Low-Light Image Enhancement Based on Generative Adversarial Network
Image enhancement is considered to be one of the complex tasks in image processing. When the images are captured under dim light, the quality of the images degrades due to low visibility degenerating the vision-based algorithms’ performance that is built for very good quality images with better visi...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-11-01
|
| Series: | Frontiers in Genetics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fgene.2021.799777/full |
| _version_ | 1818405879327752192 |
|---|---|
| author | Nandhini Abirami R. Durai Raj Vincent P. M. |
| author_facet | Nandhini Abirami R. Durai Raj Vincent P. M. |
| author_sort | Nandhini Abirami R. |
| collection | DOAJ |
| description | Image enhancement is considered to be one of the complex tasks in image processing. When the images are captured under dim light, the quality of the images degrades due to low visibility degenerating the vision-based algorithms’ performance that is built for very good quality images with better visibility. After the emergence of a deep neural network number of methods has been put forward to improve images captured under low light. But, the results shown by existing low-light enhancement methods are not satisfactory because of the lack of effective network structures. A low-light image enhancement technique (LIMET) with a fine-tuned conditional generative adversarial network is presented in this paper. The proposed approach employs two discriminators to acquire a semantic meaning that imposes the obtained results to be realistic and natural. Finally, the proposed approach is evaluated with benchmark datasets. The experimental results highlight that the presented approach attains state-of-the-performance when compared to existing methods. The models’ performance is assessed using Visual Information Fidelitysse, which assesses the generated image’s quality over the degraded input. VIF obtained for different datasets using the proposed approach are 0.709123 for LIME dataset, 0.849982 for DICM dataset, 0.619342 for MEF dataset. |
| first_indexed | 2024-12-14T09:03:03Z |
| format | Article |
| id | doaj.art-9f150441369549baa33e0179b06f4e30 |
| institution | Directory Open Access Journal |
| issn | 1664-8021 |
| language | English |
| last_indexed | 2024-12-14T09:03:03Z |
| publishDate | 2021-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Genetics |
| spelling | doaj.art-9f150441369549baa33e0179b06f4e302022-12-21T23:08:46ZengFrontiers Media S.A.Frontiers in Genetics1664-80212021-11-011210.3389/fgene.2021.799777799777Low-Light Image Enhancement Based on Generative Adversarial NetworkNandhini Abirami R.Durai Raj Vincent P. M.Image enhancement is considered to be one of the complex tasks in image processing. When the images are captured under dim light, the quality of the images degrades due to low visibility degenerating the vision-based algorithms’ performance that is built for very good quality images with better visibility. After the emergence of a deep neural network number of methods has been put forward to improve images captured under low light. But, the results shown by existing low-light enhancement methods are not satisfactory because of the lack of effective network structures. A low-light image enhancement technique (LIMET) with a fine-tuned conditional generative adversarial network is presented in this paper. The proposed approach employs two discriminators to acquire a semantic meaning that imposes the obtained results to be realistic and natural. Finally, the proposed approach is evaluated with benchmark datasets. The experimental results highlight that the presented approach attains state-of-the-performance when compared to existing methods. The models’ performance is assessed using Visual Information Fidelitysse, which assesses the generated image’s quality over the degraded input. VIF obtained for different datasets using the proposed approach are 0.709123 for LIME dataset, 0.849982 for DICM dataset, 0.619342 for MEF dataset.https://www.frontiersin.org/articles/10.3389/fgene.2021.799777/fullcomputer visiondeep learningfacial expression recognitionconvolutional neural networkhuman-robot interactiongenerative adversarial network |
| spellingShingle | Nandhini Abirami R. Durai Raj Vincent P. M. Low-Light Image Enhancement Based on Generative Adversarial Network Frontiers in Genetics computer vision deep learning facial expression recognition convolutional neural network human-robot interaction generative adversarial network |
| title | Low-Light Image Enhancement Based on Generative Adversarial Network |
| title_full | Low-Light Image Enhancement Based on Generative Adversarial Network |
| title_fullStr | Low-Light Image Enhancement Based on Generative Adversarial Network |
| title_full_unstemmed | Low-Light Image Enhancement Based on Generative Adversarial Network |
| title_short | Low-Light Image Enhancement Based on Generative Adversarial Network |
| title_sort | low light image enhancement based on generative adversarial network |
| topic | computer vision deep learning facial expression recognition convolutional neural network human-robot interaction generative adversarial network |
| url | https://www.frontiersin.org/articles/10.3389/fgene.2021.799777/full |
| work_keys_str_mv | AT nandhiniabiramir lowlightimageenhancementbasedongenerativeadversarialnetwork AT durairajvincentpm lowlightimageenhancementbasedongenerativeadversarialnetwork |