A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images
Abstract RGGB sensor arrays are commonly used in digital cameras and mobile photography. However, images of extreme dark-light conditions often suffer from insufficient exposure because the sensor receives insufficient light. The existing methods mainly employ U-Net variants, multi-stage camera para...
Saved in:
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2025-01-01
|
Series: | Scientific Reports |
Subjects: | |
Online Access: | https://doi.org/10.1038/s41598-025-87412-x |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832585734902513664 |
---|---|
author | Yiyao Huang Xiaobao Zhu Fenglian Yuan Jing Shi U. Kintak Jingfei Fu Yiran Peng Chenheng Deng |
author_facet | Yiyao Huang Xiaobao Zhu Fenglian Yuan Jing Shi U. Kintak Jingfei Fu Yiran Peng Chenheng Deng |
author_sort | Yiyao Huang |
collection | DOAJ |
description | Abstract RGGB sensor arrays are commonly used in digital cameras and mobile photography. However, images of extreme dark-light conditions often suffer from insufficient exposure because the sensor receives insufficient light. The existing methods mainly employ U-Net variants, multi-stage camera parameter simulation, or image parameter processing to address this issue. However, those methods usually apply color adjustments evenly across the entire image, which may cause extensive blue or green noise artifacts, especially in images with dark backgrounds. This study attacks the problem by proposing a novel multi-step process for image enhancement. The pipeline starts with a self-attention U-Net for initial color restoration and applies a Color Correction Matrix (CCM). Thereafter, High Dynamic Range (HDR) image reconstruction techniques are utilized to improve exposure using various Camera Response Functions (CRFs). After removing under- and over-exposed frames, pseudo-HDR images are created through multi-frame fusion. Also, a comparative analysis is conducted based on a standard dataset, and the results show that the proposed approach performs better in creating well-exposed images and improves the Peak-Signal-to-Noise Ratio (PSNR) by 0.16 dB compared to the benchmark methods. |
format | Article |
id | doaj-art-ba4b556c5d194d2d88ef11096cad4836 |
institution | Kabale University |
issn | 2045-2322 |
language | English |
publishDate | 2025-01-01 |
publisher | Nature Portfolio |
record_format | Article |
series | Scientific Reports |
spelling | doaj-art-ba4b556c5d194d2d88ef11096cad48362025-01-26T12:33:22ZengNature PortfolioScientific Reports2045-23222025-01-0115111710.1038/s41598-025-87412-xA two-stage HDR reconstruction pipeline for extreme dark-light RGGB imagesYiyao Huang0Xiaobao Zhu1Fenglian Yuan2Jing Shi3U. Kintak4Jingfei Fu5Yiran Peng6Chenheng Deng7Macau University of Science and Technology, Faculty of Innovation EngineeringNanchang Hangkong University, School of Information EngineeringNanchang Hangkong University, School of Information EngineeringDepartment of Mechanical and Materials Engineering, University of CincinnatiMacau University of Science and Technology, Faculty of Innovation EngineeringNanchang Hangkong University, School of Information EngineeringMacau University of Science and Technology, Faculty of Innovation EngineeringMacau University of Science and Technology, Faculty of Innovation EngineeringAbstract RGGB sensor arrays are commonly used in digital cameras and mobile photography. However, images of extreme dark-light conditions often suffer from insufficient exposure because the sensor receives insufficient light. The existing methods mainly employ U-Net variants, multi-stage camera parameter simulation, or image parameter processing to address this issue. However, those methods usually apply color adjustments evenly across the entire image, which may cause extensive blue or green noise artifacts, especially in images with dark backgrounds. This study attacks the problem by proposing a novel multi-step process for image enhancement. The pipeline starts with a self-attention U-Net for initial color restoration and applies a Color Correction Matrix (CCM). Thereafter, High Dynamic Range (HDR) image reconstruction techniques are utilized to improve exposure using various Camera Response Functions (CRFs). After removing under- and over-exposed frames, pseudo-HDR images are created through multi-frame fusion. Also, a comparative analysis is conducted based on a standard dataset, and the results show that the proposed approach performs better in creating well-exposed images and improves the Peak-Signal-to-Noise Ratio (PSNR) by 0.16 dB compared to the benchmark methods.https://doi.org/10.1038/s41598-025-87412-xExtremely dark-lightImage enhancementSelf-attention U-NetHDR reconstruction pipeline |
spellingShingle | Yiyao Huang Xiaobao Zhu Fenglian Yuan Jing Shi U. Kintak Jingfei Fu Yiran Peng Chenheng Deng A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images Scientific Reports Extremely dark-light Image enhancement Self-attention U-Net HDR reconstruction pipeline |
title | A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images |
title_full | A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images |
title_fullStr | A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images |
title_full_unstemmed | A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images |
title_short | A two-stage HDR reconstruction pipeline for extreme dark-light RGGB images |
title_sort | two stage hdr reconstruction pipeline for extreme dark light rggb images |
topic | Extremely dark-light Image enhancement Self-attention U-Net HDR reconstruction pipeline |
url | https://doi.org/10.1038/s41598-025-87412-x |
work_keys_str_mv | AT yiyaohuang atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT xiaobaozhu atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT fenglianyuan atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT jingshi atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT ukintak atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT jingfeifu atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT yiranpeng atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT chenhengdeng atwostagehdrreconstructionpipelineforextremedarklightrggbimages AT yiyaohuang twostagehdrreconstructionpipelineforextremedarklightrggbimages AT xiaobaozhu twostagehdrreconstructionpipelineforextremedarklightrggbimages AT fenglianyuan twostagehdrreconstructionpipelineforextremedarklightrggbimages AT jingshi twostagehdrreconstructionpipelineforextremedarklightrggbimages AT ukintak twostagehdrreconstructionpipelineforextremedarklightrggbimages AT jingfeifu twostagehdrreconstructionpipelineforextremedarklightrggbimages AT yiranpeng twostagehdrreconstructionpipelineforextremedarklightrggbimages AT chenhengdeng twostagehdrreconstructionpipelineforextremedarklightrggbimages |