CN114049342A - Denoising model generation method, system, device and medium - Google Patents

Denoising model generation method, system, device and medium Download PDF

Info

Publication number
CN114049342A
CN114049342A CN202111392331.0A CN202111392331A CN114049342A CN 114049342 A CN114049342 A CN 114049342A CN 202111392331 A CN202111392331 A CN 202111392331A CN 114049342 A CN114049342 A CN 114049342A
Authority
CN
China
Prior art keywords
sem image
sem
characteristic information
neural network
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111392331.0A
Other languages
Chinese (zh)
Inventor
燕燕
时雪龙
周涛
许博闻
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai IC R&D Center Co Ltd
Shanghai IC Equipment Material Industry Innovation Center Co Ltd
Original Assignee
Shanghai IC R&D Center Co Ltd
Shanghai IC Equipment Material Industry Innovation Center Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai IC R&D Center Co Ltd, Shanghai IC Equipment Material Industry Innovation Center Co Ltd filed Critical Shanghai IC R&D Center Co Ltd
Priority to CN202111392331.0A priority Critical patent/CN114049342A/en
Publication of CN114049342A publication Critical patent/CN114049342A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30148—Semiconductor; IC; Wafer

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)

Abstract

The invention provides a method, a system, equipment and a medium for generating a denoising model, wherein the method comprises the following steps: acquiring an SEM image set after wafer photoetching; superposing and averaging the SEM image pixel values with the same coordinate in the SEM image set to obtain a target image set; performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information; acquiring target image characteristic information of the target image set; and training a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model. According to the method, the characteristic information of the target image in the target image set is obtained, the SEM image characteristic information and the target image characteristic information are used for training the convolutional neural network model to obtain the denoising model, the denoising model is adopted to denoise the SEM image, and the denoising reliability and efficiency are improved.

