Home > Research > Publications & Outputs > Variant-Depth Neural Networks for Deblurring Tr...

Electronic data

Links

Text available via DOI:

View graph of relations

Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems. / Wang, Qian; Guo, Cai; Dai, Hong-Ning et al.
In: IEEE Transactions on Intelligent Transportation Systems, Vol. 24, No. 6, 6, 01.06.2023, p. 5792-5802.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Wang, Q, Guo, C, Dai, H-N & Xia, M 2023, 'Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems', IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 6, 6, pp. 5792-5802. https://doi.org/10.1109/tits.2023.3255839

APA

Wang, Q., Guo, C., Dai, H.-N., & Xia, M. (2023). Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems. IEEE Transactions on Intelligent Transportation Systems, 24(6), 5792-5802. Article 6. https://doi.org/10.1109/tits.2023.3255839

Vancouver

Wang Q, Guo C, Dai HN, Xia M. Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems. IEEE Transactions on Intelligent Transportation Systems. 2023 Jun 1;24(6):5792-5802. 6. Epub 2023 Apr 4. doi: 10.1109/tits.2023.3255839

Author

Wang, Qian ; Guo, Cai ; Dai, Hong-Ning et al. / Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems. In: IEEE Transactions on Intelligent Transportation Systems. 2023 ; Vol. 24, No. 6. pp. 5792-5802.

Bibtex

@article{fddf5801699446ccbcef46b7527f81bc,
title = "Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems",
abstract = "Intelligent transportation systems (ITS) with surveillance cameras capture traffic images or videos. However, images or videos in ITS often encounter blurs due to various reasons. Considering resource limitations, although recent technologies make progress in image-deblurring, there are still challenges in applying image-deblurring models in practical transportation systems: the model size and the running time. This work proposes an artful variant-depth network (VDN) to address the challenges. We design variant-depth sub-networks in a coarse-to-fine manner to improve the deblurring effect. We also adopt a new connection namely stack connection to connect all sub-networks to reduce the running time and model size while maintaining high deblurring quality. We evaluate the proposed VDN with the state-of-the-art (SOTA) methods on several typical datasets. Results on Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) show that the VDN outperforms SOTA image-deblurring methods. Furthermore, the VDN also has the shortest running time and the smallest model size.",
keywords = "Computer Science Applications, Mechanical Engineering, Automotive Engineering",
author = "Qian Wang and Cai Guo and Hong-Ning Dai and Min Xia",
year = "2023",
month = jun,
day = "1",
doi = "10.1109/tits.2023.3255839",
language = "English",
volume = "24",
pages = "5792--5802",
journal = "IEEE Transactions on Intelligent Transportation Systems",
issn = "1524-9050",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

RIS

TY - JOUR

T1 - Variant-Depth Neural Networks for Deblurring Traffic Images in Intelligent Transportation Systems

AU - Wang, Qian

AU - Guo, Cai

AU - Dai, Hong-Ning

AU - Xia, Min

PY - 2023/6/1

Y1 - 2023/6/1

N2 - Intelligent transportation systems (ITS) with surveillance cameras capture traffic images or videos. However, images or videos in ITS often encounter blurs due to various reasons. Considering resource limitations, although recent technologies make progress in image-deblurring, there are still challenges in applying image-deblurring models in practical transportation systems: the model size and the running time. This work proposes an artful variant-depth network (VDN) to address the challenges. We design variant-depth sub-networks in a coarse-to-fine manner to improve the deblurring effect. We also adopt a new connection namely stack connection to connect all sub-networks to reduce the running time and model size while maintaining high deblurring quality. We evaluate the proposed VDN with the state-of-the-art (SOTA) methods on several typical datasets. Results on Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) show that the VDN outperforms SOTA image-deblurring methods. Furthermore, the VDN also has the shortest running time and the smallest model size.

AB - Intelligent transportation systems (ITS) with surveillance cameras capture traffic images or videos. However, images or videos in ITS often encounter blurs due to various reasons. Considering resource limitations, although recent technologies make progress in image-deblurring, there are still challenges in applying image-deblurring models in practical transportation systems: the model size and the running time. This work proposes an artful variant-depth network (VDN) to address the challenges. We design variant-depth sub-networks in a coarse-to-fine manner to improve the deblurring effect. We also adopt a new connection namely stack connection to connect all sub-networks to reduce the running time and model size while maintaining high deblurring quality. We evaluate the proposed VDN with the state-of-the-art (SOTA) methods on several typical datasets. Results on Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) show that the VDN outperforms SOTA image-deblurring methods. Furthermore, the VDN also has the shortest running time and the smallest model size.

KW - Computer Science Applications

KW - Mechanical Engineering

KW - Automotive Engineering

U2 - 10.1109/tits.2023.3255839

DO - 10.1109/tits.2023.3255839

M3 - Journal article

VL - 24

SP - 5792

EP - 5802

JO - IEEE Transactions on Intelligent Transportation Systems

JF - IEEE Transactions on Intelligent Transportation Systems

SN - 1524-9050

IS - 6

M1 - 6

ER -