Closed-loop automatic defect inspection and classification

Patent metadata

Field	Value
Publication number	US-10901402-B2
Application number	US-201816174070-A
Country	US
Kind code	B2
Filing date	Oct 29, 2018
Priority date	Jul 22, 2013
Publication date	Jan 26, 2021
Grant date	Jan 26, 2021

Abstract

Inspection apparatus includes an imaging module, which is configured to capture images of defects at different, respective locations on a sample. A processor is coupled to process the images so as to automatically assign respective classifications to the defects, and to autonomously control the imaging module to continue capturing the images responsively to the assigned classifications.

First claim

What is claimed is:

1. A system comprising:
a memory; and
a processor, operatively coupled with the memory, to:
receive a first image of a defect on a sample from a first image acquisition mode, wherein the first image acquisition mode uses a first type of detector that collects a first type of image information, wherein the first type of detector is a scanning electronic microscope (SEM);
assign a first classification to the defect on the sample based on the first image;
determine, based on the first classification to the defect, whether to capture a second image of the defect on the sample and which second image acquisition mode to use;
select, based on the first classification to the defect, a second type of detector from a group consisting of an optical inspection tool, a dispersive X-ray tool, and the SEM acquiring a tilt image with respect to the first image;
responsive to determining that a second image of the defect is to be captured from the second image acquisition mode, instruct the selected second type of detector to capture the second image and to collect a second type of image information from the second image;
iteratively update, using an integrated closed-loop automatic defect classification (ADC) algorithm, defect classification criteria based on the second type of image information collected by the second type of detector; and
assign a second classification to the defect based on the updated defect classification criteria applied on the second image.

2. The system of claim 1 , wherein the processor is further to:
determine that information from the first image of the defect is insufficient to classify the defect, wherein the determination of whether to capture the second image is based on the determination that the information from the first image of the defect is insufficient.

3. The system of claim 2 , wherein the instructing of the second type of detector is based on the determination that the information from the first image of the defect is insufficient.

4. The system of claim 1 , wherein the processor is further to:
determine a number of defects on the sample that have been classified; and
determine to capture additional images of additional defects on the sample based on whether the number of defects on the sample that have been classified satisfies a threshold number.

5. A method comprising:
receiving a first image of a defect on a sample from a first image acquisition mode, wherein the first image acquisition mode uses a first type of detector that collects a first type of image information, wherein the first type of detector is a scanning electronic microscope (SEM);
assigning a first classification to the defect on the sample based on the first image;
determining, by a processor, based on the first classification of the defect, whether to capture a second image of the defect on the sample and which second image acquisition mode to use;
selecting, based on the first classification to the defect, a second type of detector from a group consisting of an optical inspection tool, a dispersive X-ray tool, and the SEM acquiring a tilt image with respect to the first image;
responsive to determining that a second image of the defect is to be captured from the second image acquisition mode, instructing the selected second type of detector to capture the second image and to collect a second type of image information from the second image;
iteratively updating, using an integrated closed-loop automatic defect classification (ADC) algorithm, defect classification criteria based on the second type of image information collected by the second type of detector; and
assigning a second classification to the defect based on the updated defect classification criteria applied on the second image.

6. The method of claim 5 , further comprising:
determine that information from the first image of the defect is insufficient to classify the defect, wherein the determination of whether to capture the second image is based on the determination that the information from the first image of the defect is insufficient.

7. The method of claim 6 , wherein the instructing of the second type of detector is based on the determination that the information from the first image of the defect is insufficient.

8. The method of claim 5 , further comprising:
determine a number of defects on the sample that have been classified; and
determine to capture additional images of additional defects on the sample based on whether the number of defects on the sample that have been classified satisfies a threshold number.

9. A non-transitory computer readable medium comprising instructions, which when executed by a processor, cause the processor to perform operations comprising:
receiving a first image of a defect on a sample from a first image acquisition mode, wherein the first image acquisition mode uses a first type of detector that collects a first type of image information, wherein the first type of detector is a scanning electronic microscope (SEM);
assigning a first classification to the defect on the sample based on the first image;
determining, based on the first classification to the defect, whether to capture a second image of the defect on the sample and which second image acquisition mode to use;
selecting, based on the first classification to the defect, a second type of detector from a group consisting of an optical inspection tool, a dispersive X-ray tool, and the SEM acquiring a tilt image with respect to the first image;
responsive to determining that a second image of the defect is to be captured from the second image acquisition mode, instructing the selected second type of detector to capture the second image and to collect a second type of image information from the second image;
iteratively updating, using an integrated closed-loop automatic defect classification (ADC) algorithm, defect classification criteria based on the second type of image information collected by the second type of detector; and
assigning a second classification to the defect based on the updated defect classification criteria applied on the second image.

10. The non-transitory computer readable medium of claim 9 , wherein the operations further comprise:
determining that information from the first image of the defect is insufficient to classify the defect, wherein the determination of whether to capture the second image is based on the determination that the information from the first image of the defect is insufficient.

11. The non-transitory computer readable medium of claim 10 , wherein the instructing of the second type of detector is based on the determination that the information from the first image of the defect is insufficient.

Assignees

• Applied Materials Israel Ltd

Inventors

• Greenberg Gadi
• Kaizerman Idan
• Zohar Zeev

Classifications

• H10P74/23

  characterised by multiple measurements, corrections, marking or sorting processes · CPC title

• Y02P90/02

  Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS] · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• H01J2237/2817

  Pattern inspection · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title