An efficient framework for deep learning‐based light‐defect image enhancement
Abstract The enhancement of light‐defect images such as extremely low‐light, low‐light and dim‐light has always been a research hotspot. Most of the existing methods are excellent in specific illuminations, and there is much room for improvement in processing light‐defect images with different illum...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-05-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12125 |
| _version_ | 1798035586187001856 |
|---|---|
| author | Chengxu Ma Daihui Li Shangyou Zeng Junbo Zhao Hongyang Chen |
| author_facet | Chengxu Ma Daihui Li Shangyou Zeng Junbo Zhao Hongyang Chen |
| author_sort | Chengxu Ma |
| collection | DOAJ |
| description | Abstract The enhancement of light‐defect images such as extremely low‐light, low‐light and dim‐light has always been a research hotspot. Most of the existing methods are excellent in specific illuminations, and there is much room for improvement in processing light‐defect images with different illuminations. Therefore, this study proposes an efficient framework based on deep learning to enhance various light‐defect images. The proposed framework estimates the reflectance component and illumination component. Next, we propose a generator guided by an attention mechanism in the reflectance part to repair the light‐defect in the dark. In addition, we design a colour loss function for the problem of colour distortion in the enhanced images. Finally, the illumination map of the light‐defect images is adjusted adaptively. Extensive experiments are conducted to demonstrate that our method can not only deal with the images with different illuminations but also enhance the images with clearer details and richer colours. At the same time, we prove its superiority by comparing it with state‐of‐the‐art methods under both visual quality comparison and quantitative comparison of various datasets and real‐world images. |
| first_indexed | 2024-04-11T21:00:13Z |
| format | Article |
| id | doaj.art-70fb74cd0d914decb0069bad9ea3413c |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-04-11T21:00:13Z |
| publishDate | 2021-05-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-70fb74cd0d914decb0069bad9ea3413c2022-12-22T04:03:32ZengWileyIET Image Processing1751-96591751-96672021-05-011571553156610.1049/ipr2.12125An efficient framework for deep learning‐based light‐defect image enhancementChengxu Ma0Daihui Li1Shangyou Zeng2Junbo Zhao3Hongyang Chen4College of Electronic Engineering Guangxi Normal University Guilin Guangxi People's Republic of ChinaCollege of Electronic Engineering Guangxi Normal University Guilin Guangxi People's Republic of ChinaCollege of Electronic Engineering Guangxi Normal University Guilin Guangxi People's Republic of ChinaCollege of Electronic Engineering Guangxi Normal University Guilin Guangxi People's Republic of ChinaCollege of Electronic Engineering Guangxi Normal University Guilin Guangxi People's Republic of ChinaAbstract The enhancement of light‐defect images such as extremely low‐light, low‐light and dim‐light has always been a research hotspot. Most of the existing methods are excellent in specific illuminations, and there is much room for improvement in processing light‐defect images with different illuminations. Therefore, this study proposes an efficient framework based on deep learning to enhance various light‐defect images. The proposed framework estimates the reflectance component and illumination component. Next, we propose a generator guided by an attention mechanism in the reflectance part to repair the light‐defect in the dark. In addition, we design a colour loss function for the problem of colour distortion in the enhanced images. Finally, the illumination map of the light‐defect images is adjusted adaptively. Extensive experiments are conducted to demonstrate that our method can not only deal with the images with different illuminations but also enhance the images with clearer details and richer colours. At the same time, we prove its superiority by comparing it with state‐of‐the‐art methods under both visual quality comparison and quantitative comparison of various datasets and real‐world images.https://doi.org/10.1049/ipr2.12125Optical, image and video signal processingComputer vision and image processing techniquesNeural nets |
| spellingShingle | Chengxu Ma Daihui Li Shangyou Zeng Junbo Zhao Hongyang Chen An efficient framework for deep learning‐based light‐defect image enhancement IET Image Processing Optical, image and video signal processing Computer vision and image processing techniques Neural nets |
| title | An efficient framework for deep learning‐based light‐defect image enhancement |
| title_full | An efficient framework for deep learning‐based light‐defect image enhancement |
| title_fullStr | An efficient framework for deep learning‐based light‐defect image enhancement |
| title_full_unstemmed | An efficient framework for deep learning‐based light‐defect image enhancement |
| title_short | An efficient framework for deep learning‐based light‐defect image enhancement |
| title_sort | efficient framework for deep learning based light defect image enhancement |
| topic | Optical, image and video signal processing Computer vision and image processing techniques Neural nets |
| url | https://doi.org/10.1049/ipr2.12125 |
| work_keys_str_mv | AT chengxuma anefficientframeworkfordeeplearningbasedlightdefectimageenhancement AT daihuili anefficientframeworkfordeeplearningbasedlightdefectimageenhancement AT shangyouzeng anefficientframeworkfordeeplearningbasedlightdefectimageenhancement AT junbozhao anefficientframeworkfordeeplearningbasedlightdefectimageenhancement AT hongyangchen anefficientframeworkfordeeplearningbasedlightdefectimageenhancement AT chengxuma efficientframeworkfordeeplearningbasedlightdefectimageenhancement AT daihuili efficientframeworkfordeeplearningbasedlightdefectimageenhancement AT shangyouzeng efficientframeworkfordeeplearningbasedlightdefectimageenhancement AT junbozhao efficientframeworkfordeeplearningbasedlightdefectimageenhancement AT hongyangchen efficientframeworkfordeeplearningbasedlightdefectimageenhancement |