Description

Denoising model generation method, system, device and medium
Technical Field
The invention relates to the technical field of semiconductor manufacturing processes, in particular to a denoising model generation method, a denoising model generation system, denoising model generation equipment and denoising model generation media.
Background
Under the advanced lithography process node, random noise generated by the dispersion of polymer molecules and the distribution of a photoacid generator can affect the accuracy of an Optical Proximity Correction (OPC) model. In order to build an accurate OPC model, a selected pattern or critical dimension value is usually measured many times, and the influence of random noise is reduced by averaging, but the measurement and calculation of this method is complicated, the subjectivity is strong, the efficiency is low, and the reliability is insufficient.
Disclosure of Invention
The invention aims to provide a method, a system, equipment and a medium for generating a denoising model, which solve the problems of low efficiency and insufficient reliability of acquiring a denoising image, realize the rapid acquisition of the denoising image and increase the reliability of the denoising image.
In order to achieve the above object, in a first aspect, the present invention provides a method for generating a denoising model, including:
acquiring an SEM image set after wafer photoetching;
superposing and averaging the SEM image pixel values with the same coordinate in the SEM image set to obtain a target image set; performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information; acquiring target image characteristic information of the target image set; and training a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model.
The beneficial effects are that: the method comprises the steps of obtaining an SEM image set after wafer photoetching, superposing SEM image pixel values of the same coordinates in the SEM image set, then averaging to obtain a target image set, utilizing a convolutional neural network model to carry out feature extraction on SEM images in the SEM image set to obtain SEM image feature information, obtaining a denoising model by obtaining the target image feature information in the target image set and utilizing the SEM image feature information and the target image feature information to train the convolutional neural network model, and achieving denoising on the SEM images by adopting the denoising model, so that the denoising reliability and efficiency are improved.
Optionally, the set of SEM images includes the SEM images at N identical coordinates, where N is greater than or equal to 2. The beneficial effects are that: by acquiring N SEM images under the same coordinate, wherein N is more than or equal to 2, various random noises generated by the images under the coordinate are ensured to be acquired as much as possible, and the reliable target image set is obtained conveniently by means of superposition and averaging.
Optionally, performing feature extraction on the SEM image in the SEM image set by using a convolutional neural network model to obtain SEM image feature information, including: inputting the SEM image into the convolutional neural network model to obtain characteristic information of the SEM image; training a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model, comprising:
calculating a loss value between the SEM image characteristic information and the target image characteristic information by using a loss function; and adjusting parameters of the convolutional neural network model according to the loss value to obtain the denoising model. The beneficial effects are that: the loss value between the SEM image characteristic information and the target image characteristic information is calculated by utilizing the loss function, and the parameter of the convolutional neural network model is adjusted according to the loss value, so that the denoising model containing the parameter is obtained, and the efficiency and the reliability of obtaining the denoised SEM image are improved by adopting the denoising model.
Optionally, before acquiring the SEM image set after wafer lithography, the method includes: generating a spatial characteristic image, and performing photoetching on the wafer; and scanning the photoetched wafer by adopting a scanning electron SEM to obtain the SEM image set. The beneficial effects are that: photoetching a wafer by generating a spatial characteristic image, scanning the photoetched wafer by adopting a scanning electron SEM to obtain an SEM image set, and calculating according to the SEM image set to obtain a target image set and an SEM image.
In a second aspect, an embodiment of the present invention provides a method for establishing an OPC model, where based on the denoising model, the method includes:
obtaining an SEM image of the wafer after photoetching; inputting the SEM image into the denoising model to obtain a denoising image; and optimizing an OPC model according to the denoised image.
The beneficial effects are that: the acquired SEM image after wafer photoetching is denoised by the denoising model to obtain a denoised image, and the denoising image is adopted to optimize the OPC model, so that the reliability and the efficiency of acquiring the OPC model are improved.
In a third aspect, an embodiment of the present invention provides a system for generating a denoising model, including: the acquisition module is used for acquiring an SEM image set after wafer photoetching; the processing module is connected with the acquisition module and is used for superposing and averaging the pixel values of the SEM images with the same coordinate in the SEM image set to obtain a target image set; the processing module is further used for performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information; the acquisition module is further configured to acquire target image feature information of the target image set; and the training module is connected with the processing module and is used for training the convolutional neural network model by utilizing the SEM image characteristic information and the target image characteristic information to obtain a denoising model.
The beneficial effects are that: the method comprises the steps that an SEM image set after wafer photoetching is obtained through an obtaining module, a processing module superposes SEM image pixel values of the same coordinates in the SEM image set to obtain an average value, a target image set is obtained, a convolutional neural network model is used for carrying out feature extraction on SEM images in the SEM image set to obtain SEM image feature information, the obtaining module is further used for obtaining target image feature information of the target image set, a training module trains the convolutional neural network model through the SEM image feature information and the target image feature information to obtain a denoising model, denoising of the SEM images is achieved through the denoising model, and reliability and efficiency of denoising are improved.
