on the specified structure unit based on the defect imaging script to generate a simulation model with defects.

[0037] The system defines the type, distribution rule and position range of the defect to be inserted according to the defect introduction condition set by the user or generated by the algorithm. The defect introduction condition can include: defect type: such as crack, particle contamination, aperture size, bridging, etc.; defect occurrence probability: used to control the distribution density of defects in the target structure unit or target area; defect position range: can be specified as the center area, edge area or specific coordinate range of the structure unit, supporting local or random insertion. According to the defect type, the system queries the second imaging feature mapping table to determine the corresponding defect imaging feature item. Based on the matched defect imaging feature, the system calls the corresponding defect imaging script that has been made. The script contains control instructions and parameter settings required to restore this type of defect in the simulation model, such as defect morphology, size, position rule, etc. In the specified spatial position of the target structure unit, the system executes the simulation instructions in the defect script to insert the simulated defect. The defect introduction process can include: morphology structure change: such as generating linear cracking, circular aperture reduction, particle attachment or interrupted structure, etc.; visual property adjustment: such as reducing local brightness, introducing boundary blur, enhancing local contrast or disturbing texture continuity; defect boundary control: ensure that the defect region and the surrounding structure form a significant gray or texture difference, consistent with the real image performance. The system integrates the defect structure into the wafer three-dimensional model, and then constructs a simulation model with real defect distribution and visual features.

[0038] Step S6: Based on the defect-free simulation model and the simulation model with defects, a simulation image of a simulated scanning electron microscope is generated.

[0039] For step S6, specifically, for the aforementioned completed defect-free simulation model and simulation model with defects, image rendering and image conversion techniques can be used to generate simulation images that conform to the visual style of scanning electron microscope images. This process collects model surface responses by simulating lighting conditions and observation angles, and performs image post-processing to output a simulation grayscale image. The final output grayscale image, as a simulated scanning electron microscope imaging result, includes the surface details of the complete structure, material property performance, and visual features after defect introduction (such as dark bands, breakpoints, highlighted particles, etc.).

[0040] It can be found that, compared with the related art, in the scheme provided by the embodiment of the application, by establishing an imaging feature mapping table, automatically analyzing wafer three-dimensional model process attributes, intelligently calling a simulation script to construct a defect-free and defective model, and batch manufacturing a high-simulation scanning electron microscope image through virtual simulation means, the automation and standardization degree of simulation image generation is improved, the imaging details of the wafer structure in the normal and abnormal states are effectively restored, the actual defect image acquisition cost is significantly reduced, the diversity and coverage of the simulation image data set are improved, and then the efficiency and accuracy of defect detection model training and verification are improved.

[0041] Second embodiment

[0042] The second embodiment of the application relates to a simulation image generation method of a simulation scanning electron microscope. The second embodiment is an improvement on the basis of the first embodiment, and the specific improvement is that in the second embodiment of the application, a specific implementation manner for constructing an imaging feature mapping table required for a scanning electron microscope simulation image is provided. That is, step S1 can further include the following steps:

[0043] Step S101: scanning electron microscope image acquisition is performed on a plurality of wafer samples with known material types, process recipes and defect information;

[0044] Step S102: standard imaging features and defect imaging features corresponding to structural units are extracted from the acquired images;

[0045] Step S103: the extracted standard imaging features are bound with the material types and process recipes corresponding thereto to form a first imaging feature mapping table;

[0046] Step S104: the extracted defect imaging features are bound with the defect types corresponding thereto to form a second imaging feature mapping table.

[0047] In the embodiment, wafer batch samples containing multiple structural units are selected, and the material composition and process path thereof can be obtained through process trace records and manufacturing execution system (MES) data. A plurality of groups of scanning electron microscope images (SEM images) in a defect-free state are acquired from wafer samples with known material and process records. For each structural unit of the SEM image, the material type (such as aluminum, tungsten, and polysilicon) and the process recipe (such as PVD deposition, CVD deposition, and dry etching) actually adopted are classified and analyzed, and the imaging features of the structural unit are analyzed, including but not limited to surface smoothness, brightness distribution, texture roughness, contrast characteristics, etc. The material type, the process recipe and the extracted standard imaging features are mapped to form a first imaging feature mapping table, as shown in the following table.

[0048] Table 1 Mapping relationship between material type, process recipe and standard imaging features

[0049] Materials Process Standard imaging features Aluminum Chemical vapor deposition High imaging brightness, smooth surface texture, high contrast Aluminum Anodization Medium imaging brightness, rough surface texture, medium contrast Silicon Chemical vapor deposition Low imaging brightness, rough surface texture, medium contrast Tungsten Chemical vapor deposition High imaging brightness, rough surface texture, medium to high contrast

[0050] Each feature type in the standard imaging features, such as imaging brightness, surface texture, image contrast, etc., is quantitatively valued by quantifiable parameters. For example, imaging brightness is quantitatively described by average gray value, surface texture is quantitatively described by gray variance or texture energy, and image contrast is quantitatively described by contrast factor or local contrast.

[0051] Exemplarily, imaging brightness is quantitatively divided by gray mean value (8-bit image value range 0-255), such as using 5-level classification standard: low (80-110), medium low (110-140), medium (140-180), medium high (180-220) and high (220-255); or using 3-level classification standard: low (80-130), medium (130-200) and high (200-255).

[0052] The quantifiable parameters mentioned above can be extracted from real scanning electron microscope images by image analysis algorithm and normalized to support the uniform comparison of features between different sample image data, and can also be used as input parameters to drive the automatic generation and matching of scripts, such as generating highlight-smooth-high contrast standard imaging simulation scripts. Since each script parameter is derived from data extraction and normalization processing of real scanning electron microscope images, the simulation images can highly restore the actual observation results in brightness, texture, boundary characteristics, etc., significantly enhancing the authenticity and generalization ability of the training images.

