IP Library Granted Patent US 12,731,217
Granted Patent B2
US 12,731,217 · App. 18/417,246 · Granted Sep 8, 2026

Image processor and image processing system including the same

Inventors: Giyoung Heo (Suwon-si, KR); Daeho Lee (Suwon-si, KR); Minseok Lee (Suwon-si, KR)
Assignee: Samsung Electronics Co., Ltd.
G06T5/50G06T5/70G06T2207/10144G06T2207/20208
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,731,217
App. No.
18/417,246
Granted
Sep 8, 2026
Kind
B2
Abstract

An image processor including an interface circuit configured to receive a plurality of images corresponding to a plurality of exposure times from an external sensor, and a high dynamic range (HDR) synthesis circuit configured to synthesize a first HDR image, based on a first image and a second image among the plurality of images, synthesize a second HDR image, based on a third image among the plurality of images and the first HDR image, apply a first weight including weight values respectively corresponding to pixels of the first HDR image to the first HDR image, and apply a second weight including weight values respectively corresponding to pixels of the third image to the third image to synthesize the second HDR image.

Claims (60)

1 . An image processor comprising:

an interface circuit configured to receive a plurality of images corresponding to a plurality of exposure times from an external sensor, each of the plurality of exposure times having different lengths; and

a high dynamic range (HDR) synthesis circuit configured to

synthesize a first HDR image, based on a first image and a second image among the plurality of images,

synthesize a second HDR image, based on a third image among the plurality of images and the first HDR image,

apply a first weight including weight values respectively corresponding to pixels of the first HDR image to the first HDR image to synthesize the second HDR image, wherein the first weight comprises the weight values based on a synthesis ratio of the first image and the second image when the first HDR image is synthesized, and

apply a second weight including weight values respectively corresponding to pixels of the third image to the third image to synthesize the second HDR image.

2 . The image processor of claim 1 , wherein the HDR synthesis circuit comprises:

a mixer configured to synthesize the first HDR image or the second HDR image; and

a direct memory access (DMA) configured to store the first weight.

3 . The image processor of claim 1 , wherein

the plurality of exposure times comprise a first exposure time, a second exposure time shorter than the first exposure time, and a third exposure time shorter than the second exposure time,

the first image is an image corresponding to the first exposure time,

the second image is an image corresponding to the second exposure time, and

the third image is an image corresponding to the third exposure time.

4 . The image processor of claim 2 , wherein the mixer is configured to generate the first weight including weight values based on the synthesis ratio of the first image and the second image when synthesizing the first HDR image, and transmit the first weight to the DMA.

5 . The image processor of claim 1 , further comprising a post-processing circuit configured to perform a post-processing operation for noise removal, based on the second HDR image and a third weight including weight values corresponding to pixels of the second HDR image,

wherein

the HDR synthesis circuit generates a weight map, and

the weight map comprises the first weight, the second weight and the third weight.

6 . The image processor of claim 5 , wherein the weight map comprises a weight based on at least one of noise variance and motion for each pixel of the plurality of images.

7 . The image processor of claim 5 , wherein the post-processing circuit comprises a noise reduction circuit.

8 . The image processor of claim 5 , wherein the HDR synthesis circuit comprises:

a mixer configured to synthesize the first HDR image or the second HDR image;

a write direct memory access (WDMA) configured to receive the weight map from the mixer and store the received weight map; and

a read direct memory access (RDMA) configured to receive the stored weight map from the WDMA and transmit the weight map received from the WDMA to the mixer.

9 . The image processor of claim 5 , wherein the third weight comprises weight values based on a synthesis ratio of the third image and the first HDR image when the second HDR image is synthesized.

10 . An operating method of an image processor including a high dynamic range (HDR) synthesis circuit, the method comprising:

obtaining a plurality of images corresponding to a plurality of exposure times;

synthesizing a first HDR image, based on a first image and a second image each having different exposure times among the plurality of images; and

applying a first weight including weight values respectively corresponding to pixels of the first HDR image to the first HDR image, and applying a second weight including weight values respectively corresponding to pixels of a third image to the third image, to synthesize a second HDR image, wherein the first weight comprises the weight values based on a synthesis ratio of the first image and the second image when the first HDR image is synthesized,

the plurality of images including the third image,

an exposure time corresponding to the third image being shorter than an exposure time corresponding to the first image and an exposure time corresponding to the second image.

