US 11,676,264 B2
System and method for determining defects using physics-based image perturbations
Martin Plihal, Pleasanton, CA (US); Saravanan Paramasivam, Chennai (IN); Jacob George, Cochin (IN); Niveditha Lakshmi Narasimhan, Chennai (IN); Sairam Ravu, Chennai (IN); Somesh Challapalli, Hyderabad (IN); and Prasanti Uppaluri, Saratoga, CA (US)
Assigned to KLA Corporation, Milpitas, CA (US)
Filed by KLA Corporation, Milpitas, CA (US)
Filed on Jul. 21, 2020, as Appl. No. 16/935,159.
Claims priority of provisional application 62/898,761, filed on Sep. 11, 2019.
Prior Publication US 2021/0027445 A1, Jan. 28, 2021
Int. Cl. G06T 7/00 (2017.01); G06N 20/00 (2019.01); G06F 18/214 (2023.01); G06F 18/21 (2023.01); G06V 20/69 (2022.01)
CPC G06T 7/001 (2013.01) [G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06N 20/00 (2019.01); G06V 20/69 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30148 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A system, comprising:
a controller having one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions configured to cause the one or more processors to:
receive one or more training images of one or more defects of a training specimen;
select one or more defect types from the one or more defects for augmentation based on the frequency of the defect types in the one or more training images;
generate one or more augmented images of the one or more defects of the training specimen, wherein the one or more processors selects one or more defect types for augmentation based on the frequency of defect types in the one or more training images;
generate a machine learning classifier based on the one or more augmented images of the one or more defects of the training specimen;
receive one or more target images of one or more target features of a target specimen; and
determine one or more defects of the one or more target features with the machine learning classifier.