Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network
With the wave of artificial intelligence and deep learning sweeping the world, there are many algorithms based on deep learning for image defog research. However, there is still serious color distortion, contrast reduction, incomplete fog removal, and other problems. To solve these problems, this pa...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/10/2186 |
| _version_ | 1797600319317737472 |
|---|---|
| author | Junkai Zhang Xiaoming Sun Yan Chen Yan Duan Yongliang Wang |
| author_facet | Junkai Zhang Xiaoming Sun Yan Chen Yan Duan Yongliang Wang |
| author_sort | Junkai Zhang |
| collection | DOAJ |
| description | With the wave of artificial intelligence and deep learning sweeping the world, there are many algorithms based on deep learning for image defog research. However, there is still serious color distortion, contrast reduction, incomplete fog removal, and other problems. To solve these problems, this paper proposes an improved image defogging network based on the traditional cycle-consistent adversarial network. We add the self-attention module and atrous convolution multi-scale feature fusion module on the basis of the traditional CycleGAN network to enhance the feature extraction capability of the network. The perceptual loss function is introduced into the loss function of the model to enhance the texture sense of the generated image. Finally, by comparing several typical defogging algorithms, the superiority of the defogging model proposed in this paper is proved qualitatively and quantitatively. Among them, on the indoor synthetic data set, the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measurement (SSIM) of the network designed by us can reach 23.22 and 0.8809, respectively. On the outdoor synthetic data set, the PSNR and SSIM of our designed network can be as high as 25.72 and 0.8859, respectively. On the real data set, the PSNR and SSIM of our designed network can reach 21.02 and 0.8166, respectively. It is proved that the defogging network in this paper has good practicability and universality. |
| first_indexed | 2024-03-11T03:47:48Z |
| format | Article |
| id | doaj.art-7b053dbc15dc4cfb85e4f5f0992a6ceb |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T03:47:48Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-7b053dbc15dc4cfb85e4f5f0992a6ceb2023-11-18T01:08:56ZengMDPI AGElectronics2079-92922023-05-011210218610.3390/electronics12102186Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial NetworkJunkai Zhang0Xiaoming Sun1Yan Chen2Yan Duan3Yongliang Wang4School of Measurement-Control Technology and Communications Engineering, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Measurement-Control Technology and Communications Engineering, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Measurement-Control Technology and Communications Engineering, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Measurement-Control Technology and Communications Engineering, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Measurement-Control Technology and Communications Engineering, Harbin University of Science and Technology, Harbin 150080, ChinaWith the wave of artificial intelligence and deep learning sweeping the world, there are many algorithms based on deep learning for image defog research. However, there is still serious color distortion, contrast reduction, incomplete fog removal, and other problems. To solve these problems, this paper proposes an improved image defogging network based on the traditional cycle-consistent adversarial network. We add the self-attention module and atrous convolution multi-scale feature fusion module on the basis of the traditional CycleGAN network to enhance the feature extraction capability of the network. The perceptual loss function is introduced into the loss function of the model to enhance the texture sense of the generated image. Finally, by comparing several typical defogging algorithms, the superiority of the defogging model proposed in this paper is proved qualitatively and quantitatively. Among them, on the indoor synthetic data set, the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measurement (SSIM) of the network designed by us can reach 23.22 and 0.8809, respectively. On the outdoor synthetic data set, the PSNR and SSIM of our designed network can be as high as 25.72 and 0.8859, respectively. On the real data set, the PSNR and SSIM of our designed network can reach 21.02 and 0.8166, respectively. It is proved that the defogging network in this paper has good practicability and universality.https://www.mdpi.com/2079-9292/12/10/2186image defoggingcycle-consistent adversarial networkmulti-scale featuresself-attention moduleatrous convolution |
| spellingShingle | Junkai Zhang Xiaoming Sun Yan Chen Yan Duan Yongliang Wang Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network Electronics image defogging cycle-consistent adversarial network multi-scale features self-attention module atrous convolution |
| title | Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network |
| title_full | Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network |
| title_fullStr | Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network |
| title_full_unstemmed | Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network |
| title_short | Single-Image Defogging Algorithm Based on Improved Cycle-Consistent Adversarial Network |
| title_sort | single image defogging algorithm based on improved cycle consistent adversarial network |
| topic | image defogging cycle-consistent adversarial network multi-scale features self-attention module atrous convolution |
| url | https://www.mdpi.com/2079-9292/12/10/2186 |
| work_keys_str_mv | AT junkaizhang singleimagedefoggingalgorithmbasedonimprovedcycleconsistentadversarialnetwork AT xiaomingsun singleimagedefoggingalgorithmbasedonimprovedcycleconsistentadversarialnetwork AT yanchen singleimagedefoggingalgorithmbasedonimprovedcycleconsistentadversarialnetwork AT yanduan singleimagedefoggingalgorithmbasedonimprovedcycleconsistentadversarialnetwork AT yongliangwang singleimagedefoggingalgorithmbasedonimprovedcycleconsistentadversarialnetwork |