公开(公告)号：US12229921B2

公开(公告)日：2025-02-18

申请号：US17554827

申请日：2021-12-17

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Quockhanh Dinh ,  Kyonghwan Jin ,  Kwangpyo Choi

IPC: G06T5/50 ,  G06N3/04 ,  G06T5/20 ,  G06V10/74 ,  G06V10/77 ,  G06V10/82 ,  H04N19/85

Abstract: An image processing apparatus, including a memory configured to store one or more instructions; and a processor configured to execute the one or more instructions stored in the memory to cause the image processing apparatus to: identify, in a previous frame, a prediction sample corresponding to a current sample of a current frame, generate a prediction frame for the current frame by changing a sample value of a collocated sample of the previous frame according to a sample value of the prediction sample, derive a weight by comparing a sample value of the current sample with the sample value of the prediction sample, apply the weight to a collocated sample of the prediction frame to obtain a weighted prediction frame, and obtain a current output frame by processing the current frame and the weighted prediction frame through a neural network comprising a convolution layer.

公开(公告)号：US12096131B2

公开(公告)日：2024-09-17

申请号：US17512338

申请日：2021-10-27

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Quockhanh Dinh ,  Kyonghwan Jin ,  Youngo Park ,  Kwangpyo Choi

IPC: G06T7/90 ,  G06F18/214 ,  G06F18/40 ,  G06N20/00 ,  G06T3/4015 ,  G06T3/4038 ,  G06T5/20 ,  G06T5/70 ,  H04N23/81 ,  H04N23/90

CPC classification number: H04N23/81 ,  G06F18/2148 ,  G06F18/40 ,  G06N20/00 ,  G06T3/4015 ,  G06T3/4038 ,  G06T5/20 ,  G06T5/70 ,  G06T7/90 ,  H04N23/90 ,  G06T2207/10016 ,  G06T2207/10024 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20182

Abstract: A method includes obtaining, by a second processor of the electronic device, from a first processor of the electronic device, a control signal for obtaining image data, loading any one learning model of at least one learning model into a memory, obtaining, by using the camera module, raw image data of the object, from light reflected from the object, the raw image data being configured to have a specified color array consisting of a plurality of colors with respect to a plurality of pixels, obtaining, by using the loaded any one learning model, a color data set with respect to the plurality of pixels from the obtained raw image data, the color data set including a plurality of pieces of color data classified according to the plurality of colors, and obtaining the noise-reduced image data of the object by using the obtained color data set.