Development of a Modified CycleGAN Model with Residual Blocks and Perceptual Loss for Image Dehazing

Sani Moch Sopian; Arief Suryadi Satyawan; Mokhammad Mirza Etnisa Haqiqi; Helfy Susilawati; Beni Wijaya; Khaulyca Arva Artemysia; Firman; Muhammad Ikbal Samie

doi:10.34148/teknika.v14i2.1235

Authors

Sani Moch Sopian Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia https://orcid.org/0009-0003-5783-0865 (unauthenticated)
Arief Suryadi Satyawan Research Center for Telecommunication – BRIN, Indonesia
Mokhammad Mirza Etnisa Haqiqi Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia
Helfy Susilawati Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia
Beni Wijaya Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia
Khaulyca Arva Artemysia Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia
Firman Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia
Muhammad Ikbal Samie Department of Electrical Engineering, Faculty of Engineering, Garut University, West Java, Indonesia

DOI:

https://doi.org/10.34148/teknika.v14i2.1235

Keywords:

Dehazing, CycleGAN, Deep Learning, Residual Block, Attention Mechanism

Abstract

Fog reduces image contrast and clarity, creating challenges for applications such as autonomous driving and remote sensing. This study proposes a series of CycleGAN modifications for single image dehazing using unpaired data, integrating residual blocks, attention mechanisms, VGG19-based perceptual loss, and haze-aware loss. Among ten architectural variants, Modification 10 combining perceptual and haze-aware loss achieved the best overall performance. Quantitatively, it showed stable generator losses (0.91 for Gen G, 0.57 for Gen F), with improved discriminator performance (Disc X: 0.59, Disc Y: 0.47), indicating better training stability and image realism. Additionally, it offered competitive PSNR (7.99), strong SSIM (0.4202), and low LPIPS (0.6577), confirming its effectiveness in both pixel-level accuracy and perceptual quality. Qualitatively, this model generated clearer, more natural images with improved edge sharpness and detail preservation. These findings demonstrate that the modified CycleGAN significantly enhances dehazing performance and presents a valuable contribution to deep learning-based image restoration.

Downloads

Download data is not yet available.

References

[1] D. Engin, A. Genç, and H. K. Ekenel, “Cycle-Dehaze: Enhanced CycleGAN for Single Image Dehazing,” May 2018, [Online]. Available: http://arxiv.org/abs/1805.05308

[2] Y. Qiu, Y. Lu, Y. Wang, and H. Jiang, “IDOD-YOLOV7: Image-Dehazing YOLOV7 for Object Detection in Low-Light Foggy Traffic Environments,” Sensors, vol. 23, no. 3, Feb. 2023, doi: 10.3390/s23031347.

[3] S. Lee, S. Yun, J. H. Nam, C. S. Won, and S. W. Jung, “A review on dark channel prior based image dehazing algorithms,” Dec. 01, 2016, Springer International Publishing. doi: 10.1186/s13640-016-0104-y.

[4] D. Berman, T. Treibitz, and S. Avidan, “Non-local Image Dehazing,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, Dec. 2016, pp. 1674–1682. doi: 10.1109/CVPR.2016.185.

[5] B. Cai, X. Xu, K. Jia, C. Qing, and D. Tao, “DehazeNet: An end-to-end system for single image haze removal,” IEEE Transactions on Image Processing, vol. 25, no. 11, pp. 5187–5198, Nov. 2016, doi: 10.1109/TIP.2016.2598681.

[6] B. Li, X. Peng, Z. Wang, J. Xu, and D. Feng, “An All-in-One Network for Dehazing and Beyond,” Jul. 2017, [Online]. Available: http://arxiv.org/abs/1707.06543

[7] X. Liu, Y. Ma, Z. Shi, and J. Chen, “GridDehazeNet: Attention-Based Multi-Scale Network for Image Dehazing,” Aug. 2019, [Online]. Available: http://arxiv.org/abs/1908.03245

[8] J. Y. Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks,” in Proceedings of the IEEE International Conference on Computer Vision, Institute of Electrical and Electronics Engineers Inc., Dec. 2017, pp. 2242–2251. doi: 10.1109/ICCV.2017.244.

[9] J. Chen, C. Wu, H. Chen, and P. Cheng, “Unsupervised dark-channel attention-guided cyclegan for single-image dehazing,” Sensors (Switzerland), vol. 20, no. 21, pp. 1–15, Nov. 2020, doi: 10.3390/s20216000.

[10] Y. Liu, H. Al-Shehari, and H. Zhang, “Attention mechanism enhancement algorithm based on cycle consistent generative adversarial networks for single image dehazing,” J Vis Commun Image Represent, vol. 83, Feb. 2022, doi: 10.1016/j.jvcir.2021.103434.

[11] H. Ni and W. Su, “A Self-Attention CycleGAN for Unsupervised Image Hazing,” Big Data and Cognitive Computing, vol. 9, no. 4, Apr. 2025, doi: 10.3390/bdcc9040096.

[12] J. Wang, Y. zhang, and Z. Liu, “Unsupervised dehazing of multi-scale residuals based on weighted contrast learning,” Aug. 22, 2024. doi: 10.21203/rs.3.rs-4812948/v1.

[13] H. Liu, X. Li, and T. Tan, “Interaction-Guided Two-Branch Image Dehazing Network,” 2024. [Online]. Available: https://link.springer.com/conference/accv

[14] X. Chen et al., “Unpaired Deep Image Dehazing Using Contrastive Disentanglement Learning,” Mar. 2022, [Online]. Available: http://arxiv.org/abs/2203.07677

[15] A. Jain, “Feature Fusion Attention Network with CycleGAN for Image Dehazing, De-Snowing and De-Raining,” Mar. 2025, [Online]. Available: http://arxiv.org/abs/2503.06107

[16] D. Yang, Z. Zhu, H. Ge, H. Qiu, H. Wang, and C. Xu, “A Lightweight Neural Network for the Real-Time Dehazing of Tidal Flat UAV Images Using a Contrastive Learning Strategy,” Drones, vol. 8, no. 7, Jul. 2024, doi: 10.3390/drones8070314.