[0053] To construct the mapping relationship between defect types and defect imaging features, wafer samples containing multiple typical defect types are selected, including defect instances such as particle contamination, cracks, broken lines, aperture abnormalities, etc. For the defect regions labeled in the scanning electron microscope images, local image features are extracted by image analysis algorithm, and combined with the defect type labels to establish the mapping relationship between defect types and defect imaging features, forming a second imaging feature mapping table, as shown in the following table.

[0054] Table 2 Mapping relationship between defect types and defect imaging features

[0055] Defect types Standard imaging features Cracks Low imaging brightness, elongated dark bands, sharp edges, high contrast Particle contamination High imaging brightness, tight boundaries, small area, local high-light spots Discontinuity Imaging texture interruption, structure gap, gray level drop, edge discontinuity Aperture undersize Pattern area reduction, edge blur, brightness drop, low contrast

[0056] The image characteristic types such as brightness change, edge definition, contour continuity in the defect imaging feature can also be quantified and processed numerically. For example, the average gray value of the crack region is set to be low (such as ≤100), the linearity is ≥0.9, and the local contrast is ≥0.85; the gray peak value of the particle pollution region is ≥220, the area is <30px², and the contour tightness is ≥0.95. These parameters are used as input parameters to drive the automatic generation of the script, such as generating a crack defect imaging simulation script, and can also be used as a script calling condition to support automatic matching of defect simulation scripts in the defect insertion process.

[0057] It can be found that, in the scheme provided by the embodiments of the application, the mapping relationship between the material type, the process recipe and the standard imaging feature, and the mapping relationship between the defect type and the defect imaging feature are established, forming a bidirectional imaging feature mapping system covering the normal structure state and the defect state. The mapping system not only realizes the mapping normalization and standardization of the process attribute of the structure unit to the visual performance, but also supports automatic retrieval and calling of the corresponding standard imaging script or defect imaging script, so as to improve the execution efficiency of the image simulation process.

[0058] Third embodiment

[0059] The third embodiment of the application relates to a simulation image generation method of a scanning electron microscope. The third embodiment is an improvement based on the first embodiment, and the specific improvement is that in the third embodiment of the application, a specific implementation manner of parameterizing, structuring and scripting the imaging feature is provided. That is, step S2 can further include the following steps:

[0060] Step S201: for each standard imaging feature, determine the corresponding surface characteristic simulation parameter, and the surface characteristic simulation parameter includes the light reflectivity, the surface roughness factor, the texture disturbance amplitude and the contour definition factor;

[0061] Step S202: according to the surface characteristic simulation parameter, make the first instruction set for the surface modeling and rendering control of the structure unit, and organize the first instruction set to form a standard imaging script file;

[0062] Step S203: for each defect imaging feature, determine the corresponding defect simulation parameter, and the defect simulation parameter includes the defect morphology type, the defect size range, the positioning rule of the defect in the structure unit and the brightness weight;

[0063] Step S204: according to the defect simulation parameter, make the second instruction set for the modeling and rendering control of the defect region, and organize the second instruction set to form a defect imaging script file.

[0064] Exemplarily, for each standard imaging feature, the system can determine the surface property simulation parameters corresponding to it according to the visual indicators extracted in the image analysis process. These parameters are used to control the visual performance of the structure unit in the three-dimensional simulation image, such as light reflectivity: the electron beam reflection intensity of the corresponding material under a specific process, affecting the imaging brightness; surface roughness factor: reflecting the degree of surface texture fluctuation, controlling the texture smoothness; texture disturbance amplitude: defining the amplitude of micro-scale noise, used to simulate process residues or particle structures; contour clarity factor: adjusting the degree of edge sharpening, determining the clear performance of the structure contour in the image. For example, a certain structure unit matches the standard imaging feature of "high brightness-smooth-high contrast", and the surface property simulation parameters can be defined as: "reflectivity = 0.9, roughness factor = 0.1, disturbance amplitude = low, contour clarity = high".

[0065] The system automatically generates a first instruction set for three-dimensional structure surface modeling and image rendering control according to the above-mentioned surface property simulation parameters, according to the mapping logic of corresponding parameter modeling or rendering behavior, including: surface material configuration instructions (setting reflectivity, glossiness, etc.), light simulation instructions (controlling electron beam incident angle and response coefficient), texture generation instructions (applying disturbance function to synthesize local texture), boundary sharpening instructions (performing image sharpening or gradient enhancement processing) and other instructions. The instruction set is encapsulated into a standard imaging script file in JSON, Python-like, YAML syntax format, etc. and stored. The script path or identifier is bound to the entry in the first imaging feature mapping table, facilitating subsequent calling and use.

[0066] The instruction set describes the simulation engine's behavior in material setting, texture disturbance generation, light model configuration, and edge rendering adjustment in the form of structured commands, ensuring that the visual performance of the structure unit in the simulation image is consistent with the standard imaging feature it maps to.

[0067] For each defect imaging feature, the system extracts corresponding defect simulation parameters according to its visual performance and spatial distribution. These parameters are used to control the geometric morphology and image abnormal features of the defect, such as defect topography type: such as crack (linear opening), particle (circular protrusion), aperture anomaly (shape contraction), etc.; defect size range: indicating the length, width, area, etc. range limit of the defect; defect positioning rule: defining the defect insertion position (such as structure unit edge, center or random); brightness weight: controlling the gray scale offset degree of the defect area, reflecting the brightness / darkness abnormality degree in the imaging. For example, for a crack defect, its crack defect imaging simulation parameters can be defined as: "type = linear crack, length = 30-50px, position = structure edge, brightness drop = 50%".

[0068] Based on the aforementioned defect simulation parameters, the system generates a second set of instructions for local structural modification and image perturbation control, and organizes it into a defect imaging script file. This script includes: defect geometry modeling instructions (for drawing the shape of the defect region), local attribute perturbation instructions (for offsetting or gradient control of grayscale values ​​in a specified region), position constraint instructions (for determining the defect insertion point according to positioning rules), and edge blurring instructions (for natural defect transition). This script will be locally called on the defect-free simulation model, enabling the system to accurately insert defects that conform to realistic representations into the image.