Optionally, the set of SEM images includes the SEM images at N identical coordinates, where N is greater than or equal to 6. The beneficial effects are that: by acquiring N SEM images under the same coordinate, wherein N is more than or equal to 6, various random noises generated by the images under the coordinate are ensured to be acquired as much as possible, and the reliable target image set is obtained conveniently by means of superposition and averaging.
Optionally, performing feature extraction on the SEM image in the SEM image set by using a convolutional neural network model to obtain SEM image feature information, including: inputting the SEM image into the convolutional neural network model to obtain characteristic information of the SEM image; the training module trains a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model, and the training module comprises: calculating a loss value between the SEM image characteristic information and the target image characteristic information by using a loss function; and adjusting parameters of the convolutional neural network model according to the loss value to obtain the denoising model. The beneficial effects are that: the loss value between the SEM image characteristic information and the target image characteristic information is calculated by utilizing the loss function, and the parameter of the convolutional neural network model is adjusted according to the loss value, so that the denoising model containing the parameter is obtained, and the efficiency and the reliability of obtaining the denoised SEM image are improved by adopting the denoising model.
Optionally, before the acquiring unit acquires the SEM image set after the wafer lithography, the acquiring unit includes: generating a spatial characteristic image, and performing photoetching on the wafer; and scanning the photoetched wafer by adopting a scanning electron SEM to obtain the SEM image set. The beneficial effects are that: photoetching a wafer by generating a spatial characteristic image, scanning the photoetched wafer by adopting a scanning electron SEM to obtain an SEM image set, and calculating according to the SEM image set to obtain a target image set and an SEM image.
In a fourth aspect, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and an execution program stored in the memory and executable on the processor, and the processor executes the execution program to implement the method described above.
The electronic equipment has the beneficial effects that: the running program is executed by the processor to realize the running of the method.
In a fifth aspect, the embodiment of the present invention provides a computer-readable storage medium, on which an operating program is stored, and the operating program, when executed by a processor, implements the method described above.
The computer-readable storage medium of the present invention is advantageous in that the execution of the method described above is realized by executing an execution program.
Drawings
FIG. 1 is a schematic flow chart of a conventional method for obtaining a denoised image;
FIG. 2 is a schematic flow chart of a method for generating a denoising model according to an embodiment of the present invention;
FIG. 3 is a flowchart of training a convolutional neural network model according to an embodiment of the present disclosure;
FIG. 4 is a schematic comparison diagram of an image denoised by the denoising model according to the embodiment of the present invention;
FIG. 5 is a flowchart illustrating a method for establishing an OPC model according to an embodiment of the present invention;
FIG. 6 is a schematic flowchart of a process for building an OPC model by using a denoising model according to an embodiment of the present invention;
fig. 7 is a schematic diagram of a system for generating a denoising model according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the present invention, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention. Unless defined otherwise, technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs. As used herein, the word "comprising" and similar words are intended to mean that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items.
At present, in order to establish an accurate OPC model, a selected graph or a critical dimension value is usually measured for multiple times, and as shown in fig. 1, the influence of random noise is reduced by averaging the multiple measurements, but the measurement and calculation of the method are complex, the subjectivity is strong, the efficiency is low, and the reliability is insufficient.
For solving the problems in the prior art, an embodiment of the present invention provides a method for generating a denoising model, which is shown in fig. 2 and fig. 3 and includes:
s201: and acquiring an SEM image set after the wafer is subjected to photoetching.
Before the step, a spatial characteristic image is generated according to a mask plate pattern, then a photoetching process is carried out on the wafer, namely, a photoetching image is formed on the surface of the wafer, and scanning is carried out on the photoetching wafer by adopting a scanning electronic SEM (scanning Electron microscope) to obtain an SEM image set. It should be noted that the set of SEM images includes SEM images at different coordinates, and the number of SEM images at each same coordinate system is at least 6, and the size of the SEM images is determined as the case may be.
S202: and superposing the SEM image pixel values with the same coordinate in the SEM image set, and averaging to obtain a target image set.
In the step, because the obtained SEM images have noise randomly after the wafer is subjected to photoetching, the pixel values of the SEM images with the same coordinate are superposed and averaged to offset the randomly generated noise, so that a more reliable target image set is obtained.
S203: and performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information.
In the step, feature extraction is performed on the SEM images in the SEM image set through a convolutional neural network model to obtain SEM image feature information.
S204: and acquiring the characteristic information of the target image set.
S205: and training a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model.
In the step, the SEM image characteristic information and the target image characteristic information are used for training the convolutional neural network model, and parameters in the convolutional neural network model are adjusted to obtain a denoising model. Namely, the denoising model is obtained by training a convolutional neural network model.
Specifically, the SEM image set includes a training image and a verification image, a loss value between the training image feature information and the target image feature information is calculated using a loss function, and a parameter of the convolutional neural network model is adjusted according to the loss value to obtain the denoising model. And when the precision of the image output by the trained convolutional neural network model and the precision of the image in the target image set meet the requirement, stopping the iterative update of parameters in the convolutional neural network model to obtain a denoising model, wherein the denoising model can be put into practical application at the moment.
