Article Details

Title · Authors · Abstract · Keywords · DOI · References

Road Traffic Dehazing Method Based on Improved AOD-Net Algorithm

SONG Yubo, LI Zixuan, LI Zhen

(Mechatronics T&R Institute, Lanzhou Jiaotong University, Lanzhou 730070, China)

Section：Artificial Intelligence · Vol. 2025, Issue 10 · Pages：39-44,49 · 2096-4706（2025）10-0039-07 · CLC number：TP391 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2025.10.008

Abstract

To address the problems of detail information loss and reduction of image clarity after traffic image dehazing, an improved AOD-Net algorithm is proposed. Aiming at the problem of information loss caused by insufficient feature extraction depth, a cascaded convolutional network is designed to more accurately identify and extract fine-grained features in the image. At the same time, in order to solve the problem that the AOD-Net algorithm does not fully consider the feature weight and easily leads to information redundancy, an adaptive weight distribution mechanism is introduced to dynamically adjust according to the importance of different information, so as to avoid the loss of detailed information. In addition, the clarity of the image after dehazing is improved by introducing the Smooth L1 loss function optimization model. The experiment is carried out on the public data set RESIDE. The results show that compared with the baseline model, the Peak Signal-to-Noise Ratio (PSNR) of the improved algorithm is increased by 0.52 dB, the Structural Similarity (SSIM) is increased by 0.0868, and the information entropy is increased by 0.76. The image after dehazing is clearer, which effectively improves the image quality. Compared with other algorithms, this method shows significant advantages in processing traffic scene images.

Keywords

image dehazing; AOD-Net algorithm; cascaded convolution; Attention Mechanism

References

  1. [1] SONG Y D，HE Z Q，QIAN H，et al. Vision Transformers for Single Image Dehazing [J].IEEE Transactions on Image Processing，2022：1927-1941.
  2. [2] Mao Deqian, Gao Shanshan, Lv Haixia, etc Multi scale image dehazing network based on local Transformer [J/OL]. Journal of Computer Aided Design and Graphics, 2025:1-14. [2022-04-16] http://kns.cnki.net/kcms/detail/11.2925.TP.20240204.1554.047.html.
  3. [3] LIU F L. An Overview of Image Enhancement Dehazing Algorithms [C]//The 3rd International Conference on Signal Processing and Machine Learning（CONF-SPML 2023）.Oxford：[sn]，2023：469-473.
  4. [4] WANG B，HU L，WEI B W，et al. Nighttime Image Dehazing Using Color Cast Removal and Dual Path Multi-Scale Fusion Strategy [J].Frontiers of Computer Science，2022，16（4）：164706.
  5. [5] Zheng Fengxian, Wang Xiali, He Dandan, etc A review of single image dehazing algorithm research [J]. Computer Engineering and Applications, 2022, 58 (3): 1-14
  6. [6] FU M H，LIU H，YU Y K，et al. DW-GAN: A Discrete Wavelet Transform GAN for NonHomogeneous Dehazing [C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops （CVPRW）.Nashville：IEEE，2021：203-212.
  7. [7] LI T Y，RONG S H，ZHAO W F，et al. Underwater Image Enhancement Using Adaptive Color Restoration and Dehazing [J].Optics Express，2022，30（4）：6216-6235.
  8. [8] HE K M，SUN J，TANG X O. Single Image Haze Removal Using Dark Channel Prior [J].IEEE Trans on Pattern Analysis and Machine Intelligence，2011，33（12）：2341-2353.
  9. [9] ZHU Q S，MAI J M，SHAO L. A fast Single Image Haze Removal Algorithm Using Color Attenuation Prior [J].IEEE Trans on Image Processing，2015，24（11）：3522-3533.
  10. [10] Hu Mengting, Fan Tao, from the Kingdom A traffic image dehazing algorithm based on adaptive polymorphic feature fusion [J]. Computer Application Research, 2023, 40 (7): 2210-2216
  11. [11] Zeng Ying, Liu Xin, Chen Jiyou, etc Single image dehazing based on enhanced multi-scale generative adversarial networks [J]. Small Microcomputer Systems, 2023, 44 (2): 370-375
  12. [12] Liu Zhe, Liang Yudong, Li Jiaying Adaptive Image dehazing Algorithm Based on Dynamic Convolutional Kernel [J]. Computer Science, 2023, 50 (6): 200-208
  13. [13] CAI B L，XU X M，JIA K，et al. Dehaze Net: An End-to-End System for Single Image Haze Removal [J].IEEE Trans on Image Processing，2016，25（11）：5187-5198.
  14. [14] REN W Q，PAN J S，ZHANG H，et al. Single Image Dehazing Via Multi-Scale Convolutional Neural Networks With Holistic Edges [J].International Journal of Computer Vision，2020，128（1）：240-259.
  15. [15] CHEN D D，HE M M，FAN Q N，et al. Gated Context Aggregation Network for Image Dehazing and Deraining [C]//2019 IEEE Winter Conference on Applications of Computer Vision（WACV）.Waikoloa：IEEE，2019：1375-1383.
  16. [16] LI B Y，PENG X L，WANG Z Y，et al. AOD-NeT: All-in-One Dehazing Network [J].2017 IEEE International Conference on Computer Vision.Venice：IEEE，2017，2017：4780-4788.
  17. [17] Qin X，WANG Z L，BAI Y C，et al. FFA-Net: Feature Fusion Attention Network for Single Image Dehazing [J/OL].[2024-10-28].arXiv:1911.07559 [cs.CV].https://arxiv.org/abs/1911.07559.
  18. [18] Zhang Qian, Chen Ziqiang, Jiang Hongyue, etc Improved lightweight image dehazing algorithm for AOD Net [J]. Laboratory Research and Exploration, 2022, 41 (7): 18-22

About the authors

Song Yubo (1977-), male, Han ethnicity, Liaoning Fuxin, associate professor, PhD, research interests: machine vision, image processing, intelligent control of automation equipment; Li Zixuan (2000-), female, Han ethnicity, from Baoding, Hebei Province, is currently pursuing a master's degree in image processing and object detection; Li Zhen (1998-), male, Han ethnicity, from Wuhan, Hubei, is currently pursuing a master's degree in infrared image processing and deep learning algorithm improvement.