US 12,236,666 B2
Image processing circuit, system-on-chip including the same, and method of enhancing image quality
Geunhee Yang, Hwaseong-si (KR); Donghoon Kim, Seongnam-si (KR); Maiyuran Wijay, San Diego, CA (US); Oleg Sergeyevich Khoruzhiy, San Diego, CA (US); Rama Mythili Vadali, Vista, CA (US); Song Ni, San Diego, CA (US); Dusic Yoo, Seoul (KR); Jaewon Yu, Yongin-si (KR); Seunghyung Lee, Yongin-si (KR); and Venkata Rama Krishna Meka, Rajahmundry (IN)
Assigned to Samsung Electronics Co., Ltd., Gyeonggi-do (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed on Mar. 15, 2022, as Appl. No. 17/695,142.
Claims priority of provisional application 63/177,027, filed on Apr. 20, 2021.
Claims priority of application No. 10-2021-0108957 (KR), filed on Aug. 18, 2021.
Prior Publication US 2022/0335709 A1, Oct. 20, 2022
Int. Cl. G06V 10/764 (2022.01); G06N 3/08 (2023.01); G06T 3/40 (2024.01); G06T 5/70 (2024.01); G06T 5/73 (2024.01); G06V 10/26 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06V 20/00 (2022.01)
CPC G06V 10/764 (2022.01) [G06N 3/08 (2013.01); G06T 3/40 (2013.01); G06T 5/70 (2024.01); G06T 5/73 (2024.01); G06V 10/267 (2022.01); G06V 10/774 (2022.01); G06V 20/00 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06V 10/82 (2022.01)] 20 Claims
OG exemplary drawing
 
1. An image processing circuit comprising:
a tuning circuit configured to:
receive a segmentation map comprising pixel-by-pixel class inference information of a first image and a confidence map comprising confidence of the class inference information,
determine classes of respective pixels of the first image, correction effects for each pixel of the first image, and correction values indicating intensity of the correction effects based on the segmentation map and the confidence map, and
generate a correction map based on the classes and the correction values of the respective pixels; and
at least one correcting circuit configured to generate an enhanced image by applying correction effects according to the correction values to the respective pixels of the first image based on the correction map, wherein
the segmentation map is generated by enlarging a low-resolution segmentation map,
the confidence map is generated by enlarging a low-resolution confidence map, and
each of the low-resolution segmentation map and the low-resolution confidence map is generated using a lower resolution version of the first image.