In this embodiment, a target image set is obtained by obtaining an SEM image set after wafer lithography, superimposing SEM image pixel values of the same coordinates in the SEM image set, and then averaging, feature extraction is performed on an SEM image in the SEM image set by using a convolutional neural network model to obtain SEM image feature information, and a denoising model is obtained by obtaining target image feature information in the target image set and training the convolutional neural network model by using the SEM image feature information and the target image feature information. With reference to fig. 4, fig. 4 is a schematic comparison diagram of the image denoising by using the denoising model, and the denoising of the SEM image is realized by using the denoising model, so that the denoising reliability and efficiency are improved, and the image quality is improved.
In this embodiment, the convolutional neural network model has 15 layers in total, 13 layers in total except for the input layer and the output layer, and 13 layers in total have the same structure, the convolution kernel size is 3 × 3, and 64 convolution kernels are arranged in each layer. Each convolutional layer is followed by Batch Normalization (BN) and uses a linear rectification function (ReLU) as the activation function. The input layer only performs convolution and activation, and the output layer only performs convolution operation. In practice, however, other convolutional neural network models may be used.
In another embodiment of the disclosure, a method for establishing an OPC model is provided, which is shown in fig. 5 and 6, and includes:
s501, obtaining an SEM image of the wafer after photoetching.
And scanning the photoetched wafer by adopting a scanning electron SEM (scanning Electron microscope), and then acquiring a scanned SEM image.
S502, inputting the SEM image into the denoising model to obtain a denoising image.
In the step, the denoising model is used to obtain the denoised image, so that random noise which may exist is filtered, and the precision of the image is improved.
And S503, optimizing an OPC model according to the denoised image.
In the step, the noise-removed image is adopted to establish the OPC model, so that the reliability and the efficiency of obtaining the OPC model are improved.
In an embodiment of the disclosure, a system for generating a denoising model is provided, and as shown in fig. 7, the system includes: the processing module 702 is connected to the obtaining module 701 and is configured to superimpose and average SEM image pixel values of the same coordinates in the SEM image set to obtain a target image set. The processing module 702 is further configured to perform feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information, and the obtaining module 701 is further configured to obtain target image feature information of the target image set. The training module 703 is connected to the processing module 702, and trains a convolutional neural network model by using the SEM image feature information and the target image feature information to obtain a denoising model.
In the embodiment, an obtaining module 701 obtains an SEM image set after wafer lithography, a processing module 702 superimposes SEM image pixel values of the same coordinates in the SEM image set to obtain an average value, a target image set is obtained, a convolutional neural network model is used to perform feature extraction on an SEM image in the SEM image set to obtain SEM image feature information, the obtaining module 701 is further used to obtain target image feature information of the target image set, a training module 703 trains the convolutional neural network model by using the SEM image feature information and the target image feature information to obtain a denoising model, denoising of the SEM image is realized by using the denoising model, and reliability and efficiency of denoising are improved.
Optionally, the set of SEM images includes the SEM images at N identical coordinates, where N is greater than or equal to 2. In practical applications, N is usually 6. By acquiring N SEM images under the same coordinate, various random noises generated by the images under the coordinate are ensured to be acquired as much as possible, and the superposition and the averaging are facilitated to obtain a reliable target image set.
Optionally, the training module 703 trains a convolutional neural network model by using the SEM image feature information and the target image feature information to obtain a denoising model, specifically, calculates a loss value between the SEM image feature information and the target image feature information by using a loss function, and adjusts a parameter of the convolutional neural network model according to the loss value to obtain the denoising model including the parameter. Therefore, the denoising model is adopted to improve the efficiency and reliability of obtaining the denoised SEM image.
Optionally, before acquiring the SEM image set after the wafer lithography, the acquiring unit 701 generates a spatial feature image according to a mask pattern, performs lithography on the wafer, and scans the wafer after the lithography by using a scanning electron SEM to obtain the SEM image set. And calculating according to the SEM image set to obtain a target image set and an SEM image.
In a further embodiment of the present disclosure, an electronic device is provided, which includes a memory, a processor, and a running program stored in the memory and executable on the processor, and the processor executes the running program to implement the method as described above.
In another embodiment of the present disclosure, a computer-readable storage medium is provided, on which an operating program is stored, which when executed by a processor implements the above-described method.
Each functional unit in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a storage medium. Based on such understanding, the technical solutions of the embodiments of the present application may be substantially implemented or may be part of or all of the technical solutions of the embodiments of the present application may be embodied in the form of a software product stored in a storage medium or a processor executing all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as flash memory, removable hard disk, read-only memory, random access memory, magnetic disk or optical disk
The above description is only a specific implementation of the embodiments of the present application, but the scope of the embodiments of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the embodiments of the present application should be covered by the scope of the embodiments of the present application. Therefore, the protection scope of the embodiments of the present application shall be subject to the protection scope of the claims.

