CN-115698690-B - Image processing method, shape inspection method, image processing system, and shape inspection system

CN115698690BCN 115698690 BCN115698690 BCN 115698690BCN-115698690-B

Abstract

The invention provides an image processing method, a shape checking method, an image processing system and a shape checking system. The image processing method acquires data of an estimated photographed image used when comparing an estimated photographed image obtained from reference data of a sample with an actual photographed image of the sample by using a system including an input receiving unit that receives input of the reference data, process information of the sample, and learned model data, an estimating unit that calculates photographed image statistics indicating probability distribution of values that can be taken by the data of the photographed image by using the reference data, the process information, and the model data, and an output unit that outputs the photographed image statistics and generates an estimated photographed image from the photographed image statistics. In this way, when the simulated image estimated from the design data and the actually captured image are compared, the time required for the estimation can be shortened, and the comparison can be performed in real time.

Inventors

OUCHI MASANORI
ISHIKAWA MASANORI
NITTA YASUTAKA
SHINDO HIROYUKI

Assignees

株式会社日立高新技术

Dates

Publication Date: 20260505
Application Date: 20200616

Claims (14)

1. An image processing method for acquiring data of an estimated photographed image obtained from reference data of a sample to be inspected and an actual photographed image of the sample, using a system having an input receiving unit, an estimating unit, and an output unit, The image processing method comprises the following steps: An input step of receiving input of the reference data, step information including at least one of a manufacturing condition of the sample and a photographing condition of the photographed image, and learned model data; An estimating step of calculating a captured image statistic as a probability density function representing a probability distribution of each pixel value of the captured image data using the reference data, the step information, and the model data, and An output unit configured to output the captured image statistics, The estimated captured image can be generated from the captured image statistics, The process information is combined with the feature quantity of the captured image.
2. The image processing method according to claim 1, wherein, The system also includes a machine learning portion and a storage portion, The image processing method further includes a learning necessity determining step of determining necessity of learning for the model data by the machine learning section, In the learning necessity determining step, when the necessity of learning is determined to be necessary, Receiving input of a learning data set including the reference data for learning, the process information and the captured image, A comparison of the captured image statistics and the data of the captured images of the learning dataset is made, Updating the model data based on the result of the comparison, In the learning necessity determining step, when the necessity of learning is determined to be unnecessary, The storage unit stores parameters used by the estimation unit in calculating the captured image statistics as the model data.
3. The image processing method according to claim 1, wherein, The image processing method further includes a step of evaluating an influence of the process information on the sample using the captured image statistics.
4. The image processing method according to claim 1, wherein, The captured image statistics include an average image and a standard deviation image.
5. The image processing method according to claim 1, wherein, The test sample is a semiconductor circuit.
6. A shape inspection method for inspecting a shape of the sample using the captured image statistics obtained by the image processing method as set forth in claim 1, wherein, The system also comprises a template image making part and a pattern matching processing part, The input receiving section receives an input of data of the captured image, The template image creating unit creates a template image based on the captured image statistics, The pattern matching processing section performs pattern matching of the template image and the photographed image, The output section outputs a result of the pattern matching.
7. A shape inspection method for inspecting a shape of the sample using the captured image statistics obtained by the image processing method as set forth in claim 2, wherein, The system also comprises a template image making part and a pattern matching processing part, The input receiving section receives an input of data of the captured image, The template image creating unit creates a template image based on the captured image statistics, The pattern matching processing section performs pattern matching of the template image and the photographed image, The output section outputs a result of the pattern matching.
8. An image processing system that acquires data of an estimated photographed image obtained from reference data of a sample to be inspected when comparing the estimated photographed image with an actual photographed image of the sample, The image processing system is provided with: An input receiving unit that receives input of the reference data, process information including at least one of a manufacturing condition of the sample and a shooting condition of the shot image, and learned model data; An estimation unit that calculates a captured image statistic as a probability density function representing a probability distribution of each pixel value of the captured image data using the reference data, the process information, and the model data, and An output unit configured to output the captured image statistic, The estimated captured image can be generated from the captured image statistics, The process information is combined with the feature quantity of the captured image.
9. The image processing system of claim 8, wherein, The image processing system further includes a machine learning section and a storage section, The machine learning section determines necessity of learning for the model data, In the case where the machine learning section determines that the necessity of learning is required, Receiving input of a learning data set including the reference data for learning, the process information and the captured image, A comparison of the captured image statistics and the data of the captured images of the learning dataset is made, Updating the model data based on the result of the comparison, In the case where the machine learning section determines that the necessity of learning is not required, The storage unit stores parameters used by the estimation unit in calculating the captured image statistics as the model data.
10. The image processing system of claim 8, wherein, And evaluating an influence of the process information on the sample using the captured image statistics.
11. The image processing system of claim 8, wherein, The captured image statistics include an average image and a standard deviation image.
12. The image processing system of claim 8, wherein, The test sample is a semiconductor circuit.
13. A shape inspection system comprising the image processing system according to claim 8, further comprising a template image creating unit and a pattern matching unit, wherein the shape of the sample is inspected using the captured image statistics, The input receiving section receives an input of data of the captured image, The template image creating unit creates a template image based on the captured image statistics, The pattern matching processing section performs pattern matching of the template image and the photographed image, The output section outputs a result of the pattern matching.
14. A shape inspection system comprising the image processing system according to claim 9, further comprising a template image creating unit and a pattern matching unit, wherein the shape of the sample is inspected using the captured image statistics, The input receiving section receives an input of data of the captured image, The template image creating unit creates a template image based on the captured image statistics, The pattern matching processing section performs pattern matching of the template image and the photographed image, The output section outputs a result of the pattern matching.

