Bright Channel Attention Optimized Dark Channel Prior for Unsupervised Nighttime Image Dehazing

10.6919/ICJE.202205_8(5).0084

Xuan Gong

Nighttime Image Dehazing ； Unsupervised Learning ； Attention Mechanism ； Generative Adversarial Network ； Bright Channel

International Core Journal of Engineering

8卷5期（2022 / 05 / 01）

655 - 665

英文

Images captured in nighttime hazy scenes often have complex lighting conditions and color casts, so nighttime image dehazing is more challenging. An unsupervised nighttime image dehazing algorithm with bright channel attention to optimize the dark channel prior is proposed. First, the algorithm uses the dark channel prior to restore the visibility of the image; then, the method uses the bright channel attention mechanism to guide the generative adversarial network to optimize the dehazing result of the dark channel prior. In order to enable the model to better restore nighttime foggy images, a bright channel attention mechanism is proposed to guide the generator model, so that the learned generator model can focus on the restoration of degraded areas of the image. The experimental results show that the proposed algorithm can achieve good dehazing effect on both real nighttime hazy images and synthetic datasets, and can well solve the color shift of nighttime hazy images.

1. LI Boyi, Peng Xiulian, Wang Zhangyang, et al. End-to-end united video dehazing and detection[C]// 2018 the AAAI Conference on Artificial Intelligence. New Orleans : AAAI, 2018: 7016-7023.
   連結：
2. DI Wu, Qingsong ZHU. The Latest Research Progress of Image Dehazing [J]. Acta Automatica Sinica,2015,41(2):221-239.
   連結：
3. BERMAN D, TREIBITZ T, AVIDAN S, Non-local Image Dehazing[C]// 2016 the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas: IEEE, 2016:1674-1682.
   連結：
4. HE Kaiming, SUN Jian, TANG Xiaoou, Single image haze removal using dark channel prior[J], IEEE Transactions Pattern Analysis and Machine Intelligence, 2011,33(12):2341-2353.
   連結：
5. BERMAN D, TREIBITZ T, AVIDAN S, Single Image Dehazing Using Haze-Lines[J], IEEE Transactions Pattern Analysis and Machine Intelligence, 2018,42(3):720-734.
   連結：
6. CAI Bolun, XU Xiangmin, JIA Kui, et al. DehazeNet: An End-to-End System for Single Image Haze Removal[J], IEEE Transactions Image Processing, 2016, 25(11):5187-5198.
   連結：
7. LI Boyi, PENG Xiulian, WANG Zhangyang, et al. AOD-Net: All-in-One Dehazing Network[C]// 2017 IEEE International Conference on Computer Vision (ICCV), Venice : IEEE 2017:4780-4788.
   連結：
8. PEI Soochang, LEE Tzuyen, Nighttime haze removal using color transfer pre-processing and Dark Channel Prior[C]// 2012 19th IEEE International Conference on Image Processing, Florida:IEEE, 2012:957-960.
   連結：
9. ZHANG Jing, CAO Yang, WANG Zengfu, Nighttime haze removal based on a new imaging model[C]// 2014 IEEE International Conference on Image Processing, Paris : IEEE,2014:4557-4561.
   連結：
10. LI Yu, TAN R T, BROWN M S, Nighttime Haze Removal with Glow and Multiple Light Colors[C]// 2015 IEEE International Conference on Computer Vision, Santiago : IEEE , 2015: 226-234.
    連結：
11. Aiping, NanWang. Nighttime Image Dehazing Algorithm by Structure-Texture Image Decomposition[J]. Laser & Optoelectronics Progress,2018,55(6):95-102.
    連結：
12. ZHANG Jing, CAO Yang, KANG Yu, Fast Haze Removal for Nighttime Image Using Maximum Reflectance Prior[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu : IEEE , 2017:7016-7024.
    連結：
13. YU Teng, SONG Kang, MIAO Pu, et al. Nighttime Single Image Dehazing via Pixel-Wise Alpha Blending[J], IEEE Access. 2019,7(1):6264-6275.DOI:10.1109/ACCESS.2019.2936049.
    連結：
14. TANG Qufang, YANG Jie, HE Xiangjian, et al. Nighttime image dehazing based on Retinex and dark channel prior using Taylor series expansion[J], Computer Vision and Image Understanding, 2021,202(1):103086.
    連結：
15. KUANAR S, MAHAPATRA D, BILASM, Multi-path dilated convolution network for haze and glow removal in nighttime images[J]. The Visual Computer, 2021,1-14.
    連結：
16. FENG Mengyao, TU Teng, JING Mingtao, et al. Learning a convolutional auto-encoder for nighttime image dehazing[J], Information. 2020,11(9): 424.
    連結：
17. ZHANG Jing, CAO Yang, ZHA Zhengjun, et al. Nighttime dehazing with a synthetic benchmark[C]// 2020 28th ACM International Conference on Multimedia, Seattle:ACM, 2020:2355–2363.
    連結：
18. ZHANG He, SINDAGI V, PATEL V, Joint Transmission Map Estimation and Dehazing Using Deep Networks[J], IEEE Transactions on Circuits and Systems for Video Technology,2020,30(7):1975-1986.
    連結：
19. LI Runde, PAN Jinshan, LI Zechao, et al. Single Image Dehazing via Conditional Generative Adversarial Network[C]// 2018 IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City:IEEE, 2018:8202-8211.
    連結：
20. ENGIN D, GENC A, EKENEL H K, Cycle-Dehaze: Enhanced CycleGAN for Single Image Dehazing[C]// 2018 IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City:IEEE, 2018:825-833.
    連結：
21. NAYAR S K, NARASIMHAN, Vision in bad weather[C]// the 7th IEEE International Conference on Computer Vision, Kerkyra: IEEE, 1999:820-827.
    連結：
22. RONNEBERGER O,FISCHER P,BROX T, U-Net: Convolutional Networks for Biomedical Image Segmentation[C]// International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich: Springer,2015:234-241.
    連結：
23. ZHU Junyan, PARK T, ISOLA P, et al. Unpaired image-to-image using cycle-consistent adversarial networks[C]// 2017 IEEE International Conference on Computer Vision, Venice:IEEE,2017:2232-2969.
    連結：
24. Wang Daolai, Zhang Tianyu. Review and analysis of image defogging algorithm [J]. Chinese Journal of Graphics ,2020,41(6):861-870.
25. GOODGELLOW I, POUGET A J, MIRZA M, et al. Generative adversarial nets[C]// International Conference on Neural Information Processing Systems, Montreal: Neural information processing systems foundation, 2014: 2672-2680.