Claims (9)

1. A method for generating a denoising model, comprising:
acquiring an SEM image set after wafer photoetching;
superposing and averaging the SEM image pixel values with the same coordinate in the SEM image set to obtain a target image set;
performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information;
acquiring target image characteristic information of the target image set;
and training a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model.
2. The method of claim 1, wherein the set of SEM images includes the SEM images at N identical coordinates, N being equal to or greater than 2.
3. The method according to claim 1 or 2, wherein performing feature extraction on the SEM images in the set of SEM images by using a convolutional neural network model to obtain SEM image feature information comprises:
inputting the SEM image into the convolutional neural network model to obtain characteristic information of the SEM image;
training a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model, comprising:
calculating a loss value between the SEM image characteristic information and the target image characteristic information by using a loss function;
and adjusting parameters of the convolutional neural network model according to the loss value to obtain the denoising model.
4. A method for building an OPC model, based on the denoising model of any one of claims 1 to 3, the method comprising:
obtaining an SEM image of the wafer after photoetching;
inputting the SEM image into the denoising model to obtain a denoising image;
and optimizing an OPC model according to the denoised image.
5. A system for generating a denoising model, comprising:
the acquisition module is used for acquiring an SEM image set after wafer photoetching;
the processing module is connected with the acquisition module and is used for superposing and averaging the pixel values of the SEM images with the same coordinate in the SEM image set to obtain a target image set;
the processing module is further used for performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information;
the acquisition module is further configured to acquire target image feature information of the target image set;
and the training module is connected with the processing module and is used for training the convolutional neural network model by utilizing the SEM image characteristic information and the target image characteristic information to obtain a denoising model.
6. The system of claim 5, wherein the set of SEM images includes the SEM images at N identical coordinates, where N is equal to or greater than 6.
7. The system according to claim 5 or 6, wherein performing feature extraction on the SEM images in the SEM image set by using a convolutional neural network model to obtain SEM image feature information comprises:
inputting the SEM image into the convolutional neural network model to obtain characteristic information of the SEM image;
the training module trains a convolutional neural network model by using the SEM image characteristic information and the target image characteristic information to obtain a denoising model, and the training module comprises:
calculating a loss value between the SEM image characteristic information and the target image characteristic information by using a loss function;
and adjusting parameters of the convolutional neural network model according to the loss value to obtain the denoising model.
8. An electronic device comprising a memory, a processor, and a running program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 3 or 4 when executing the running program.
9. A readable storage medium on which an operating program is stored, wherein the operating program, when executed by a processor, implements the method of any one of claims 1 to 3, or claim 4.
CN202111392331.0A 2021-11-19 2021-11-19 Denoising model generation method, system, device and medium Pending CN114049342A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111392331.0A CN114049342A (en) 2021-11-19 2021-11-19 Denoising model generation method, system, device and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111392331.0A CN114049342A (en) 2021-11-19 2021-11-19 Denoising model generation method, system, device and medium

Publications (1)

Publication Number Publication Date
CN114049342A true CN114049342A (en) 2022-02-15

Family

ID=80211136

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111392331.0A Pending CN114049342A (en) 2021-11-19 2021-11-19 Denoising model generation method, system, device and medium

Country Status (1)

Country Link
CN (1) CN114049342A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117094909A (en) * 2023-08-31 2023-11-21 青岛天仁微纳科技有限责任公司 Nanometer stamping wafer image acquisition processing method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117094909A (en) * 2023-08-31 2023-11-21 青岛天仁微纳科技有限责任公司 Nanometer stamping wafer image acquisition processing method
CN117094909B (en) * 2023-08-31 2024-04-02 青岛天仁微纳科技有限责任公司 Nanometer stamping wafer image acquisition processing method

Similar Documents

Publication Publication Date Title
CN110603626B (en) Wafer inspection using differential images
JP5276854B2 (en) Pattern generation apparatus and pattern shape evaluation apparatus
CN110361926B (en) Optical proximity effect correction model, establishment method thereof and mask forming method
CN115797300A (en) Edge detection method and device based on adaptive gradient threshold canny operator
WO2022121736A1 (en) Cdsem image virtual measurement method based on machine learning
CN114049342A (en) Denoising model generation method, system, device and medium
JP5810031B2 (en) Semiconductor circuit pattern measuring apparatus and method
TWI836130B (en) Method for enhancing an image and inspection system
TWI617899B (en) Method for determining the dose corrections to be applied to an ic manufacturing process by a matching procedure
CN102376093B (en) Calibration method of camera
JP2014206550A (en) Pattern shape evaluation device
CN115752760A (en) Phase recovery algorithm suitable for micro-vibration environment
CN112797917B (en) High-precision digital speckle interference phase quantitative measurement method
TWI741541B (en) Image analysis system and image analysis method
CN110969582B (en) Remote sensing image blind restoration method based on multi-source reference information
JP7042358B2 (en) Image processing method and image processing device
CN110047032B (en) Local self-adaptive mismatching point removing method based on radial basis function fitting
JP2021132159A (en) Feature value measurement method and feature value measurement device
Sezestre et al. Model-based contour extraction: an enabler for very low-frame SEM images metrology
JP7094594B2 (en) Processing condition estimation device, method and program
TWI846977B (en) Processing condition estimation device, processing condition estimation method, and processing condition estimation recording medium
TWI845970B (en) Inspection equipment and measuring equipment
CN114897717B (en) Multiframe correction blind deconvolution method for sun image restoration
US20230298854A1 (en) Method of performing metrology on a microfabrication pattern
CN113554585B (en) Image analysis system and image analysis method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination