Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Journal of Artificial Intelligence Practice
  4. Vol 8, Issue 3, 2025

Research on Underground Non-uniform Fog Removal Method Based on Enhanced Parallel Attention Mechanism

Download as PDF

DOI: 10.23977/jaip.2025.080317 | Downloads: 1 | Views: 23

Author(s)

Huan Zhang 1, Lingfei Cheng 1, Kui Tang 1

Affiliation(s)

1 School of Physics and Electronic Information, Henan Polytechnic University, Jiaozuo, Henan, 454003, China

Corresponding Author

Lingfei Cheng

ABSTRACT

The image quality in the underground environment is limited by insufficient lighting and the interference of non-uniform dust and mist generated by work activities. This non-uniform fog results in low image visibility, blurry details, and color distortion, which hinders underground safety monitoring. For this purpose, a model was designed for the removal of non-uniform fog underground. Firstly, the module includes multi-scale convolution and parallel attention mechanism. Multi scale convolution can obtain more feature information from images in order to restore texture information. Parallel attention can better capture multi-dimensional global information, improve the comprehensiveness of feature extraction, and perform well in removing non-uniform fog. In addition, the SE attention module is introduced to automatically learn the sensitivity of different channels to fog concentration, with high weights for dense fog areas, enhancing the dehazing effect. Finally, the PSNR and SSIM of the Haze4K dataset were verified to be 32.18 and 0.963, respectively. The validation indicators for the self-made non-uniform fog dataset are PSNR of 32.37dB and SSIM of 0.981. This provides a certain reference value for obtaining high-quality images for underground monitoring.

KEYWORDS

Downhole Image, Image Dehazing, U-Net, Deep Learning

CITE THIS PAPER

Huan Zhang, Lingfei Cheng, Kui Tang, Research on Underground Non-uniform Fog Removal Method Based on Enhanced Parallel Attention Mechanism. Journal of Artificial Intelligence Practice (2025) Vol. 8: 134-143. DOI: http://dx.doi.org/10.23977/jaip.2025.080317.

REFERENCES

[1] McCartney, E.J. Optics of the Atmosphere: Scattering by Molecules and Particles[M]. New York: Wiley, 1976.
[2] Nayar, S.K.; Narasimhan, S.G. Vision in bad weather[C]. In Proceedings of the Seventh IEEE International Conference on Computer Vision. Kerkyra, Greece: IEEE, 1999: 820–827. 
[3] He, K.; Sun, J.; Tang, X. Single image haze removal using dark channel prior[C]. In Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL, USA: IEEE, 2009: 1956–1963. 
[4] Wang, Y.; Wei, S.; Duan, Y.; Wu, H. Defogging algorithm of underground coal mine image based on adaptive dual-channel prior[J]. Journal of Mine Automation, 2022, 48: 46–51+84. 
[5] Cao, H.; Yao, S.; Wang, Z. Defogging algorithm of underground coal mine dust and fog image based on boundary constraint[J]. Journal of Mine Automation, 2022, 48, 139–146. 
[6] Li, B.; Peng, X.; Wang, Z.; Xu, J.; Feng, D. AOD-Net: All-in-One Dehazing Network[C]. In Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy: IEEE, 2017: 4780–4788. 
[7] Wang, K., Liu, Y., Yang, Y., Zhang, G., & Qian, W. Single Image Dehazing Based on Haze Prior Residual Perception Learning[J]. Circuits, Systems, and Signal Processing, 2025. doi: 10.1007/s00034-025-03058-0.
[8] Cai, B.; Xu, X.; Jia, K.; Qing, C. Tao, D. Dehazenet: An end-to-end system for single image haze removal[J]. IEEE Transactions on Image Processing, 2016, 25(11), 5187–5198. 
[9] Huang, H., Ouyang, H., Dong, Y., He, X., & Zhao, X. Fast dehazing for large format oblique images based on improved dark channel prior[J].Optics and Precision Engineering, 2025, 33(3): 476–485. doi: 10.37188/OPE. 20253303.0476.
[10] Li, X., Xia, F., Zhang, K., Wang, H., & Xie, T. Enhanced edge feature extraction dual branch fusion network for real image dehazing[J]. Optics and Precision Engineering, 2025, 33(2), 247–261. doi: 10.37188/OPE.20253302.0247.
[11] Qin, X.; Wang, Z.; Bai, Y.; Xie, X.; Jia, H. FFA-Net: Feature fusion attention network for single image dehazing[C]. Proceedings of the AAAI Conference on Artificial Intelligence. 2020, 34: 11908–11915. 
[12] Chen, D.; He, M.; Fan, Q.; Liao, J.; Zhang, L.; Hou, D.; Yuan, L.; Hua, G. Gated Context Aggregation Network for Image Dehazing and Deraining[C]. In Proceedings of the 2019 IEEE Winter Conference on Applications of Computer Vision (WACV). Waikoloa, HI, USA: IEEE, 2019: 1375–1383. 
[13] Li, H.-Y., Qiao, R.-C., Li, H.-J., & Chen, Q. CNN-Transformer Dehazing Algorithm Based on Global Residual Attention and Gated Feature Fusion[J]. Journal of Northeastern University, 2025, 46(1): 26–34. doi: 10.12068/j.issn. 1005-3026.2025.20239041.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 15934
Visits: 544804

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.