DATA AUGMENTATION FOR A LEARNING-BASED VEHICLE MAKE-MODEL AND LICENSE PLATE MATCHING SYSTEM

Abstract

The most important requirement for deep learning algorithms to run with a low error ratio is the realization of the training process with a sufficient amount of data. Using synthetic data is one of the most common approaches when the data set is not enough for training. Synthetic data production must be based on a real dataset to improve the prediction and classification abilities of the deep learning algorithms. The enrichment of the existing dataset using different techniques such as modified copies of existing data is called data augmentation. It can sometimes be difficult to generate enough datasets according to the type of problem, especially in image classification. In such cases, a dataset can be generated by duplicating and/or modifying existing pictures of the objects. In this study, data augmentation for a learning-based vehicle make-model and license plate matching system has been performed and a new vehicle image dataset has been generated. The proposed approach which has been used in creating the dataset is presented in detail. The generated new vehicle image dataset is available to developers as open-source.

Keywords

• data augmentation
• deep learning
• dataset
• open-source
• synthetic data generation

References

1. F. Cen, X. Zhao, W. Li, G. Wang, Deep Feature Augmentation for Occluded Image Classification, Pattern Recognit. 111 (2020) 107737. https://doi.org/10.1016/j.patcog.2020.107737.
2. H.C. Shin, K. Il Lee, C.E. Lee, Data augmentation method of object detection for deep learning in maritime image, Proc. - 2020 IEEE Int. Conf. Big Data Smart Comput. BigComp 2020. (2020) 463–466. https://doi.org/10.1109/BigComp48618.2020.00-25.
3. H. Zheng, H. Shang, Z. Sun, X. Fu, J. Yao, J. Huang, Supervised Augmentation: Leverage Strong Annotation for Limited Data, Proc. - Int. Symp. Biomed. Imaging. 2020-April (2020) 1134–1138. https://doi.org/10.1109/ISBI45749.2020.9098607.
4. D. Zhao, G. Yu, P. Xu, M. Luo, Equivalence between dropout and data augmentation: A mathematical check, Neural Networks. 115 (2019) 82–89. https://doi.org/10.1016/j.neunet.2019.03.013.
5. A. Sakai, Y. Minoda, K. Morikawa, Data augmentation methods for machine-learning-based classification of bio-signals, BMEiCON 2017 - 10th Biomed. Eng. Int. Conf. 2017-January (2017) 1–4. https://doi.org/10.1109/BMEiCON.2017.8229109.
6. A. Mikołajczyk, M. Grochowski, Data augmentation for improving deep learning in image classification problem, 2018 Int. Interdiscip. PhD Work. IIPhDW 2018. (2018) 117–122. https://doi.org/10.1109/IIPHDW.2018.8388338.
7. J. Nalepa, G. Mrukwa, S. Piechaczek, P.R. Lorenzo, M. Marcinkiewicz, B. Bobek-billewicz, P. Wawrzyniak, P. Ulrych, J. Szymanek, M. Cwiek, W. Dudzik, M. Kawulok, M.P. Hayball, DATA AUGMENTATION VIA IMAGE REGISTRATION Future Processing , Gliwice , Poland Institute of Informatics , Silesian University of Technology , Gliwice , Poland Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology , Gliwice , Poland Feedba, (2019) 4250–4254.
8. H. Li, J. Rao, L. Zhou, J. Zhang, Valid data augmentation by patch alpha matting, 2019 IEEE 4th Int. Conf. Signal Image Process. ICSIP 2019. (2019) 361–366. https://doi.org/10.1109/SIPROCESS.2019.8868572.

9. J. Nalepa, M. Myller, M. Kawulok, Training- And Test-Time Data Augmentation for Hyperspectral Image Segmentation, IEEE Geosci. Remote Sens. Lett. 17 (2020) 292–296. https://doi.org/10.1109/LGRS.2019.2921011.
10. B. Huval, T. Wang, S. Tandon, J. Kiske, W. Song, J. Pazhayampallil, M. Andriluka, P. Rajpurkar, T. Migimatsu, R. Cheng-Yue, F. Mujica, A. Coates, A.Y. Ng, An Empirical Evaluation of Deep Learning on Highway Driving, (2015) 1–7. http://arxiv.org/abs/1504.01716.
11. A.T. Sasongko, G. Jati, M.I. Fanany, W. Jatmiko, Dataset of vehicle images for Indonesia toll road tariff classification, Data Br. 32 (2020) 106061. https://doi.org/10.1016/j.dib.2020.106061.
12. E. Osaba, Benchmark dataset for the Asymmetric and Clustered Vehicle Routing Problem with Simultaneous Pickup and Deliveries, Variable Costs and Forbidden Paths, Data Br. 29 (2020) 105142. https://doi.org/10.1016/j.dib.2020.105142.
13. W. Yang, Z. Li, C. Wang, J. Li, A multi-task Faster R-CNN method for 3D vehicle detection based on a single image, Appl. Soft Comput. J. 95 (2020) 106533. https://doi.org/10.1016/j.asoc.2020.106533.
14. J. Arun Pandian, G. Geetharamani, B. Annette, Data Augmentation on Plant Leaf Disease Image Dataset Using Image Manipulation and Deep Learning Techniques, Proc. 2019 IEEE 9th Int. Conf. Adv. Comput. IACC 2019. (2019) 199–204. https://doi.org/10.1109/IACC48062.2019.8971580.
15. N. Varela, C.G. Zoe, R. Ternera-Muñoz Yesith, F. Esmeral-Romero Ernesto, N.A.L. Zelaya, Method for classifying images in databases through deep convolutional networks, Procedia Comput. Sci. 175 (2020) 135–140. https://doi.org/10.1016/j.procs.2020.07.022.
16. Y. Bengio, Learning deep architectures for AI, Found. Trends Mach. Learn. 2 (2009) 1–27. https://doi.org/10.1561/2200000006.
17. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, R. Salakhutdinov, Dropout: A Simple Way to Prevent Neural Networks from Overfitting, Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. 15 (2018) 7642–7651. https://doi.org/10.1109/CVPR.2018.00797.
18. B. Ağgül, GitHub, (2020). https://github.com/burakaggul/Vehicle-brand-model-recognition-with-deep-learning-using-keras.
19. Craigslist.org, (2020). https://cfl.craigslist.org/.
20. Autoscout24, (2020). https://www.autoscout24.com.tr/.
21. J. Shijie, W. Ping, J. Peiyi, H. Siping, Research on data augmentation for image classification based on convolution neural networks, Proc. - 2017 Chinese Autom. Congr. CAC 2017. 2017-January (2017) 4165–4170. https://doi.org/10.1109/CAC.2017.8243510.
22. T.D. Pham, Geostatistical Simulation of Medical Images for Data Augmentation in Deep Learning, IEEE Access. 7 (2019) 68752–68763. https://doi.org/10.1109/ACCESS.2019.2919678.