[0069] It is easy to see that the solution provided in this application establishes an integrated implementation method for modeling and rendering control driven by image feature parameters by introducing an automatic generation mechanism for standard imaging scripts and defect imaging scripts. This avoids redundant definitions and manual modeling operations, reduces human intervention, and improves processing efficiency. Furthermore, by decoupling the simulation script from the mapping logic, it supports on-demand expansion of script templates to adapt to different structural or process conditions, improving the system's maintainability and adaptability.

[0070] It should be noted that the third embodiment of this application may also be an improvement based on any one or more of the first to second embodiments.

[0071] Fourth embodiment

[0072] The fourth embodiment of this application relates to a method for generating simulated images using a scanning electron microscope. The fourth embodiment is an improvement upon the first embodiment, specifically in that it provides a mechanism for efficiently extracting process attribute information corresponding to structural units from a three-dimensional wafer model. Specifically, step S3 may further include the following steps:

[0073] Step S301: Read the wafer 3D model and parse the data structure of the wafer 3D model. The data structure is a 3D modeling data system used to describe the mapping relationship between the geometric information of each structural unit in the model and the corresponding metadata entries.

[0074] Step S302: Extract the material type, process formula information and mask graphic information contained in the metadata entry corresponding to each structural unit to obtain the associated process attribute information.

[0075] In this embodiment, the system first loads the input wafer three-dimensional model file, which is composed of multiple structure units, each of which represents a certain geometric microstructure (such as metal lines, contact holes, vias, dielectric layers, etc.) on the wafer. Each structure unit not only contains its geometric spatial attributes (such as position, size, layer number), but also is associated with a set of metadata entries through an index field or an embedded field, which stores process parameter information related to the structure unit in the manufacturing process. The data structure describing the mapping relationship between the geometric information of each structure unit and the corresponding metadata entries can be an object-oriented modeling system or constructed in the form of a hierarchical graph (such as Octree or NetCDF), the core feature of which is to clearly define the retrievable mapping relationship between the structure unit entity and its physical / technological attributes, so that the system can directly extract the rendering control information required from the three-dimensional model structure in the simulation preparation stage.

[0076] The system analyzes the metadata entries associated with each structure unit, extracts the material type, process recipe information, and mask pattern information contained therein, and forms a process attribute description vector for the structure unit. For example, the process attributes of a certain structure unit after analysis are: "material type = aluminum, process recipe = PVD deposition and dry etching, mask pattern = line width 0.12 μm, rectangular opening." The system compares the extracted process attribute information with the first imaging feature mapping table to determine the standard imaging script required to be called by the structure unit, and provides a process-level decision basis for whether to add defects, what kind of defects to add, and where to add defects.

[0077] In this embodiment, the wafer three-dimensional model is constructed, including the following steps:

[0078] Step A1: Construct an initial three-dimensional model of the target wafer, which is a thin cylindrical structure used to represent the base form of the wafer;

[0079] Step A2: According to the preset process flow, read each process step in turn, and determine the adjustment mode and adjustment area of the wafer three-dimensional model based on the process step;

[0080] Step A3: Determine the adjustment amplitude of the wafer three-dimensional model in combination with the process recipe parameters corresponding to the current process step and the historical measurement data;

[0081] Step A4: Adjust the structure of the wafer three-dimensional model based on the adjustment mode, adjustment area, and adjustment amplitude;

[0082] The process of steps A2 to A4 is executed in a loop until the process flow is completed, and the three-dimensional model structure of the target wafer is obtained.

[0083] Specifically, the system supports voxel modeling or mesh modeling by loading a three-dimensional modeling engine module to construct an editable geometric model structure. The size (diameter and thickness) of the initial three-dimensional model can be set according to the specific wafer specification (such as 8 inches, 12 inches, etc.) to represent the substrate of the bare wafer. The initial three-dimensional model does not contain any pattern information and represents the silicon wafer entity in the unprocessed state, which is the starting point of the entire virtual modeling process.

[0084] The modeling engine stretches a two-dimensional cross-section with a circular bottom surface along the Z-axis direction in a three-dimensional coordinate system to generate a cylinder with a specified height. The spatial coordinate system of the three-dimensional model is established with the center point of the cylinder bottom surface as the origin. The X-Y axis plane is the wafer surface, and the Z-axis is the thickness direction. By constructing the initial three-dimensional model, the system can provide a unified geometric reference framework for accepting subsequent material stacking, etching, pattern division, and attribute embedding structure adjustment operations in each process step to ensure that all processing steps are superimposed and transformed in a unified coordinate system.

[0085] In step A2, the process steps can include but are not limited to deposition, oxidation, photolithography, etching, doping, cleaning, and planarization, etc. For different types of process steps, the system uses the following processing strategies to determine the adjustment method and adjustment area.

[0086] When the process step is an oxidation step, if a silicon dioxide layer is generated directly on the exposed silicon surface, the system sets the adjustment method to uniformly cover an oxidation structure on the entire substrate surface of the wafer three-dimensional model.

[0087] When the process step is a deposition step, the system sets the adjustment method to add material structures in the wafer three-dimensional model, and the adjustment area is determined according to the mask image or the photoresist protection area of the current step, usually referring to the surface area not covered by the photoresist.

[0088] When the process step is a photolithography step, the system sets the adjustment method to remove or retain the photoresist structure, and the adjustment area is determined according to the pattern information of the mask image and the type of photoresist used (positive or negative), to simulate the pattern structure of the developed photoresist.

[0089] When the process step is a doping (ion implantation) step, the system sets the adjustment method to introduce doping attribute information such as ion species, implantation dose, and spatial distribution in the adjustment area, which is the area not covered by the photoresist to represent the implantation window.

[0090] When the process step is an etching step, the system sets the adjustment method to remove material structures, and the adjustment area is determined according to the exposed area defined by the pattern in the loaded mask image. The system will reduce the material volume in this area by a specified depth or proportion.

