# US 11,443,419 B2 · Reference image generation method and pattern inspection method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,443,419 B2
- Document ID: US11443419B2
- Status: Granted
- Kind: B2
- Application number: 16/817,932
- Filed: 2020-03-13
- Published: 2020-09-17
- Granted: 2022-09-13
- Earliest claimed: 2019-03-13
- Assignees: NuFlare Technology, Inc.
- Applicants: NuFlare Technology, Inc.
- Inventors: Yoshitaka Yasui; Ikunao Isomura
- Web page: https://patentgrep.com/patent/US11443419B2

## CPC

- G06T 7/001: {using an image reference approach}
- G06T 7/12: Edge-based segmentation
- G06T 7/13: Edge detection
- G06T 2207/10061: from scanning electron microscope
- G06T 2207/30148: Semiconductor; IC; Wafer
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

To include reading design data of a plurality of patterns formed on a sample and characteristic information indicating characteristics of each of the patterns from a storage device, the characteristic information being additionally written in the design data, dividing a pattern formed region of the sample on which the patterns are formed, into a plurality of regions where the characteristics are different from each other on a basis of the characteristic information, calculating parameter information according to the characteristics with respect to each of the regions, where the parameter information is provided for generating a reference image from the design data to be used in an inspection of the patterns, and generating the reference image from the design data on a basis of the calculated parameter information.

## Claims

### Claim 1

A reference image generation method comprising:
reading design data of a plurality of patterns formed on a sample and characteristic information indicating characteristics of each of the patterns from a storage device, the characteristic information being additionally written in the design data;
dividing a pattern formed region of the sample on which the patterns are formed, into a plurality of regions where the characteristics within each region are of a similar pattern type and characteristics of different of the plurality of regions are different from each other, on a basis of the characteristic information and pattern type;
calculating parameter information according to the characteristics with respect to each of the regions, where the parameter information is filter coefficients, so that each of the plurality of regions has a respective filter characteristic, for generating a reference image by calculating with the design data to be used in an inspection of the patterns; and
generating the reference image from the design data on a basis of the calculated parameter information.

### Claim 2 (depends on 1)

The method of claim 1, wherein the characteristic information is pattern type information indicating types of the patterns.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
before the dividing of the pattern formed region into the regions,
calculating a density and number of edges of each of the patterns of the design data with respect to each unit of the design data having a same image size as a generation unit of the reference image;
generating the pattern type information by determining a type of each of the patterns with respect to each unit of the design data on a basis of the calculated density and the calculated number of edges; and
adding the pattern type information to the design data with respect to each unit of the design data.

### Claim 4 (depends on 3)

The method of claim 3, wherein the generating of the pattern type information comprises determining a type of each of the patterns by performing labeling to classify a plurality of patterns where densities and numbers of edges are close to each other into a same group on a basis of the calculated density and the calculated number of edges.

### Claim 5 (depends on 1)

The method of claim 1, wherein the characteristic information is at least either inspection rank information indicating inspection ranks of the patterns or contour information indicating contours of the patterns.

### Claim 6 (depends on 5)

The method of claim 5, comprising
before the dividing of the pattern formed region into the regions,
adding at least either the inspection rank information or the contour information to the design data in response to a user operation.

### Claim 7

A pattern inspection method comprising:
reading design data of a plurality of patterns formed on a sample and characteristic information indicating characteristics of each of the patterns from a storage device, the characteristic information being additionally written in the design data;
dividing a pattern formed region of the sample on which the patterns are formed, into a plurality of regions where the characteristics within each region are of a similar pattern type and characteristics of different of the plurality of regions are different from each other, on a basis of the characteristic information;
calculating parameter information according to the characteristics with respect to each of the regions, where the parameter information is filter coefficients, so that each of the plurality of regions has a respective filter characteristic, for generating a reference image by calculating with the design data to be used in an inspection of the patterns;
generating the reference image from the design data on a basis of the calculated parameter information; and
inspecting a defect of the patterns on a basis of the reference image and an optical image obtained by imaging the sample.
