IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement
Inadequate illumination often causes severe image degradation, such as noise and artifacts. These types of images do not meet the requirements of advanced visual tasks, so low-light image enhancement is currently a flourishing and challenging research topic. To alleviate the problem of low brightnes...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/14/3162 |
| _version_ | 1797589517703577600 |
|---|---|
| author | Chao Xie Hao Tang Linfeng Fei Hongyu Zhu Yaocong Hu |
| author_facet | Chao Xie Hao Tang Linfeng Fei Hongyu Zhu Yaocong Hu |
| author_sort | Chao Xie |
| collection | DOAJ |
| description | Inadequate illumination often causes severe image degradation, such as noise and artifacts. These types of images do not meet the requirements of advanced visual tasks, so low-light image enhancement is currently a flourishing and challenging research topic. To alleviate the problem of low brightness and low contrast, this paper proposes an improved zero-shot Retinex network, named IRNet, which is composed of two parts: a Decom-Net and an Enhance-Net. The Decom-Net is designed to decompose the raw input into two maps, i.e., illuminance and reflection. Afterwards, the subsequent Enhance-Net takes the decomposed illuminance component as its input, enhances the image brightness and features through gamma transformation and a convolutional network, and fuses the enhanced illumination and reflection maps together to obtain the final enhanced result. Due to the use of zero-shot learning, no previous training is required. IRNet depends on the internal optimization of each individual input image, and the network weights are updated by iteratively minimizing a series of designed loss functions, in which noise reduction loss and color constancy loss are introduced to reduce noise and relieve color distortion during the image enhancement process. Experiments conducted on public datasets and the presented practical applications demonstrate that our method outperforms other counterparts in terms of both visual perception and objective metrics. |
| first_indexed | 2024-03-11T01:07:36Z |
| format | Article |
| id | doaj.art-4f731553c44144fd8c396118e3e2dcf1 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T01:07:36Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-4f731553c44144fd8c396118e3e2dcf12023-11-18T19:06:46ZengMDPI AGElectronics2079-92922023-07-011214316210.3390/electronics12143162IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image EnhancementChao Xie0Hao Tang1Linfeng Fei2Hongyu Zhu3Yaocong Hu4College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, ChinaCollege of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, ChinaCollege of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, ChinaCollege of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, ChinaSchool of Electrical Engineering, Anhui Polytechnic University, Wuhu 241000, ChinaInadequate illumination often causes severe image degradation, such as noise and artifacts. These types of images do not meet the requirements of advanced visual tasks, so low-light image enhancement is currently a flourishing and challenging research topic. To alleviate the problem of low brightness and low contrast, this paper proposes an improved zero-shot Retinex network, named IRNet, which is composed of two parts: a Decom-Net and an Enhance-Net. The Decom-Net is designed to decompose the raw input into two maps, i.e., illuminance and reflection. Afterwards, the subsequent Enhance-Net takes the decomposed illuminance component as its input, enhances the image brightness and features through gamma transformation and a convolutional network, and fuses the enhanced illumination and reflection maps together to obtain the final enhanced result. Due to the use of zero-shot learning, no previous training is required. IRNet depends on the internal optimization of each individual input image, and the network weights are updated by iteratively minimizing a series of designed loss functions, in which noise reduction loss and color constancy loss are introduced to reduce noise and relieve color distortion during the image enhancement process. Experiments conducted on public datasets and the presented practical applications demonstrate that our method outperforms other counterparts in terms of both visual perception and objective metrics.https://www.mdpi.com/2079-9292/12/14/3162low-light image enhancementzero-shot learningRetinex theoryimage features |
| spellingShingle | Chao Xie Hao Tang Linfeng Fei Hongyu Zhu Yaocong Hu IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement Electronics low-light image enhancement zero-shot learning Retinex theory image features |
| title | IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement |
| title_full | IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement |
| title_fullStr | IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement |
| title_full_unstemmed | IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement |
| title_short | IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement |
| title_sort | irnet an improved zero shot retinex network for low light image enhancement |
| topic | low-light image enhancement zero-shot learning Retinex theory image features |
| url | https://www.mdpi.com/2079-9292/12/14/3162 |
| work_keys_str_mv | AT chaoxie irnetanimprovedzeroshotretinexnetworkforlowlightimageenhancement AT haotang irnetanimprovedzeroshotretinexnetworkforlowlightimageenhancement AT linfengfei irnetanimprovedzeroshotretinexnetworkforlowlightimageenhancement AT hongyuzhu irnetanimprovedzeroshotretinexnetworkforlowlightimageenhancement AT yaoconghu irnetanimprovedzeroshotretinexnetworkforlowlightimageenhancement |