System, a method and a computer program product for patch-based defect detection

Patent metadata

Field	Value
Publication number	US-9235885-B2
Application number	US-201313756399-A
Country	US
Kind code	B2
Filing date	Jan 31, 2013
Priority date	Jan 31, 2013
Publication date	Jan 12, 2016
Grant date	Jan 12, 2016

Abstract

A system capable of inspecting an article for defects, the system including: a patch comparator, configured to determine with respect to each of a plurality of reference patches in a reference image a similarity level, based on a predefined patch-similarity criterion and on a source patch defined in the reference image; an evaluation module, configured to rate each inspected pixel out of multiple inspected pixels of the inspection image with a representative score which is based on the similarity level of a reference patch associated with a reference pixel corresponding to the inspected pixel; a selection module, configured to select multiple selected inspected pixels based on the representative scores of the multiple inspected pixels; and a defect detection module, configured to determine a presence of a defect in the candidate pixel based on an inspected value of the candidate pixel and inspected values of the selected inspected pixels.

First claim

What is claimed is:

1. A computerized method for inspecting an article for defects based on processing of inspection images generated by collecting signals arriving from the article, the method comprising:
obtaining a candidate pixel of the inspection image, the candidate pixel being representative of a candidate article defect location;
in a reference image, defining a source patch associated with a reference source pixel of the reference image which corresponds to the candidate pixel;
in the reference image, based on the source patch and a predefined patch-similarity criterion, determining a similarity level with respect to each of a plurality of reference patches, each of which is associated with a reference image pixel;
in the inspection image, rating each inspected pixel out of multiple inspected pixels with a representative score which is based on the similarity level of a reference patch associated with a reference pixel corresponding to the inspected pixel;
in the inspection image, selecting multiple selected inspected pixels based on the representative scores of the multiple inspected pixels;
determining a presence of a defect in the candidate pixel based on an inspected value of the candidate pixel and inspected values of the selected inspected pixels; and
in each image of a plurality of images which includes the reference image:
(a) defining multiple source patches of different shapes, wherein each of the multiple source patches is associated with a reference source pixel of the image; and
(b) for each of the source patches of the image: based on the source patch and a respective patch-similarity criterion, determining a similarity level with respect to each of a plurality of reference patches, each of which is associated with a pixel of the image;
thereby for each of a plurality of pixels of the image, determining a plurality of similarity levels that are determined for multiple reference patches of different shapes which are associated with the respective pixel of the image;

wherein the rating comprises rating each inspected pixel out of at least one of the multiple inspected pixels with a representative score which is based on multiple similarity levels of reference patches associated with pixels of the plurality of images which correspond to the inspected pixel.

2. The method according to claim 1 , wherein the defining of the source patch comprises defining the source patch so that a size of a smallest rectangle in which all pixels of the source patch are enclosed is smaller than 0.05 times a size of the inspection image.

3. The method according to claim 1 , wherein the determining of the presence of the defect is based on a grey level value of the candidate pixel and on a threshold determined based on grey level values of the selected inspected pixels.

4. The method according to claim 1 , comprising:
defining in the reference image multiple source patches of different shapes, wherein each of the multiple source patches is associated with the reference source pixel; and
for each of the multiple source patches, based on that source patch and a respective patch-similarity criterion, determining a similarity level with respect to each of a respective plurality of reference patches, each of which is associated with a reference image pixel; thereby for each of a plurality of reference image pixels, determining a plurality of similarity levels that are determined for multiple reference patches, of different shapes, which are associated with the respective reference image pixel;
wherein the rating of each inspected pixel out of multiple inspected pixels comprises rating each inspected pixel out of a subgroup of the multiple inspected pixels with a representative score which is based on the plurality of similarity levels of the multiple reference patches which are associated with a reference pixel corresponding to the inspected pixel.

5. The method according to claim 1 , further comprising:
defining a plurality of reference patches in a second reference image;
based on a second source patch defined in the second reference image, determining a similarity level with respect to each of the plurality of reference patches in the second reference image to the second source patch, each of plurality of reference patches in the second reference image being associated with a pixel of the second reference image;
wherein at least one of the representative scores with which one of the multiple inspected pixels is rated is based also on a similarity level of a reference patch associated with a reference pixel of the second reference image which corresponds to the inspected pixel.

6. The method according to claim 5 , wherein the at least one representative score is determined based on a group of similarity levels of reference patches from multiple reference images, so that the score indicates a low degree of similarity between an environment of the inspected pixel and an environment of the candidate pixel if any of the respective similarity levels indicates a low degree of similarity.

7. A system capable of inspecting an article for defects, the system comprising:
a memory; and
a processor electronic device operatively coupled to the memory, the processor electronic device to:
determine with respect to each of a plurality of reference patches in a reference image a similarity level, based on a predefined patch-similarity criterion and on a source patch which is defined in the reference image and which is associated with a reference source pixel of the reference image; wherein the reference source pixel corresponds to a candidate pixel of an inspection image that is generated by collecting signals arriving from the article, the candidate pixel being representative of a candidate article defect location; wherein each of the plurality of reference patches is associated with a reference image pixel;
rate each inspected pixel out of multiple inspected pixels of the inspection image with a representative score which is based on the similarity level of a reference patch associated with a reference pixel corresponding to the inspected pixel;
select multiple selected inspected pixels based on the representative scores of the multiple inspected pixels; and
determine a presence of a defect in the candidate pixel based on an inspected value of the candidate pixel and inspected values of the selected inspected pixels,
wherein in each image of a plurality of images, which includes the reference image, multiple source patches of different shapes are defined, each of the multiple source patches is associated with a reference source pixel of the image;
wherein to determine the similarity level comprises the processor electronic device to determine in each image of the plurality of images, for each of the source patches of the image a similarity level with respect to each of a plurality of reference patches based on the source patch and a respective patch-similarity criterion, wherein each of the plurality of reference patches is associated with a pixel of the image;
thereby for each of a plurality of pixels of the image, determining a plurality of similarity levels that are determined for multiple reference patches of different shapes which are associated with the respective pixel of the image;
wherein to rate each inspected pixel comprises the processor electronic device to rate each inspected pixel out of at least one of the multiple inspected pixels with a representative score which is based on multiple similarity levels of reference patches associated with pixels of the plurality of images which correspond to the inspected pixel.

8. The system according to claim 7 , wherein a size of a smallest rectangle in which all pixels of the source patch are enclos

Assignees

• Applied Materials Israel Ltd
• Applied Materials Israel Ltd

Inventors

• Amzaleg Moshe
• Dodzin Nir Ben David
• Gatt Vered
• Hanani Yair
• Rozenman Efrat

Classifications

• H10P74/203

  Structural properties, e.g. testing or measuring thicknesses, line widths, warpage, bond strengths or physical defects · CPC title

• G01N2021/95615

  with stored comparision signal · CPC title

• G01N21/95607

  using a comparative method · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• G06T2207/10061

  from scanning electron microscope · CPC title