11 . The method of claim 10 , further comprising: as an operating method of an image processor further comprising a post-processing circuit,

obtaining, by the post-processing circuit, the second HDR image and a third weight from the HDR synthesis circuit;

performing, by the post-processing circuit, a post-processing operation for removing noise, based on the second HDR image and the third weight,

wherein the synthesizing of the second HDR image further comprises generating the third weight based on a synthesis ratio of the first HDR image and the third image when the second HDR image is synthesized.

12 . The method of claim 11 , wherein the third weight comprises weight values based on the synthesis ratio of the third image and the first HDR image when the second HDR image is synthesized.

13 . The method of claim 11 , wherein the post-processing circuit comprises a noise reduction circuit.

14 . An image processing system comprising:

an image sensor configured to generate a plurality of images corresponding to a plurality of exposure times, each of the plurality of exposure times having different lengths; and

a processor configured to

receive the plurality of images from the image sensor,

synthesize a first high dynamic range (HDR) image, based on selected ones of the plurality of images having different exposure times,

synthesize a second HDR image, based on the first HDR image and any one of the plurality of images having a different exposure time from the selected ones of the plurality of images,

apply a first weight including weight values respectively corresponding to pixels of the first HDR image to the first HDR image to synthesize the second HDR image, wherein the first weight comprises the weight values based on a synthesis ratio of the first image and the second image when the first HDR image is synthesized, and

apply a second weight including weight values respectively corresponding to pixels of the any one of the plurality of images to the any one of the plurality of images to synthesize the second HDR image.

15 . The image processing system of claim 14 , wherein

the processor comprises:

a mixer configured to synthesize the first HDR image or the second HDR image and generate a weight map;

a direct memory access (DMA) configured to store the first weight; and

a post-processing circuit configured to perform a post-processing operation for removing noise, based on the second HDR image and a third weight including weight values corresponding to pixels of the second HDR image, and

the weight map comprises the first weight, the second weight and the third weight.

16 . The image processing system of claim 15 , wherein the weight map is a weight based on at least one of noise variance and motion for each pixel of the plurality of images.

17 . The image processing system of claim 15 , wherein the third weight comprises weight values based on a synthesis ratio between one of the plurality of images and the first HDR image when a second HDR image is synthesized.

18 . The image processing system of claim 14 , wherein

the plurality of exposure times comprise a first exposure time, a second exposure time shorter than the first exposure time, and a third exposure time shorter than the second exposure time,

the plurality of images comprise a first image corresponding to the first exposure time, a second image corresponding to the second exposure time, and a third image corresponding to the third exposure time,

the first HDR image is generated based on the first image and the second image, and

the any one image among the plurality of images is the third image.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 23, 2024
From: HEO, GIYOUNG; LEE, DAEHO; LEE, MINSEOK
To: SAMSUNG ELECTRONICS CO., LTD.
Reel/Frame 066206/0225 →
Priority Claims (1)
KR 10-2023-0013192 · Jan 31, 2023 · national
Continuity (1)
Related Publication 20240257310A1 · Aug 1, 2024
References Cited (13)
US 7612804B1 · Marcu · 2009 [cited by examiner]
US 8970727B2 · Shinmei · 2015 [cited by examiner]
US 9338349B2 · Sharma et al. · 2016 [cited by applicant]
US 9438827B2 · Li · 2016 [cited by applicant]
US 9628721B2 · Moon · 2017 [cited by examiner]
US 9973709B2 · You et al. · 2018 [cited by applicant]
US 10255888B2 · Narasimha et al. · 2019 [cited by applicant]
US 10638052B2 · Dhiman et al. · 2020 [cited by applicant]
US 11082625B2 · Kang et al. · 2021 [cited by applicant]
US 11290655B1 · Kim · 2022 [cited by examiner]
US 11379997B2 · Lee et al. · 2022 [cited by applicant]
US 20230206399A1 · Dong · 2023 [cited by examiner]
KR 101698876B1 · 2017 [cited by applicant]