[0091] When the process step is a cleaning step, the system sets the adjustment mode as removing the residual photoresist structure, and the adjustment region is determined according to the photoresist pattern region formed in the previous photoetch step, indicating the region where the residual photoresist may exist, and the photoresist layer is thinned in the model.

[0092] When the process step is a planarization step (such as CMP), the system sets the adjustment mode as unifying the height of the top surface of the wafer three-dimensional model, and the adjustment region is the top region of the current model or the local region where the thickness fluctuates. The system performs a geometric reconstruction operation in this region to achieve the height normalization of the structure.

[0093] In step A3, the process recipe parameters of the current process step are read, including but not limited to target thickness, etching time, deposition rate, doping attribute information, etc. The parameter values are the standard target values set by the process design stage or the production system. These parameters represent the physical structure changes that should be achieved by the process step under ideal conditions, and are the basis for calculating the adjustment amplitude.

[0094] At the same time, the system obtains the historical measurement data corresponding to the process step from the database or the MES system, i.e. the actual measurement results for the process step during the processing of multiple batches of wafers in the past, including but not limited to actual deposition thickness, etching depth, photoresist residual thickness, doping layer distribution, etc. Such data can reflect the real execution error and stability trend in the process, providing support for subsequent parameter correction.

[0095] Based on the process recipe parameters and historical measurement data, the system constructs a fitting model (such as linear regression, polynomial regression or correction function) to represent the numerical mapping relationship between the theoretical target value and the actual processing result. Through the fitting model, the target value in the process recipe parameters is corrected to obtain an adjustment amplitude value that is closer to the actual situation. This adjustment amplitude value is the spatial scale basis for subsequent structure addition, removal or attribute embedding in the wafer three-dimensional model.

[0096] Taking the oxidation process step as an example, to predict the growth rate and thickness of silicon dioxide, an oxidation thickness regression model can be constructed according to the historical measurement data (such as furnace temperature, oxygen flow, deposition time, etc.):

[0097]

[0098] wherein, is the thickness of the oxide layer, is the furnace temperature, is the oxygen flow, is the oxidation time, is the model error term.

[0099] The historical metrology data is real data obtained by measurement, which is used as the target value for training the regression model. The model can also search for the best parameter combination through an optimization algorithm to make the model output as close as possible to the set thickness value. The final output film thickness value will be used as the amplitude value for adjusting the three-dimensional model structure in step A4, which is used to generate a silicon dioxide structure layer in the model that matches the actual oxidation process, so that the model is more consistent with the actual process error.

[0100] Taking the photolithography process step as an example, the system can collect historical metrology data closely related to the pattern development result, including exposure dose, development time, photoresist thickness, and other related parameters. Based on the above multi-dimensional process data, the system can construct a line width prediction model, preferably using advanced modeling techniques such as response surface method (RSM) to express the response relationship between line width and key process parameters. The system can predict the pattern line width value in the model through this modeling method. The final pattern structure is more consistent with the actual measurement result, which improves the consistency of photolithography pattern transfer and the accuracy of three-dimensional model restoration.

[0101] Step A4 specifically includes:

[0102] When the adjustment method is material addition, a three-dimensional structure corresponding to the target material attribute is constructed above the adjustment area according to the adjustment amplitude;

[0103] When the adjustment method is material removal, the corresponding three-dimensional structure is removed from the wafer three-dimensional model with the adjustment amplitude as the depth in the adjustment area;

[0104] When the adjustment method is to remove or retain the photoresist structure, a thin cylindrical structure is first overlaid on the surface of the current wafer three-dimensional model to simulate the initial thickness of the photoresist, and then the photoresist structure is removed in the adjustment area according to the adjustment amplitude to retain the photoresist structure in the non-adjustment area;

[0105] When the adjustment method is to introduce doping attribute information, attribute markers representing ion species, concentration, or distribution are added to the wafer three-dimensional model in the adjustment area;

[0106] When the adjustment method is to unify the height of the top surface of the wafer three-dimensional model, geometric reconstruction is performed on the local structure with height difference in the adjustment area, including cutting off the structure volume higher than the target height or filling the area lower than the target height, to realize the normalization of the surface height of the wafer three-dimensional model.

[0107] The process of steps A2 to A4 is executed in a loop until the process flow is completed, and a three-dimensional model structure of the target wafer is obtained. The generated three-dimensional wafer model contains the structure layer, pattern topography and doping attribute formed by each process step in the entire process flow, and expresses the structure evolution and attribute change of the wafer at each process stage in a three-dimensional visual manner, so as to simulate the complete wafer manufacturing process. The model not only has a three-dimensional geometric structure, but also integrates electrical, process and other attribute information, and can dynamically display the structure evolution process of the wafer at each process stage through three-dimensional visualization, realizing virtual simulation and reconstruction of the entire wafer manufacturing process.

[0108] It can be found that in the scheme provided by the embodiments of the application, a data structure analysis mechanism for efficiently extracting process attribute information of a structure unit is provided, and semantic-level understanding of the structure unit in the three-dimensional wafer model is realized. The model not only has spatial configuration information, but also has process constraint attributes, thereby supporting the mechanism of process attribute driven script matching in the subsequent image simulation process, and significantly improving the intelligent level of the system in image consistency control and script calling automation.

[0109] It should be noted that the fourth embodiment of the application can also be an improvement on the basis of any one or more of the first to third embodiments.

[0110] Fifth Embodiment

[0111] The fifth embodiment of the application relates to a simulation image generation method of a scanning electron microscope. The fifth embodiment is an improvement on the basis of the first embodiment, and the specific improvement is that in the fifth embodiment of the application, a specific implementation manner for constructing a defect-free simulation model is provided. That is, step S4 can further include the following steps:

[0112] Step S401: based on the material type of the structure unit and the process recipe information, querying a first imaging feature mapping table and obtaining the corresponding standard imaging feature;

[0113] Step S402: according to the standard imaging feature, calling a corresponding standard imaging script;

[0114] Step S403: extracting surface property simulation parameters in the standard imaging script, and adjusting the surface property of the structure unit;

[0115] Step S404: by completing the surface property adjustment, a defect-free simulation model conforming to the description of the standard imaging feature is generated.

