Sage Journals: Discover world-class research

Abstract

Clear visual data is a fundamental prerequisite for real-time detection of and decision-making about abnormal traffic conditions in highway intelligent traffic systems (ITS). However, as a frequent weather phenomenon, haze can significantly reduce image contrast and obscure critical visual features, posing a serious threat to the effectiveness of highway ITS. This paper proposes an image defogging method based on spatial-frequency domain context joint convolutional network, aiming to address the limitations of existing methods in capturing frequency domain correlations and reducing spatial redundancy. Firstly, the method decomposes the haze characteristics from the spatial-frequency domain perspective, establishes a frequency domain defogging model more suitable for hazy scenes, and fully utilizes multi-scale frequency features to represent the correlation between the fog layer and the clear background. Secondly, based on the scale-space transformation and frequency-division feature mapping strategy, the feature correlations of low-frequency and high-frequency components are effectively decomposed, and redundancy is eliminated to improve computational efficiency. Then, a multi-layer channel attention mechanism is introduced, enabling the model to flexibly extract the weight information of haze distributions at different layers based on the frequency-domain feature decomposition, strengthening the attention to key features, and gradually stripping the complex overlapping fog layers through a sequential iterative approach. Finally, a defogging dataset is constructed to verify the effectiveness of the proposed method. Experiment results demonstrate that the proposed method achieves a peak signal-to-noise ratio of 30.3 dB and a structural similarity index measure of 0.9, outperforming existing methods. These results confirm the effectiveness of our method in restoring image clarity and preserving structural details, offering practical support for highway ITS under adverse weather conditions.

Keywords

traffic surveillance hazy scenes image defogging frequency domain fusion multi-channel attention mechanism

Get full access to this article

View all access options for this article.

References

Garg

Navar

S. K.

Detection and Removal of Rain from Videos. Proc., 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol. 1, 2004, p. I.

Barnum

Kanade

Narasimhan

S. G.

Spatio-Temporal Frequency Analysis for Removing Rain and Snow from Videos. Proc., First International Workshop on Photometric Analysis for Computer Vision, Vol. 6, 2007, p. 8.

Zhen

Jihong

A New Algorithm of Rain (Snow) Removal in Video. Journal of Multimedia, Vol. 8, No. 2, 2013, p. 168.

Pucheng

Yuan

Yusheng

Research Progress of Video Image Rain Removal Technology. Journal of Graphics, Vol. 38, No. 5, 2017, pp. 629–646.

Barnum

P. C.

Narasimhan

Kanade

Analysis of Rain and Snow in Frequency Space. International Journal of Computer Vision, Vol. 86, No. 2, 2010, pp. 256–274.

Kang

L. W.

Lin

C. W.

Y. H.

Automatic Single-Image-Based Rain Streaks Removal Via Image Decomposition. IEEE Transactions on Image Processing, Vol. 21, No. 4, 2011, pp. 1742–1755.

Luo

Removing Rain from a Single Image Via Discriminative Sparse Coding. Proc., IEEE International Conference on Computer Vision, Vol. 21, 2015, pp. 3397–3405.

Tan

R. T.

Guo

, et al. Rain Streak Removal Using Layer Priors. Proc., IEEE Conference on Computer Vision and Pattern Recognition, Vol. 3, 2016, pp. 2736–2744.

Han

Pool

Narang

Mao

Gong

Tang

Dally

W. J.

Dsd: Dense-Sparse-Dense Training for Deep Neural Networks. arXiv preprint arXiv:1607.04381, 2016.

10.

Luo

J. H.

Zhang

Zhou

H. Y.

Xie

C. W.

Lin

Thinet: Pruning CNN Filters for a Thinner Net. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 41, No. 10, 2018, pp. 2525–2538.

11.

Tung

Mori

Clip-q: Deep Network Compression Learning by in-Parallel Pruning-Quantization. Proc., IEEE Conference on Computer Vision and Pattern Recognition, Vol. 12, 2018, pp. 7873–7882.

12.

Xie

Girshick

Dollár

Aggregated Residual Transformations for Deep Neural Networks. Proc., IEEE Conference on Computer Vision and Pattern Recognition, Vol. 5, 2017, pp. 1492–1500.

13.

Howard

A. G.

Zhu

Chen

Kalenichenko

Wang

Weyand

Adam

Mobilenets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv preprint arXiv:1704.04861, 2017.

14.

Chen

Kalantidis

Yan

Feng

Multi-Fiber Networks for Video Recognition. Proc., European Conference on Computer Vision, Vol. 18, 2018, pp. 352–367.

15.

