Methods for image simulation, pseudo-random defect dataset generation, and micro and nano defects detection

Patent metadata

Field	Value
Publication number	US-12307654-B1
Application number	US-202519031144-A
Country	US
Kind code	B1
Filing date	Jan 17, 2025
Priority date	Mar 29, 2024
Publication date	May 20, 2025
Grant date	May 20, 2025

Abstract

The embodiments of the present disclosure provide a method for an image simulation generation based on a near-field simulation of a computational electromagnetic field, the method comprising: constructing a simulated three-dimensional model based on model parameters; constructing a simulated Kohler illumination model based on a light source parameter; using a degree of similarity change in a synthesized image under incremental aperture diaphragm sampling points as a criterion for approximate convergence of the simulation to determine a count of samples to be used for a balancing combination of simulation cost and accuracy; an optical simulation is performed based on the simulated three-dimensional model and the simulated Kohler illumination model, and a far-field electromagnetic field distribution data is obtained to obtain a simulated image by synthesizing the image. In the generation of a large count of simulated images on the basis of pseudo-random defect dataset generation may be further realized, in the acquisition of a large count of datasets, the dataset of defect inspection model may be trained, in order to achieve a direct detection for patterned wafer defective images and solve the problem of difficult access to reference images in a process of patterned wafer defect detection.

First claim

What is claimed is:

1. A method for an image simulation generation based on a near-field simulation of a computational electromagnetic field, wherein the method comprises:
constructing, based on model parameters, a simulated three-dimensional model, the model parameters including a model material parameter, a model three-dimensional geometric structure parameter, and a model boundary computation domain;
constructing, based on a light source parameter, a simulated Kohler illumination model, the light source parameter including an initial wavelength of a plane wave, a polarization amplitude of the plane wave, an angle of incidence of the plane wave, and an azimuthal angle of an incident plane wave;
using a degree of similarity change in a synthesized image under incremental aperture diaphragm sampling points as a criterion for approximate convergence of the simulation to determine a count of samples to be used for a balancing combination of simulation cost and accuracy;
performing an optical simulation based on the simulated three-dimensional model and the simulated Kohler illumination model, the optical simulation including simulating by simulation environment constructed by combining the simulated three-dimensional model and the simulated Kohler illumination model and synthesizing an image, by obtaining far-field electromagnetic field distribution data to synthesizing an image to obtain a simulated image;
wherein the using the degree of similarity change in the synthesized image under incremental aperture diaphragm sampling points as a criterion for approximate convergence of the simulation to determine a count of samples to be used for balancing the combination of the simulation cost and accuracy includes:
giving an initial sampling density n input and judgment thresholds T 1 , T 2 , T* for an image similarity;
defining n=n input and increasing a sampling density sequentially according to rule of n=n+1 to obtain a simulated image I n corresponding to each of different sampling densities;
assessing an image similarity of the simulated image I n with adjacent simulated images I n−1 , and I n−2 using the Structural Similarity Index Measure (SSIM) as a metric; when SSIM (I n , I n−1 )>T 1 and SSIM (I n , I n−2 )>T 2 are satisfied, determining that the current n is a sampling point when the simulated image has sufficiently converged; defining n 0 *=n; wherein at this time, n 0 * is the sampling point when the simulated image has sufficiently converged, and the corresponding simulated image is I 0 *;
redefining n=n input and increasing a sampling density sequentially according to the rule of n=n+1 again to obtain the simulated image I n corresponding to each of different sampling densities;
comparing an image similarity between the obtained simulated image I n with the simulated image I 0 * corresponding to the convergent sampling point no; wherein when SSIM (I n , I 0 *)>T*, n≤n 0 *, and n is an odd number are satisfied, the n is a desired optimal sampling value, and defining n best =n to yield the optimal sampling density n best ;
calculating the Structural Similarity Index Measure (SSIM) by a following equation:

SSIM
⁡
(

I
x

,

I
y

)

=

(

2
⁢

μ
x

⁢

μ
y

+

C
1

)

⁢

(

2
⁢

σ
xy

+

μ
x

+

C
2

)

(

μ
x
2

+

μ
y
2

+

C
1

)

⁢

(

σ
x
2

+

σ
y
2

+

C
2

)

,

(
1
)

wherein I x and I y represent two images at sampling points x and y, respectively, μ x and μ y are averages of all pixels in the two images, σ x and σ y are grayscale standard deviation of the two images, σ xy is a covariance, and C 1 and C 2 are empirical constants;
performing an optical simulation based on the simulated three-dimensional model and the simulated Kohler illumination model, by obtaining far-field electromagnetic field distribution data to synthesizing an image to obtain a simulated image includes:
obtaining near-field electric fie

Assignees

• Univ Nanjing Aeronautics & Astronautics
• Suzhou Research Institute Of Nanjing Univ Of Aeronautics And Astronautic
• Suzhou Research Institute Of Nanjing Univ Of Aeronautics And Astronautics

Inventors

• Meng Kai
• Zhou Junquan
• Zhang Hangying
• Meng Chao
• Lou Peihuang
• Qian Xiaoming
• Wu Xing

Classifications

• G03F7/706837

  Data analysis, e.g. filtering, weighting, flyer removal, fingerprints or root cause analysis · CPC title

• G03F7/7065

  Defects, e.g. optical inspection of patterned layer for defects · CPC title

• G06T15/506

  Illumination models · CPC title

• G06T2207/10056

  Microscopic image · CPC title

• G06T2207/20081

  Training; Learning · CPC title