Abstract
Due to the problem of reduced visibility caused by haze pollution, it has brought great inconvenience to outdoor activities and traffic travel. Therefore, haze control has become one of the topics closely concerned by all sectors of society. Real time and accurate visibility detection is one of the important links to effectively prevent the impact of sudden fog or haze on driving safety, but the cost performance, accuracy and popularity of existing methods and equipment need to be improved. In this paper, we proposed a vision-based haze visibility detection method that achieves a better balance in the abovementioned aspects. Specifically, an improved AlexNet network with multi-scale feature mechanism is used to capture richer haze spatial details. Then, channel attention is employed to emphasize the characters of the feature maps. Finally, densely connection network is utilized to improve the contribution of the interconnected information flows. The experimental results show that the algorithm is consistent with the human observation, which meets the safety requirements. Moreover, our method achieves 95.6% accuracy with 11.68% improvement.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Steffens, C.: Measurement of visibility by photo-graphic photometry. Ind. Eng. Chem. 41(11), 2396–2399 (1949)
Busch, C., Debes, E.: Wavelet transform for analyzing fog visibility. IEEE Intell. Syst. 6, 66–71 (1998)
Chen, Z., Zhou, Q., Chen, Q.: Video visibility detection algorithm based on wavelet transformation. Chin. J. Sci. Instrum. 31(1), 92–98 (2010)
Chen, Z.Z., Chen, Q.M.: Video contrast visibility detection algorithm and its implementation based on camera self-calibration. J. Electron. Inf. Technol. 32(12), 2907–2912 (2010)
Li, B., Dong, R., Chen, Q.M.: Automatic calibration method for PTZ camera. J. Beijing Univ. Posts Telecommun. 32(1), 24–29 (2010)
An, M.W., Guo, Z.L., Li, J., et al.: Visibility detection based on traffic camera imagery. In: International Conference on Information Sciences and Interaction Sciences, pp. 411–414. IEEE (2010)
Hautiere, N., Tarel, J.P., Lavenant, J., et al.: Automatic fog detection and estimation of visibility distance through use of an onboard camera. Mach. Vis. Appl. 17(1), 8–20 (2006). https://doi.org/10.1007/s00138-005-0011-1
Chen, Z.: PTZ visibility detection based on image luminance changing tendency. In: International Conference on Optoelectronics and Image Processing (ICOIP), pp. 15–19. IEEE (2016)
Graves, N., Newsam, S.: Using visibility cameras to estimate atmospheric light extinction. In: Workshop on Applications of Computer Vision (WACV), pp. 577–584. IEEE (2011)
Song, H.J., Gao, Y.Y., Chen, Y.Z.: Traffic visibility estimation based on dynamic camera calibration. Chin. J. Comput. 38(6), 1172–1187 (2015)
Belaroussi, R., Gruyer, D.: Road sign-aided estimation of visibility conditions. In: 2015 14th IAPR International Conference on Machine Vision Applications (MVA), pp. 202–205. IEEE (2015)
Varjo, S., Kaikkonen, V., Hannuksela, J., et al.: All-in-focus image reconstruction from in-line holograms of snowflakes. In: 2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), pp. 1096–1101. IEEE (2015)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classificationwith deep convolutional neural networks. In: Advance in Neural Information Processing Systems, pp. 1097–1105 (2012)
Kikuchi, R., Soffer, B.H.: Maximum entropy image restoration. I. The entropy expression. JOSA 67, 1656–1665 (1977)
Paszke, A., Gross, S., Chintala, S., et al.: Automatic differentiation in PyTorch. In: Workshop on Machine Learning Systems, NIPS (2017)
Kingma, D., Ba, J.: Adam: a method for stochastic optimization. In: ICLR (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Tao, J., Wu, Y., Shao, Q., Yan, S. (2022). Multi-scale Feature Based Densely Channel Attention Network for Vision-Based Haze Visibility Detection. In: Liu, H., et al. Intelligent Robotics and Applications. ICIRA 2022. Lecture Notes in Computer Science(), vol 13458. Springer, Cham. https://doi.org/10.1007/978-3-031-13841-6_51
Download citation
DOI: https://doi.org/10.1007/978-3-031-13841-6_51
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-13840-9
Online ISBN: 978-3-031-13841-6
eBook Packages: Computer ScienceComputer Science (R0)