US 11,668,696 B2
Mask structure optimization device, mask structure optimization method, and program
Issei Sato, Wako (JP); Masahiro Kazama, Tokyo (JP); Masashi Ugawa, Tokyo (JP); Hiroaki Adachi, Tokyo (JP); and Fumiya Shimada, Tokyo (JP)
Assigned to RIKEN; and THINKCYTE, INC.
Filed by RIKEN, Saitama (JP); and ThinkCyte, Inc., Tokyo (JP)
Filed on Mar. 29, 2022, as Appl. No. 17/706,962.
Application 17/706,962 is a division of application No. 16/965,311, granted, now 11,314,994, previously published as PCT/JP2019/003120, filed on Jan. 30, 2019.
Claims priority of application No. 2018-014150 (JP), filed on Jan. 30, 2018.
Prior Publication US 2022/0222935 A1, Jul. 14, 2022
Int. Cl. G06V 10/88 (2022.01); G01N 15/14 (2006.01); G01N 33/483 (2006.01); G06V 10/50 (2022.01); G06V 20/69 (2022.01); G06F 18/2115 (2023.01); G06F 18/21 (2023.01); G06N 3/045 (2023.01)
CPC G06V 10/88 (2022.01) [G01N 15/1429 (2013.01); G01N 33/4833 (2013.01); G06F 18/2115 (2023.01); G06F 18/2163 (2023.01); G06N 3/045 (2023.01); G06V 10/507 (2022.01); G06V 20/695 (2022.01); G06V 20/698 (2022.01)] 8 Claims
OG exemplary drawing
 
7. A mask structure optimization method comprising:
a classification target image size acquiring step of acquiring a size of a classification target image which is an image including a classification target;
a mask size setting step of setting a size of a mask applied to the classification target image;
an initial value setting step of setting an initial value for a mask pattern of the mask;
a convolutional processing step of executing convolutional processing for the classification target image and an image of the mask; and
a mask pattern optimizing step of optimizing the mask pattern of the mask on the basis of results of the convolutional processing executed in the convolutional processing step,
wherein the mask is an optical mask having a plurality of regions with different optical characteristics.