Abstract
The complex glyphs of license plates usually comes with a long-tail distribution, leading to poor recognition performance of the tail class. Supplementing the training data with generated license plates is an effective solution for this issue. However, for complex glyphs, the previous methods are prone to generate incomplete structures and blurry strokes. The first reason is that the small portion of complex glyphs on the license plate contributes little to the overall loss. Secondly, due to the complex structure and dense strokes, the glyphs are prone to be generated inaccurately. To solve the above problems, firstlly, we propose a divide-and-conquer method that generates complex and simple glyphs separately and then fuses them into a complete license plate, thus enhancing the generation of complex glyphs in loss computation. Secondly, we increase the generated resolution of complex glyph to enable the model to learn dense structures and fine strokes. Besides, considering the computational cost, low-resolution generation is used for the rest of the simple glyphs. Extensive experiments demonstrate that our method can significantly enhances the realism of the complex glyph, and generated images can boost recognition performance by 3\(\%\) on SYSU. Additionally, we provide a dataset of 30,000 generated Chinese license plates with uniform Chinese distribution to promote research (https://github.com/ICIG2023-91/GCLPD).
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
Barratt, S.T., Sharma, R.: A note on the inception score. CoRR abs/1801.01973 (2018)
Chen, X., Xie, Y., Sun, L., Lu, Y.: DGFont++: robust deformable generative networks for unsupervised font generation. CoRR abs/2212.14742 (2022)
Gatys, L.A., Ecker, A.S., Bethge, M.: Image style transfer using convolutional neural networks. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition, pp. 2414–2423 (2016)
Goodfellow, I.J., et al.: Generative adversarial networks. CoRR abs/1406.2661 (2014)
Gupta, A., Andrew Zisserman, A.V.: Synthetic data for text localisation in natural images. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition, pp. 2315–2324 (2016)
Hassan, A.U., Ahmed, H., Choi, J.: Unpaired font family synthesis using conditional generative adversarial networks. Knowl. Based Syst. 229, 107304 (2021)
Isola, P., Zhu, J., Zhou, T., Efros, A.A.: Image-to-image translation with conditional adversarial networks, pp. 5967–5976 (2017)
Jiang, Y., Lian, Z., Tang, Y., Xiao, J.: SCFont: structure-guided Chinese font generation via deep stacked networks. In: The Thirty-Third AAAI Conference on Artificial Intelligence, AAAI 2019, pp. 4015–4022. AAAI Press (2019)
Johnson, J., Alahi, A., Fei-Fei, L.: Perceptual losses for real-time style transfer and super-resolution. In: Computer Vision - ECCV 2016–14th European Conference, vol. 9906, pp. 694–711 (2016)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: 3rd International Conference on Learning Representations (2015)
Lyu, P., Bai, X., Yao, C., Zhu, Z., Huang, T., Liu, W.: Auto-encoder guided GAN for Chinese calligraphy synthesis. In: 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017, pp. 1095–1100. IEEE (2017)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention, vol. 9351, pp. 234–241 (2015)
Sajjadi, M.S.M., Bachem, O., Lucic, M., Bousquet, O., Gelly, S.: Assessing generative models via precision and recall. CoRR abs/1806.00035 (2018)
Sun, M., Zhou, F., Yang, C., Yin, X.: Image generation framework for unbalanced license plate data set. In: 2019 International Conference on Data Mining Workshops, pp. 883–889 (2019)
Sun, Y.F., Liu, Q., Chen, S.L., Zhou, F., Yin, X.C.: Robust Chinese license plate generation via foreground text and background separation. In: Image and Graphics - 11th International Conference, vol. 12890, pp. 290–302 (2021)
Wang, X., Man, Z., You, M., Shen, C.: Adversarial generation of training examples: applications to moving vehicle license plate recognition. CoRR abs/1707.03124 (2017)
Wu, C., Xu, S., Song, G., Zhang, S.: How many labeled license plates are needed? In: Pattern Recognition and Computer Vision - First Chinese Conference, vol. 11259, pp. 334–346 (2018)
Wu, L., Zhang, C., Liu, J., Han, J., Liu, J., Ding, E., Bai, X.: Editing text in the wild. In: Proceedings of the 27th ACM International Conference on Multimedia, pp. 1500–1508 (2019)
Xu, Z., et al.: Towards end-to-end license plate detection and recognition: A large dataset and baseline. In: Computer Vision - ECCV 2018–15th European Conference, vol. 11217, pp. 261–277 (2018)
Yu, D., Li, X., Zhang, C., Liu, T., Han, J., Liu, J., Ding, E.: Towards accurate scene text recognition with semantic reasoning networks. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12110–12119 (2020)
Zhang, L., Wang, P., Li, H., Li, Z., Shen, C., Zhang, Y.: A robust attentional framework for license plate recognition in the wild. IEEE Trans. Intell. Transp. Syst. 22(11), 6967–6976 (2021)
Zhao, Y., Yu, Z., Li, X., Cai, M.: Chinese license plate image database building methodology for license plate recognition. J. Electron. Imaging 28(1), 013001 (2019)
Zhou, W., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)
Zhu, J.Y., Park, T., Isola, P., Efros, A.A.: Unpaired image-to-image translation using cycle-consistent adversarial networks. In: IEEE International Conference on Computer Vision, pp. 2242–2251 (2017)
Acknowledgement
The research is supported by National Key Research and Development Program of China (2020AAA0109700), National Natural Science Foundation of China (62076024, 62006018, U22B2055).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Chen, YX. et al. (2023). Complex Glyph Enhancement for License Plate Generation. In: Lu, H., et al. Image and Graphics. ICIG 2023. Lecture Notes in Computer Science, vol 14355. Springer, Cham. https://doi.org/10.1007/978-3-031-46305-1_25
Download citation
DOI: https://doi.org/10.1007/978-3-031-46305-1_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-46304-4
Online ISBN: 978-3-031-46305-1
eBook Packages: Computer ScienceComputer Science (R0)