Chollet

Xception: Deep Learning with Depthwise Separable Convolutions. Proc., IEEE Conference on Computer Vision and Pattern Recognition, Vol. 1, 2017, pp. 1251–1258.

16.

Huang

Zeng

Huang

Ding

Paisley

Removing Rain from Single Images Via a Deep Detail Network. Proc., IEEE Conference on Computer Vision and Pattern Recognition, Vol. 25, 2017, pp. 3855–3863.

17.

Chen

Zhang

Zhou

Zhao

Removing Rain Based on a Cycle Generative Adversarial Network. IEEE Conference on Industrial Electronics and Applications, Vol. 4, 2018, pp. 621–626.

18.

Lin

Liu

Zha

Recurrent Squeeze-and-Excitation Context Aggregation Net for Single Image Deraining. Proc., European Conference on Computer Vision, Vol. 2, 2018, pp. 254–269.

19.

Zheng

Liu

Zhang

Residual Multiscale Based Single Image Deraining. BMVC, Vol. 5, 2019, p. 147.

20.

Zhang

Patel

V. M.

Density-Aware Single Image De-Raining Using a Multi-Stream Dense Network. Proc., IEEE Conference on Computer Vision and Pattern Recognition, Vol. 3, 2018, pp. 695–704.

21.

Liang

Huang

Ding

Paisley

Lightweight Pyramid Networks for Image Deraining. IEEE Transactions on Neural Networks and Learning Systems, Vol. 15, 2018, pp. 1–14.

22.

Rahaman

Baratin

Arpit

Draxler

Lin

Hamprecht

Courville

On the Spectral Bias of Neural Networks. International Conference on Machine Learning, Vol. 4, 2019, pp. 5301–5310.

23.

Sun

Tang

Single Image Haze Removal Using Dark Channel Prior. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 33, No. 12, 2010, pp. 2341–2353.

24.

Zhu

Zhang

Wang

Feng

Duan

Spectral Dual-Channel Encoding for Image Dehazing. IEEE Transactions on Circuits and Systems for Video Technology, Vol. 33, No. 11, 2023, pp. 6236–6248.

25.

Sahu

Seal

Bhattacharjee

Frischer

Krejcar

A Novel Parameter Adaptive Dual Channel MSPCNN Based Single Image Dehazing for Intelligent Transportation Systems. IEEE Transactions on Intelligent Transportation Systems, Vol. 24, No. 3, 2022, pp. 3027–3047.

26.

Tang

Cruz

J. D.

Dehazing and Detection of Foggy Traffic Images Using Bilateral Filtering and Improved YOLOX-S Algorithm. International Conference on Electronic Information Technology and Computer Engineering, Vol. 5, 2024, pp. 250–254.

27.

Lindeberg

Scale-Space Theory in Computer Vision. Springer Science & Business Media, 2013.

28.

Chen

Fan

Yan

Kalantidis

Rohrbach

Feng

Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution. Proc., IEEE/CVF International Conference on Computer Vision, Vol. 21, 2019, pp. 3435–3444.

29.

Bai

Kolter

J. Z.

Koltun

An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv preprint arXiv:1803.01271, 2018.

30.

Shen

Sun

Squeeze-and-Excitation Networks. IEEE Conference on Computer Vision and Pattern Recognition, Vol. 6, 2018, pp. 7132–7141.

31.

Guo

Domain Generalized Person Re-Identification With Attention Mechanism. Control and Decision. https://doi.org/10.13195/j.kzyjc.2020.1844.

32.

Ioffe

Szegedy

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. International Conference on Machine Learning, Vol. 12, 2015, pp. 448–456.

33.

K. M.

Sun

Tang

X. O.

Single Image Haze Removal Using Dark Channel Prior. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 33, No. 12, 2011, pp. 2341–2352.

34.

Cai

B. L.

Jia

Qing

Tang

X. O.

DehazeNet: An End-to-End System for Single Image Haze Removal. IEEE Transactions on Image Processing, Vol. 25, No. 11, 2016, pp. 5187–5198.

35.

B. L.

Peng

X. L.

Wang

Z. Y.

J. Z.

Feng

AOD-Net: All-in-One Dehazing Network. IEEE International Conference on Computer Vision, Vol. 15, 2017, pp. 4770–4778.

36.

Ren

W. Q.

Zhang

J. W.

Pan

J. S.

Cao

X. C.

Liu

Yang

M. H.

Gated Fusion Network for Single Image Dehazing. IEEE Conference on Computer Vision and Pattern Recognition. Vol. 6, 2018, pp. 3253–3261.

Traffic Image Defogging Method under Hazy Environments Based on Frequency Domain Fusion and Attention Mechanism

Abstract

Keywords

Get full access to this article

References