[0116] In this embodiment, the system first reads the process attribute information associated with the target structure unit, for example: the material type is aluminum, and the process recipe is CVD deposition. The system takes this information as a query key to query the first imaging feature mapping table, finds the standard imaging feature item corresponding to the process attribute combination in the mapping table, such as: “high imaging brightness, smooth surface texture, high contrast”. According to the retrieved standard imaging feature item, the standard imaging script file matched with the feature is called, and the script internally encapsulates a plurality of rendering instructions and modeling parameters. The system parses the called standard imaging script file, extracts the surface property simulation parameters therein, and according to these parameters, writes or adjusts the material properties, lighting model, surface disturbance function, etc. in the three-dimensional surface model of the structure unit, to complete the reconstruction of the visual properties.

[0117] It can be found that in the scheme provided by the embodiments of the application, by performing the above surface property adjustment operation, the system completes the conversion of the structure unit from the initial wafer three-dimensional model to the defect-free simulation state conforming to the standard imaging feature performance. At this time, the surface material, texture fluctuation, and edge processing of the structure unit all conform to the image performance of the corresponding process product in the real wafer under the scanning electron microscope.

[0118] It should be noted that the fifth embodiment of the application can also be an improvement based on any one or more of the first to fourth embodiments.

[0119] Sixth Embodiment

[0120] The sixth embodiment of the application relates to a simulation image generation method of a scanning electron microscope. The sixth embodiment is an improvement based on the first embodiment, and the specific improvement is that in the sixth embodiment of the application, a specific implementation manner of structure unit level defect generation control is provided. That is, step S5 can further include the following steps:

[0121] Step S501: determining a target structure unit for introducing a defect based on a preset defect introduction condition, and the defect introduction condition includes a defect type and a spatial position where the defect appears;

[0122] Step S502: querying a second imaging feature mapping table according to the defect type to match a corresponding defect imaging feature;

[0123] Step S503: calling a corresponding defect imaging script according to the defect imaging feature;

[0124] Step S504: extracting defect simulation parameters in the defect imaging script to perform defect simulation processing on the target structure unit to form a defect simulation model conforming to the expected defect characteristics.

[0125] In this embodiment, the system determines which structural units should introduce defects according to the defect introduction conditions set by the user or generated by the algorithm model. The defect introduction conditions include defect types such as cracks, particle contamination, aperture abnormalities, and bridging; and defect spatial positions such as the center, edge, or contact point of a structural unit, or a given three-dimensional coordinate range. Optionally, it can also include defect density, such as inserting a certain number of particles in each functional area or controlling the area distribution probability.

[0126] The defect introduction conditions can be automatically generated by one or more of the following strategies, including:

[0127] A statistical distribution model established based on defect labeling data of real SEM images is used to control the frequency, spatial distribution density, and size range of defects. For example, in the metal wiring area, a probability distribution map is established based on the high-probability area of crack defects in historical samples (such as corners and connection points) to control the spatial position and type ratio of defects.

[0128] Defect high-risk position prediction based on process simulation results (such as stress field distribution or etching deviation simulation);

[0129] A rule sampling algorithm based on defect sample enhancement strategy is used to synthesize a diversified combination of defects under controlled variables to improve data coverage.

[0130] According to the defect type, the second imaging feature mapping table is queried to match the corresponding defect imaging feature. According to the matched defect imaging feature, the corresponding defect imaging script file is loaded and called, which encapsulates the geometric modeling and visual disturbance instructions required to generate the corresponding defect area in the three-dimensional structure, such as the crack script containing line crack shape generation, gray scale adjustment, and edge sharpening control commands. Each script can be called by type / number identification.

[0131] The system parses the script file content, extracts the defect simulation parameters, and performs defect generation operations in the target structural unit. The defect simulation parameters can include: defect topography parameters: geometric shape (linear, circular), length, width, position offset; defect visual parameters: brightness disturbance value, contrast enhancement coefficient, boundary blur weight; insertion logic control: whether to overlap existing structures, whether to connect across layers, whether to deform randomly, etc.

[0132] Figures 2a to 2d The process control of defect insertion at the structural unit level in this embodiment is illustrated.

[0133] As shown in Figure 2a The initial stage of the simulation process constructs a three-dimensional model of a structural unit in a defect-free state, in which the metal layer is in a "cross" structure and there is no process abnormality or visual disturbance.

[0134] exist Figure 2b In step S501, based on the defect introduction conditions set, for example, the defect type is "crack" and the defect location is the sidewall of the metal layer, the system identifies the corresponding target structural unit and marks the spatial area where the defect needs to be inserted (as shown in the red box in the figure).

[0135] According to step S502, defect imaging features matching the crack defect type are retrieved from the second imaging feature mapping table, such as: local brightness decrease, enhanced surface texture disturbance, blurred edges, etc., and the corresponding defect imaging script is loaded accordingly. The defect script is executed to perform defect simulation processing on the target structural unit, and surface roughness disturbances are superimposed on the metal layer surface, such as... Figure 2c As shown, the fracture surface exhibits a linear fracture, simulating an origin from uneven thermal stress during deposition and etching. Cracks can cause localized signal interference and path discontinuities, thereby reducing circuit stability.

[0136] In addition to crack defects, dust interference defects can be further added, such as... Figure 2d As shown, by simulating the spatial distribution of multiple locally isolated particle shapes and superimposing them onto the metal surface texture, the visual anomalous features of local areas are enhanced. This simulates the situation of airborne particle deposition when clean environment control is insufficient. Such dust may adhere to interconnect nodes or overlapping areas, forming additional interfacial impedance, thereby causing conduction abnormalities or localized heating problems. The dust particles can be set as irregular spherical or ellipsoidal structures, with sizes ranging from several nanometers to submicrometers. The distribution position can be randomly sampled according to Gaussian or Poisson distributions, and the brightness, edge transition, and reflectivity properties of the particles can also be set as needed.

