US 12,079,310 B2
Defect classification apparatus, method and program
Ryusuke Hirai, Tokyo (JP); Saori Asaka, Yokohama Kanagawa (JP); Yukinobu Sakata, Kawasaki Kanagawa (JP); and Akiyuki Tanizawa, Kawasaki Kanagawa (JP)
Assigned to KABUSHIKI KAISHA TOSHIBA, Tokyo (JP)
Filed by KABUSHIKI KAISHA TOSHIBA, Tokyo (JP)
Filed on Aug. 31, 2021, as Appl. No. 17/463,162.
Claims priority of application No. 2021-039436 (JP), filed on Mar. 11, 2021.
Prior Publication US 2022/0292317 A1, Sep. 15, 2022
Int. Cl. G06K 9/62 (2022.01); G06F 18/241 (2023.01); G06T 7/00 (2017.01); G06V 10/75 (2022.01)
CPC G06F 18/241 (2023.01) [G06T 7/001 (2013.01); G06V 10/751 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20224 (2013.01); G06T 2207/30148 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A defect classification apparatus comprising a processor configured to:
acquire a first design image which is an image created by design software based on design data and which relates to a first inspection target;
acquire a first real image which is an image captured by imaging the first inspection target as actually produced based on the design data;
convert the first design image to a reference image by using a second real image, the reference image being a generated image representing an inspection target having no defect, and the second real image being an image captured by imaging at least a part of a second inspection target which is an actually produced inspection target having no defect;
calculate a reliability of the reference image based on a similarity between design images of the first inspection target and the second inspection target;
detect a defect in the first inspection target by comparing the reference image and the first real image; and
classify the defect, based on the calculated reliability of the reference image.