US 12,293,505 B2
Evaluation apparatus, computer-readable storage medium, evaluation method, forming system, and article manufacturing method
Fuma Kizu, Tochigi (JP); Koji Ishibashi, Tokyo (JP); and Sentaro Aihara, Tochigi (JP)
Assigned to CANON KABUSHIKI KAISHA, Tokyo (JP)
Filed by CANON KABUSHIKI KAISHA, Tokyo (JP)
Filed on Mar. 24, 2022, as Appl. No. 17/703,460.
Claims priority of application No. 2021-055838 (JP), filed on Mar. 29, 2021.
Prior Publication US 2022/0309647 A1, Sep. 29, 2022
Int. Cl. G06T 7/00 (2017.01); G03F 7/00 (2006.01); G03F 9/00 (2006.01); G06N 20/00 (2019.01)
CPC G06T 7/0004 (2013.01) [G03F 7/0002 (2013.01); G03F 7/7065 (2013.01); G03F 9/7042 (2013.01); G06N 20/00 (2019.01); G06T 2207/20081 (2013.01); G06T 2207/30148 (2013.01)] 15 Claims
OG exemplary drawing
 
1. An evaluation apparatus that evaluates a composition formed on a substrate by forming processing,
wherein the forming processing is processing of forming the composition by curing a formable material in a state in which the formable material on the substrate is in contact with a mold,
the apparatus comprises:
an obtaining device configured to obtain an image including the composition formed on the substrate by the forming processing;
a processing device configured to process the obtained image for the evaluation by using an inference model; and
a machine learning device configured to generate the inference model by machine learning, and
the processing device is configured to:
output a feature of each of one or more abnormalities in the obtained image according to the inference model,
obtain information regarding a formation region on the substrate where the composition has been formed,
determine the kind of each of the abnormalities based on the output feature of each of the abnormalities and a relationship between the information regarding the formation region and a position and a size of the abnormality, and
determine whether the image is a normal image or an image including an abnormality based on a result of the determination of the kind of the abnormalities,
wherein the machine learning device performs machine learning while using, as an input to the inference model, the image including the composition formed on the substrate by the forming processing and, as supervised data, a relationship between a plurality of images obtained in advance and feature information of respective abnormalities in the plurality of images.