[0137] The insertion of this type of particle morphology, superimposed on an existing rough and disturbed surface, can effectively simulate dust adhesion caused by environmental pollution or residual deposits in actual processes. This creates a complex defect region with multiple visual features in the simulated image, including combinations of brightness fluctuations, texture disturbances, morphological protrusions, and blurred boundaries. This significantly improves the realism and complexity of the simulated image, making it suitable for constructing more challenging training or testing samples.

[0138] Further, for the contact hole structure unit, a side wall roughness type defect is further included, and roughness disturbance (such as 50 nm scale surface fluctuation) is introduced at the side edge of the hole wall, which is used to reflect the uneven phenomenon of the hole wall caused by the fluctuation of the etching process parameters. Such a defect may affect the uniformity of metal filling and cause local impedance deviation. Alternatively, a hole diameter deviation defect can be added, specifically, a geometric change area with a hole diameter reduction of more than 5 nm is generated at the center of the contact hole, which is used to simulate the size control error in the photolithography or etching process. The insufficient hole diameter will limit the through-hole conduction capability, and in severe cases, it may even cause open circuit failure of the device.

[0139] It can be found that, in the scheme provided by the embodiments of the present application, the target structure unit can be automatically identified and matched with the corresponding defect imaging script according to the preset defect type and spatial position, avoiding the cumbersome operation of manual modeling and position setting in the traditional method, and realizing the automation and structure perception control of the defect insertion process. At the same time, the defect simulation script supports flexible configuration of the topographic parameters and visual disturbance parameters of the defect, such as size, number, distribution position, brightness change and edge blur, which can cover multiple typical defect types such as cracks, particles and abnormal hole diameters, and even allows multiple defects to be superimposed, thereby enhancing the complexity and expressiveness of the defect features in the image.

[0140] It should be noted that the sixth embodiment of the present application can also be an improvement on the basis of any one or more of the first to fifth embodiments.

[0141] Seventh Embodiment

[0142] The seventh embodiment of the present application relates to a simulation image generation method of a scanning electron microscope. The seventh embodiment is an improvement on the basis of the first embodiment, and the specific improvement is that in the seventh embodiment of the present application, a specific implementation manner of generating a gray scale image conforming to the visual style of a scanning electron microscope is provided. That is, step S6 can further include the following steps:

[0143] Step S601: Image acquisition of the defect-free simulation model and the simulation model with defects is performed through a plurality of set viewing angles, including top view, side view and inclined view;

[0144] Step S602: During the image acquisition process, color image data is generated according to the model surface normal vector, material reflectivity and simulation light source setting;

[0145] Step S603: The color image data is converted into a gray scale image according to the predefined brightness weight in the standard imaging feature and the defect imaging feature, so as to serve as a simulation image of a simulated scanning electron microscope, and the simulation image is consistent with the real scanning electron microscope image in visual style.

[0146] In this embodiment, after the construction of the defect-free simulation model and the defect simulation model is completed, multiple imaging angles are set for image acquisition. The types of viewing angles include: top view (vertical incidence): used to capture the overall profile, hierarchical structure and distribution density; side view (along the side of the structure): used to observe edge defects, profile discontinuity, etc.; oblique view (such as 45°): used to enhance the three-dimensional effect, reveal the relationship between surface disturbance and multi-layer structure. Under each viewing angle, the system simulates the electron beam scanning process with virtual SEM imaging logic, and collects the rendering results of the surface of the structure unit.

[0147] During the image acquisition process, the system combines the surface attributes of the three-dimensional model and the lighting response model to perform the following simulation on the surface of each structure unit: calculate the incident response according to the surface normal vector and the lighting angle; use material reflectivity, texture disturbance function and other parameters to form an impact on color intensity. Map the above calculation results to three-channel (R, G, B) color image data, where different channels represent simulated local reflection characteristics, material contrast, texture depth, etc. The resulting color image is not a real RGB image, but a visual response intermediate representation for subsequent grayscale synthesis.

[0148] The system performs uniform grayscale conversion processing on the generated color image data according to the brightness weight predefined in the standard imaging features and the defect imaging features (brightness performance feature quantization value of the standard imaging features, brightness weight adjustment parameter of the defect imaging features). Conversion methods include: weighted brightness method, custom mapping function (dynamically adjust the mapping ratio of R, G, B according to the material type of the structure unit), local enhancement method (perform brightness compression or contrast enhancement on high-contrast areas to enhance imaging clarity). The grayscale image obtained after conversion is the final simulation image, which is as close as possible to the visual style of the real scanning electron microscope image in terms of image texture, brightness distribution, edge clarity, etc., and can be directly used for visual algorithm training and simulation evaluation.

[0149] Figures 3a to 3c The simulation image effect diagram of the scanning electron microscope based on multi-angle acquisition in this embodiment is shown. Among them, Figure 3a The grayscale image under the side view is shown, and longitudinal dark bands appear in the vertical side wall area of the structure unit, showing linear visual disturbance introduced by crack defects, local brightness reduction and sharp edge features, highly restoring the typical imaging features of defects such as cracks in SEM images. Figure 3b The surface area in the middle shows uniform grayscale texture disturbance, simulating the micro-roughness features caused by material properties and deposition process. Figure 3c Based on Figure 3b Further introduce the particle contamination defect, multiple high-light spots appear in the image, the edge protrusion is obvious, and the local texture is disturbed, reflecting the local enhancement effect of particle adhesion on image grayscale and contrast.

