Image generation apparatus, inspection apparatus, and image generation method

Patent metadata

Field	Value
Publication number	US-11200661-B2
Application number	US-202016782243-A
Country	US
Kind code	B2
Filing date	Feb 5, 2020
Priority date	Feb 15, 2019
Publication date	Dec 14, 2021
Grant date	Dec 14, 2021

Abstract

An image generation apparatus configured to generate a substrate image for inspection regarding a defect on a substrate, the substrate having a frame pattern formed on a surface thereof, the frame pattern being a unique pattern for each kind of a treatment recipe for the substrate, the image generation apparatus including: a region estimator configured to estimate a region corresponding to the frame pattern in a substrate image of an inspection object based on an identification model, the identification model being acquired by machine learning in advance and for identifying an image of the frame pattern included in a substrate image; and an eraser configured to erase the image of the frame pattern from the substrate image of the inspection object based on an estimation result by the region estimator to generate the substrate image for inspection.

First claim

What is claimed is:

1. An image generation apparatus that generates a substrate image for inspection regarding a defect on a substrate,
the substrate having a frame pattern formed on a surface thereof, the frame pattern being a unique pattern for each kind of a treatment recipe for the substrate,
the image generation apparatus comprising:
a controller having a processor and a memory, the memory storing instructions that, when executed by the processor, cause the processor to
estimate a region corresponding to the frame pattern in a substrate image of an inspection object based on an identification model, the identification model being acquired by machine learning in advance and for identifying an image of the frame pattern included in a substrate image;
erase the image of the frame pattern from the substrate image of the inspection object based on an estimation result by the region estimator to generate the substrate image for inspection; and
acquire the identification model by the machine learning in advance using a data set including a substrate image and information on presence or absence of the image of the frame pattern in the substrate image.

2. The image generation apparatus according to claim 1 , wherein
the substrate image included in the data set includes both a frameless substrate image being a substrate image relating to a substrate which is not formed with the frame pattern but has defects, and a framed substrate image being a substrate image made by combining a substrate image relating to a substrate which has a known frame pattern and is formed with no defects with the frameless substrate image.

3. The image generation apparatus according to claim 2 , wherein
the frame pattern is at least one of a region where semiconductor devices are not formed, a boundary line between the semiconductor devices, and patterns on surfaces of the semiconductor devices.

4. The image generation apparatus according to claim 2 , wherein
the defect is at least one of a portion not having a predetermined shape in a coating film, a flaw on the substrate, and a foreign matter on the substrate.

5. The image generation apparatus according to claim 2 , wherein
the machine learning in advance is deep learning.

6. The image generation apparatus according to claim 2 , configured as an inspection apparatus, the processor being further configured to determine a kind of a defect formed on an inspection object substrate based on the substrate image for inspection, wherein
the kind of the defect is determined using machine learning.

7. The image generation apparatus according to claim 1 , wherein
the frame pattern is at least one of a region where semiconductor devices are not formed, a boundary line between the semiconductor devices, and patterns on surfaces of the semiconductor devices.

8. The image generation apparatus according to claim 1 , wherein
the defect is at least one of a portion not having a predetermined shape in a coating film, a flaw on the substrate, and a foreign matter on the substrate.

9. The image generation apparatus according to claim 1 , wherein
the machine learning in advance is deep learning.

10. The image generation apparatus according to claim 1 , configured as an inspection apparatus, the processor is further configured to determine a kind of a defect formed on an inspection object substrate based on the substrate image for inspection, using machine learning, wherein
the kind of the defect is determined using machine learning.

11. An image generation method for generating a substrate image for inspection regarding a defect on a substrate,
the substrate having a frame pattern formed on a surface thereof, the frame pattern being a unique pattern for each kind of a treatment recipe for the substrate,
the image generation method comprising:
estimating a region corresponding to the frame pattern in a substrate image of an inspection object based on an identification model, the identification model being acquired by machine learning in advance and for identifying an image of the frame pattern included in a substrate image;
erasing the image of the frame pattern from the substrate image of the inspection object based on an estimation result at the estimating to generate the substrate image for inspection; and
acquiring the identification model by the machine learning in advance using a data set including a substrate image and information on presence or absence of the image of the frame pattern in the substrate image.

12. The image generation method according to claim 11 , wherein
the frame pattern is at least one of a region where semiconductor devices are not formed, a boundary line between the semiconductor devices, and patterns on surfaces of the semiconductor devices.

13. The image generation method according to claim 11 , wherein
the defect is at least one of a portion not having a predetermined shape in a coating film, a flaw on the substrate, and a foreign matter on the substrate.

14. The image generation method according to claim 11 , wherein
the machine learning in advance is deep learning.

15. The image generation method according to claim 11 , further comprising
determining a kind of a defect formed on an inspection object substrate based on the substrate image for inspection, wherein
the kind of the defect is determined using machine learning.

Assignees

• Tokyo Electron Ltd

Inventors

• Sasaki Chie

Classifications

• G06F18/24

  Classification techniques · CPC title

• G06T7/001 Primary

  using an image reference approach · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• G06T2207/10004

  Still image; Photographic image · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title