US 11,922,619 B2
Context-based defect inspection
Brian Duffy, San Jose, CA (US); Bradley Ries, San Jose, CA (US); Laurent Karsenti, Rehovot (IL); Kuljit S. Virk, Fremont, CA (US); Asaf J. Elron, Ramat Hasharon (IL); Ruslan Berdichevsky, Zoran (IL); Oriel Ben Shmuel, Netanya (IL); Shlomi Fenster, Beit Shemesh (IL); Yakir Gorski, Givatayim (IL); Oren Dovrat, Tel Aviv-Jaffa (IL); Ron Dekel, Givatayim (IL); Emanuel Garbin, Tel Aviv-Jaffa (IL); and Sasha Smekhov, Kiryat Ono (IL)
Assigned to KLA Corporation, Milpitas, CA (US)
Filed by KLA Corporation, Milpitas, CA (US)
Filed on Mar. 29, 2023, as Appl. No. 18/128,140.
Claims priority of provisional application 63/443,050, filed on Feb. 3, 2023.
Claims priority of provisional application 63/326,268, filed on Mar. 31, 2022.
Prior Publication US 2023/0316500 A1, Oct. 5, 2023
Int. Cl. G06T 7/00 (2017.01)
CPC G06T 7/001 (2013.01) [G06T 2207/20076 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30148 (2013.01)] 27 Claims
OG exemplary drawing
 
1. An inspection system comprising:
one or more controllers including one or more processors to execute program instructions stored in a memory, the one or more program instructions configured to cause the one or more processors to:
receive one or more reference images;
receive one or more test images of a sample, the one or more test images including one or more defects of interest (DOIs);
generate one or more probabilistic context maps during inspection runtime using an unsupervised classifier;
provide the generated one or more probabilistic context maps to a supervised classifier during the inspection runtime; and
apply the supervised classifier to the received one or more test images to identify the one or more DOIs on the sample.