公开(公告)号：US20240257314A1

公开(公告)日：2024-08-01

申请号：US18538127

申请日：2023-12-13

Applicant: Samsung Electronics Co., Ltd.

Inventor： Jinwoo KANG ,  Dongbum CHOI

IPC: G06T5/50 ,  G06T5/70 ,  G06T7/20

CPC classification number: G06T5/50 ,  G06T5/70 ,  G06T7/20 ,  G06T2207/20221

Abstract: In a method of processing an image, a first composite frame image corresponding to an n-th input frame image is generated by performing a first temporal noise reduction operation based on the n-th input frame image and an (n−m)-th input frame image. The (n−m)-th input frame image is received prior to the n-th input frame image. A second composite frame image corresponding to the n-th input frame image is generated by performing a second temporal noise reduction operation based on the n-th input frame image and an (n+k)-th input frame image. The (n+k)-th input frame image is received subsequent to the n-th input frame image. The second composite frame image is provided as an n-th output frame image corresponding to the n-th input frame image.

公开(公告)号：US20190327393A1

公开(公告)日：2019-10-24

申请号：US16386855

申请日：2019-04-17

Applicant: Samsung Electronics Co., Ltd.

Inventor： Kyungdong YANG ,  Jaehyoung PARK ,  Yungmok YU ,  Dongbum CHOI ,  Jonghun WON

IPC: H04N5/222 ,  G06T7/593 ,  H04N5/225

Abstract: An electronic device is provided. The electronic device includes a first camera, a second camera spaced apart from the first camera, and a processor. The processor is configured to obtain a first image of external objects using the first camera, obtain a second image of the external objects using the second camera, identify a specified object in which pieces of depth information are generated from among the external objects included in the first image and the second image based on phase difference comparison between the first image and the second image, then select depth information about the specified object among the pieces of depth information based on a degree of spreading of a point corresponding to the specified object included in at least one of the first image and the second image, and generate depth information about the external objects including the specified object using the selected depth information.