[0150] It can be found that, in the scheme provided by the embodiments of the application, the image acquisition of the defect-free simulation model and the simulation model with defects is performed through setting multiple typical viewing angles, so as to comprehensively cover the geometric morphology and illumination response characteristics of the structure unit under different observation angles, make the generated image more close to the observation effect in the multi-angle imaging process of the real scanning electron microscope (SEM), accurately restore the imaging texture and visual contrast under different process combinations, improve the realism of the image in terms of illumination consistency and texture restoration, and significantly improve the performance of the simulation image in terms of structure readability and analysis adaptability. The generated gray-scale image not only has high restoration degree and engineering credibility, but also can be widely applied to various intelligent manufacturing and quality analysis scenes such as defect detection model training, image algorithm performance testing, and process simulation verification.

[0151] It should be noted that the seventh embodiment of the application can also be an improvement on the basis of any one or more of the first to sixth embodiments.

[0152] The step division of the above methods is only for the purpose of clear description, and can be combined into one step or some steps can be split and decomposed into multiple steps in implementation, as long as the same logical relationship is included, all within the protection scope of the application; adding insignificant modifications or introducing insignificant designs in the algorithm or process, but not changing the core design of the algorithm and process, are within the protection scope of the application.

[0153] In addition, some embodiments of the application also provide an electronic device. The electronic device can be various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and the like. The electronic device can also be various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices.

[0154] The electronic device includes one or more processors, and a memory storing computer program instructions which, when executed, cause the processor to perform a simulation image generation method of a simulation scanning electron microscope as provided by any one or more of the above embodiments. Figure 4An exemplary configuration of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, and an interface for connecting the components, including a high-speed interface and a low-speed interface. The components are interconnected through different buses, and can be mounted on a common main board or otherwise installed as needed. The processor can process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display a GUI on an external input / output device such as a display device coupled to the interface. In some other embodiments, a plurality of processors and / or buses can be used with a plurality of memories and a plurality of memory, if necessary. Also, a plurality of electronic devices can be connected, each device providing part of the necessary operations. Among them, the components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present application described and / or claimed herein.

[0155] The electronic device can further include an input device 1103 and an output device 1104. The processor 1101, the memory 1102, the input device 1103, and the output device 1104 can be connected through a bus or otherwise, Figure 4 The connection through the bus is taken as an example in the middle.

[0156] The input device 1103 can receive input digital or character information, and generate key signal input related to user settings and function control of the electronic device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 1104 can include a display device, an auxiliary lighting device (e.g., an LED), and a tactile feedback device (e.g., a vibration motor), etc. The display device can include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some embodiments, the display device can be a touch screen.

[0157] To provide interaction with the user, the electronic device can be a computer. The computer has a display device (e.g., a cathode ray tube or an LCD monitor) for displaying information to the user, and a keyboard and a pointing device (e.g., a mouse) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).

[0158] In the embodiments of the present application, the computer readable medium stores computer programs / instructions, and the computer programs / instructions are executed by the processor to implement the simulation image generation method of the simulation scanning electron microscope provided by any one or more of the above embodiments. The computer readable medium can be included in the electronic device described in the above embodiments, or can exist separately and not be assembled into the device. The computer readable medium carries one or more computer readable instructions.

[0159] The memory 1102 can be used to store non-transitory software programs, non-transitory computer executable programs and modules as a kind of non-transitory computer readable storage medium. The processor 1101 executes various functions and data processing of the server by running the non-transitory software programs, instructions and modules stored in the memory 1102, so as to implement the program instructions / modules corresponding to the method provided by any one or more of the above embodiments in the embodiments of the present application.

[0160] The memory 1102 can include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required by a function; the data storage area can store data created according to the use of the electronic device, etc. In addition, the memory 1102 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory 1102 can optionally include a memory remotely arranged with respect to the processor 1101, and these remote memories can be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0161] It should be noted that the computer readable medium described in the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the above two. The computer readable medium may, for example, but is not limited to: an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or instrument, or any combination of the above. More specific examples of computer readable storage media can include but are not limited to: an electrical connection with one or more conductive wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable medium can be any tangible medium containing or storing programs, which can be used by or in combination with an instruction execution system, device or instrument.

[0162] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory or other memory technologies, read-only optical discs, digital versatile discs or other optical storage, magnetic cassette tapes, magnetic tape discs storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device.

[0163] Computer program code for carrying out operations of the present application can be written in one or more programming languages or combinations of languages including object-oriented, such as Java, Smalltalk, C++, conventional procedural programming languages, such as the C programming language or similar programming languages. Program code can be executed entirely on a user computer, partially on a user computer, as a standalone software package, partially on a user computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network or a wide area network, or can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0164] In the above embodiments, all or part can be implemented by software, hardware, firmware or any combination thereof. For example, a dedicated integrated circuit, a general-purpose computer or any other similar hardware device can be used. In some embodiments, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drive or soft disc and similar devices. In addition, some steps or functions of the present application can be implemented by hardware, for example, as a circuit cooperating with the processor to perform each step or function.

[0165] The computer program product provided by the embodiments of the present application includes one or more computer programs / instructions, which, when executed by a processor, generate all or part of the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be magnetic media (such as floppy disk, hard disk, magnetic tape), optical media (such as DVD), or semiconductor media (such as solid state disk), etc.

[0166] The flowcharts or block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that shown in the figures. For example, two blocks noted in succession can actually be executed substantially concurrently, or they can sometimes be executed in reverse order, depending on the functionality involved. It should also be noted that each block in the block diagrams and / or flowcharts, as well as combinations of blocks in the block diagrams and / or flowcharts, can be implemented by dedicated hardware-based systems that perform the specified functions or operations, or they can be implemented by a combination of dedicated hardware and computer instructions.

[0167] The scope of the present application is defined by the appended claims rather than the description preceding it, so all changes that come within the meaning and range of equivalency of the claims are to be embraced within the scope of the present application. No reference signs in the claims should be considered as limiting the scope of the claims to the features identified by the reference signs. Furthermore, the word "comprising" does not exclude other elements or steps, and the singular "a" or "an" does not exclude a plurality. The plurality of units, or devices, or means can also be implemented by one single unit, or device, or means having multiple functions. The word "first", "second", etc. can not indicate any order, quantity, or importance, but to distinguish between different components. The terms "comprises", "comprising", "includes", "including", or "contains", "containing" should be construed to be open-ended, allowing for the possibility that other components or steps are added.

