US 11,694,078 B2
Electronic apparatus and controlling method thereof
Yongmin Tai, Suwon-si (KR); Insang Cho, Suwon-si (KR); and Chanyoung Hwang, Suwon-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Oct. 27, 2020, as Appl. No. 17/81,510.
Claims priority of application No. 10-2019-0150746 (KR), filed on Nov. 21, 2019.
Prior Publication US 2021/0158077 A1, May 27, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/62 (2022.01); G06N 3/04 (2006.01); G06N 3/08 (2006.01); G06T 3/40 (2006.01); G06F 18/2413 (2023.01); G06F 18/214 (2023.01); G06N 3/045 (2023.01); G06T 7/00 (2017.01)
CPC G06N 3/08 (2013.01) [G06F 18/214 (2023.01); G06F 18/2413 (2023.01); G06N 3/045 (2023.01); G06T 3/40 (2013.01); G06T 7/0002 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30168 (2013.01); G06T 2207/30201 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An electronic apparatus comprising: a memory that stores first information regarding a plurality of first artificial intelligence models trained to perform image processing differently from each other and second information regarding a second artificial intelligence model trained to identify a type of an image by predicting a processing result of the image by each of the plurality of first artificial intelligence models; and a processor configured to: identify a type of an input image by inputting the input image to the second artificial intelligence model stored in the memory, process the input image by inputting the input image to one of the plurality of first artificial intelligence models stored in the memory based on the identified type, and obtain a weighted value regarding a plurality of types related to the input image by inputting the input image to the second artificial intelligence model.
 
10. A method of controlling an electronic apparatus, the method comprising:
identifying a type of an input image by inputting the input image to a second artificial intelligence model trained to identify a type of an image by predicting an image processing result by each of a plurality of first artificial intelligence models trained to perform image processing differently from each other;
processing the input image by inputting the input image to one of the plurality of first artificial intelligence models based on the identified type; and
obtaining a weighted value regarding a plurality of types related to the input image by inputting the input image to the second artificial intelligence model.
 
18. A computer-implemented method of training a neural network for image processing comprising: collecting a set of digital sample images from a database; inputting the collected set of digital sample images into a plurality of first neural network models, so as to obtain a plurality of outputs, wherein the plurality of first neural network models are trained to perform image processing differently from each other; training, in a second neural network model, relationships between the set of digital sample images and a side effect type corresponding to each digital sample image using the obtained plurality of outputs, wherein the side effect type corresponds to a side effect of processing by a respective one of the plurality of first neural network models; identifying a side effect type of an input digital image by inputting the input digital image to the second neural network model; and processing the input digital image by inputting the input digital image to one of the plurality of first neural network models that corresponds to the identified defect type.