US 12,002,249 B1
Deep learning-based overlay key centering system and method thereof
Soo-Yeon Mo, Hwaseong-si (KR); Ga-Min Kim, Hwaseong-si (KR); and Hyo-Sik Ham, Hwaseong-si (KR)
Assigned to AUROS TECHNOLOGY, INC., Hwaseong-si (KR)
Filed by AUROS TECHNOLOGY, INC., Hwaseong-si (KR)
Filed on May 3, 2023, as Appl. No. 18/142,903.
Claims priority of application No. 10-2022-0151232 (KR), filed on Nov. 14, 2022.
Int. Cl. G06V 10/24 (2022.01); G06V 10/25 (2022.01); G06V 10/60 (2022.01); H04N 5/265 (2006.01)
CPC G06V 10/24 (2022.01) [G06V 10/25 (2022.01); G06V 10/60 (2022.01); H04N 5/265 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A deep learning-based overlay centering system comprising:
a server configured to collect an input data set, which comprises measurement image data of an overlay key with label data including information on a location and bounding box size of the overlay key, from at least one overlay measurement device and train a model by inputting the input data set to a model for deep learning,
wherein the server calculates a loss function by comparing result data predicted by the model with the label data, and optimizes an algorithm of the model by modifying a weight of the model so that a loss value calculated with the loss function may become smaller than a reference value, to train the model, and
wherein the label data comprises at least one of a type of the overlay key to be trained, a value obtained by dividing a x-coordinate value among center coordinates (x, y) of the overlay key to width value of measured image of the overlay key, a value obtained by dividing a y-coordinate value among center coordinates (x, y) of the overlay key to height value of measured image of the overlay key, a value obtained by dividing width of the bounding box of the overlay key to width value of measured image of the overlay key, and a value obtained by dividing height of the bounding box of the overlay key to height value of measured image of the overlay key.