[0168] The above merely provides specific examples of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims, and the above examples should be regarded as exemplary and non-limiting.

Claims

1. A method of generating a simulated image of a scanning electron microscope, characterized by, The method comprises the following steps: Based on the real wafer scanning electron microscope image data, a first imaging feature mapping table and a second imaging feature mapping table are respectively constructed, the first imaging feature mapping table is used to establish the mapping relationship between different material types and process recipes in their defect-free state and corresponding standard imaging features, and the second imaging feature mapping table is used to establish the mapping relationship between defect types and corresponding defect imaging features, which are used to describe the imaging performance of defects in the scanning electron microscope image; Based on each standard imaging feature of the first imaging feature mapping table and each defect imaging feature of the second imaging feature mapping table, corresponding standard imaging scripts and defect imaging scripts are respectively made; Obtain a three-dimensional model of a wafer containing a plurality of structure units and analyze the associated process attribute information in each structure unit, the process attribute information including material type, process recipe information and mask pattern information; Based on the process attribute information, the corresponding standard imaging script is determined by querying the first imaging feature mapping table, and the surface characteristics of the structure unit are adjusted based on the standard imaging script to generate a defect-free simulation model; Based on the generation of the defect-free simulation model, according to the preset defect introduction condition, the corresponding defect imaging script is determined by querying the second imaging feature mapping table, and defects are added on the specified structure unit based on the defect imaging script to generate a simulation model with defects; Based on the defect-free simulation model and the simulation model with defects, a simulation image of a simulated scanning electron microscope is generated.

2. The simulated image generation method according to claim 1, characterized by, The steps of constructing the first imaging feature mapping table and the second imaging feature mapping table based on the real wafer scanning electron microscope image data comprise: Collect scanning electron microscope images from a plurality of wafer samples with known material type, process recipe and defect information; Extract standard imaging features and defect imaging features corresponding to structure units from the collected images; Bind the extracted standard imaging features with their corresponding material type and process recipe to form the first imaging feature mapping table; Bind the extracted defect imaging features with their corresponding defect type to form the second imaging feature mapping table.

3. The simulated image generation method of claim 1, wherein The steps of making corresponding standard imaging scripts and defect imaging scripts based on each standard imaging feature of the first imaging feature mapping table and each defect imaging feature of the defect type and defect imaging feature mapping table comprise: For each standard imaging feature, determine the corresponding surface characteristic simulation parameters, including light reflectivity, surface roughness factor, texture disturbance amplitude and contour clarity factor; According to the surface characteristic simulation parameters, make the first instruction set of surface modeling and rendering control of the structure unit, and organize the first instruction set to form a standard imaging script file; For each defect imaging feature, determine the corresponding defect simulation parameters, including defect topography type, defect size range, defect positioning rule in the structure unit and brightness weight; According to the defect simulation parameters, a second instruction set for defect area modeling and rendering control is generated, and the second instruction set is organized to form a defect imaging script file.

4. The simulated image generation method of claim 1, wherein The step of obtaining a wafer three-dimensional model containing a plurality of structure units and analyzing associated process attribute information in each structure unit includes: reading a wafer three-dimensional model and analyzing the data structure of the wafer three-dimensional model, the data structure being a three-dimensional modeling data system for describing the mapping relationship between the geometric information of each structure unit in the model and the corresponding metadata entry; extracting the material type, process recipe information and mask pattern information contained in the corresponding metadata entry of each structure unit to obtain the associated process attribute information.

5. The simulated image generation method of claim 1, wherein The step of generating a defect-free simulation model based on the process attribute information by querying the first imaging feature mapping table to determine the corresponding standard imaging script and adjusting the surface characteristics of the structure unit based on the standard imaging script includes: querying the first imaging feature mapping table based on the material type and process recipe information of the structure unit to obtain the corresponding standard imaging feature; according to the standard imaging feature, calling the corresponding standard imaging script; extracting the surface characteristic simulation parameters in the standard imaging script to adjust the surface characteristics of the structure unit; by completing the surface characteristic adjustment, a defect-free simulation model conforming to the description of the standard imaging feature is generated.

6. The simulated image generation method of claim 1, wherein The step of generating a defect simulation model based on the defect introduction condition by querying the second imaging feature mapping table to determine the corresponding defect imaging script and adding defects on the specified structure unit based on the defect imaging script includes: based on the preset defect introduction condition, determining the target structure unit to which the defect is introduced, the defect introduction condition including the defect type and the spatial position of the defect; according to the defect type, querying the second imaging feature mapping table to match the corresponding defect imaging feature; according to the defect imaging feature, calling the corresponding defect imaging script; extracting the defect simulation parameters in the defect imaging script to perform defect simulation processing on the target structure unit to form a defect simulation model conforming to the expected defect characteristics.

7. The simulated image generation method of claim 1, wherein The step of generating a simulation image of a simulated scanning electron microscope based on the defect-free simulation model and the defect simulation model includes: image acquisition of the defect-free simulation model and the defect simulation model is performed from a plurality of preset perspectives, including top view, side view and inclined view; during the image acquisition process, color image data is generated according to the model surface normal vector, material reflectivity and simulation light source setting; according to the predefined brightness weight in the standard imaging feature and the defect imaging feature, the color image data is converted into a grayscale image as a simulation image of a simulated scanning electron microscope, and the simulation image is consistent with the real scanning electron microscope image in visual style.

8. An electronic device, comprising: The electronic device includes: One or more processors; and memory storing computer program instructions that, when executed, cause the processors to perform the method of claim 1-7.

9. A computer readable storage medium having stored thereon a computer program and / or instructions, characterized in that, The computer program and / or instructions, when executed by a processor, implement the method of claim 1-7.

10. A computer program product comprising computer programs and / or instructions, characterized in that, The computer program and / or instructions, when executed by a processor, implement the method of claim 1-7.

Images