Description

Image processing method, shape inspection method, image processing system, and shape inspection system Technical Field The invention relates to an image processing method, a shape inspection method, an image processing system, and a shape inspection system. Background Conventionally, in order to perform evaluation (defect inspection or the like) and dimensional measurement using image data, design data of an object to be evaluated or a dimensional measurement object is compared with a captured image. As an example of the object, there is a semiconductor circuit. In inspection and measurement of a semiconductor circuit (hereinafter, also simply referred to as a "circuit"), design data of the circuit and captured image data (hereinafter, also simply referred to as a "captured image") are compared to perform a process of aligning the positions thereof. This process is called pattern matching. By aligning the design data and the position of the captured image, it is possible to specify the measurement point, evaluate the degree of deviation from the circuit shape on the design data, and the like. In the circuit, there is a shape deformation caused by each condition set in the manufacturing process. In addition, differences in image quality (contrast variation, generation of image noise, and the like) due to the respective conditions set in the imaging process occur in the captured image of the circuit. In addition, even under the same conditions, the shape of the circuit and the image quality of the captured image change due to the fluctuation thereof. For example, in the case where design data is directly used as a template image in pattern matching, alignment becomes difficult due to a difference in circuit shape on the design data and circuit shape on a captured image. Therefore, in the template image, data close to the circuit shape on the captured image is preferably used, as compared with directly using the design data. Patent document 1 discloses a computer-based mounting method for generating a simulation image from design information, which includes a step of determining features of design information of an object by inputting the design information to two or more encoder layers that generate a model, and a step of generating one or more simulation images by inputting the determined features to two or more decoder layers that generate the model. Here, the simulated image represents design information represented by an image of the object generated by the image system. Patent document 1 discloses that the generative model can be replaced by a Convolutional Neural Network (CNN). Patent document 2 discloses a pattern inspection system that performs inspection of an image of an inspection target pattern using a recognizer configured by machine learning based on the image of the inspection target pattern of an electronic device and data used for manufacturing the inspection target pattern, stores a plurality of pattern images of the electronic device and pattern data used for manufacturing the pattern of the electronic device, and selects a learning pattern image used for machine learning from the plurality of pattern images based on the stored pattern data and pattern image, thereby saving labor for a truth-value manufacturing operation of learning data, realizing reduction of learning data, and shortening learning time. Prior art literature Patent literature Patent document 1 specification of U.S. Pat. No. 9965901 Patent document 2 Japanese patent application laid-open No. 2020-35282 Disclosure of Invention Problems to be solved by the invention According to the method disclosed in patent document 1, when applied to a circuit pattern to be inspected, although a circuit pattern as an analog image can be obtained, since only design data is input, a difference in conditions (hereinafter also referred to as "process information") of a manufacturing process, an imaging process, and the like cannot be specified clearly. In order to obtain the difference in the condition, it is necessary to prepare a data set of a photographed image including a circuit to be manufactured or photographed under the condition, and learn a mathematical model for simulation by condition. In order to know the influence of the process information on the circuit and the captured image thereof, it has conventionally been necessary to perform a plurality of simulations on a conditional basis. The conventional simulator uses the monte carlo method or the like, and thus takes time for simulation. Further, the process simulation of a commercially available semiconductor circuit is classified into photolithography, etching, photographing processes, and the like by process. In order to comprehensively grasp the relationship of parameters between the steps by combining these steps, it is necessary to use a simulator in a plurality of stages. However, since the simulation of the manufacturing or photographing process requires a long time for cal