The Impact of Data Augmentation Techniques on the Recognition of Script Images in Deep Learning Models

Authors

  • Wulan Sapitri Intelligent Systems Research Group, Faculty of Sains Technology, Universitas Bina Darma; Information Systems Department, Faculty of Science and Technology, Universitas Bina Darma, Indonesia
  • Yesi Novaria Kunang Intelligent Systems Research Group, Faculty of Sains Technology, Universitas Bina Darma; Information Systems Department, Faculty of Science and Technology, Universitas Bina Darma, Indonesia
  • Ilman Zuhri Yadi Intelligent Systems Research Group, Faculty of Sains Technology, Universitas Bina Darma; Information Systems Department, Faculty of Science and Technology, Universitas Bina Darma, Indonesia
  • Mahmud Mahmud Software Developer, OXY creative. Inc, Jakarta, Indonesia

DOI:

https://doi.org/10.15575/join.v8i2.1073

Keywords:

CNN, Deep Learning, Komering Script, Augmentation, Recognition

Abstract

Deep learning technology is widely used for recognizing character images, including various regional characters and diverse ancient scripts. Deep learning models require large data sets to recognize images accurately. However, creating a dataset has limitations in terms of quantity, including the Komering script dataset used in this study. Data augmentation techniques can be applied to expand the dataset by modifying existing images to increase data diversity. This study aims to investigate the impact of augmentation techniques on the performance of deep learning models in the case of Komering script recognition. The dataset consists of 500 images for five classes of Komering script characters. Three augmentation techniques, namely random rotation, height shift, and width shift, were applied to the five characters, which were then used to test the model trained to recognize characters in the Komering dataset. This research contributes to providing insights into the effect of augmentation techniques on robust confidence prediction of deep learning models for recognizing newly augmented data. The results demonstrate that the deep learning model can recognize modified data using augmentation techniques with an average accuracy of 80.05%.

References

M. Mahmud and Y. N. Kunang, “Pengembangan Aplikasi Pengenalan Aksara Komering Menggunakan Metode Deep Learning Berbasis Android,” J. Comput. Inf. Syst. Ampera, vol. 1, no. 2, pp. 99–111, May 2020, doi: 10.51519/journalcisa.v1i2.38.

Y. N. Kunang, I. Z. Yadi, Mahmud, and M. Husin, “A New Deep Learning-Based Mobile Application for Komering Character Recognition,” in 2022 5th International Seminar on Research of Information Technology and Intelligent Systems (ISRITI), May 2022, pp. 294–299. doi: 10.1109/ISRITI56927.2022.10053072.

Y. Sutal, R. K. Pingak, A. S. Ahab, and S. D. Baunsele, “Sistem Identifikasi Nominal Uang Logam Menggunakan Tensorflow dan Convolutional Neural Networkberbasis Raspberry Pi,” Kaji. Awal Ekstraksi Silika Dari Pasir Noeltoko Menggunakan X-Ray Fluoresc., no. June, pp. 278–289, 2019, [Online]. Available: https://www.conference.undana.ac.id/sainstek/article/view/84/70

P. R. Nesbit and C. H. Hugenholtz, “Enhancing UAV-SfM 3D model accuracy in high-relief landscapes by incorporating oblique images,” Remote Sens., vol. 11, no. 3, pp. 1–24, 2019, doi: 10.3390/rs11030239.

E. D. Cubuk, B. Zoph, J. Shlens, and Q. V. Le, “Autoaugment,” Proc. IEEE Int. Conf. Comput. Vis., vol. 2019-June, no. Section 3, pp. 113–123, 2019.

C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, vol. 6, no. 1, 2019, doi: 10.1186/s40537-019-0197-0.

Z. Pei, H. Xu, Y. Zhang, M. Guo, and Y. Yee-Hong, “Face recognition via deep learning using data augmentation based on orthogonal experiments,” Electron., vol. 8, no. 10, pp. 1–16, 2019, doi: 10.3390/electronics8101088.

M. Elgendi et al., “The Effectiveness of Image Augmentation in Deep Learning Networks for Detecting COVID-19: A Geometric Transformation Perspective,” Front. Med., vol. 8, no. March, pp. 1–12, 2021, doi: 10.3389/fmed.2021.629134.

Q. Zheng, M. Yang, X. Tian, N. Jiang, and D. Wang, “A full stage data augmentation method in deep convolutional neural network for natural image classification,” Discret. Dyn. Nat. Soc., vol. 2020, 2020, doi: 10.1155/2020/4706576.

Q. Wen et al., “Time Series Data Augmentation for Deep Learning: A Survey,” IJCAI Int. Jt. Conf. Artif. Intell., pp. 4653–4660, 2021, doi: 10.24963/ijcai.2021/631.

M. Kim and C. Y. Jeong, “Label-preserving data augmentation for mobile sensor data,” Multidimens. Syst. Signal Process., vol. 32, no. 1, pp. 115–129, 2021, doi: 10.1007/s11045-020-00731-2.

A. Cohen, I. Rimon, E. Aflalo, and H. H. Permuter, “A study on data augmentation in voice anti-spoofing,” Speech Commun., vol. 141, pp. 56–67, 2022, doi: 10.1016/j.specom.2022.04.005.

C. Tang, K. Ma, B. Cui, K. Ji, and A. Abraham, “Long text feature extraction network with data augmentation,” Appl. Intell., vol. 52, no. 15, pp. 17652–17667, 2022, doi: 10.1007/s10489-022-03185-0.

P. Chen, S. Liu, H. Zhao, and J. Jia, “GridMask Data Augmentation,” 2020, [Online]. Available: http://arxiv.org/abs/2001.04086

X. Guo and Q. Hu, “Low-light Image Enhancement via Breaking Down the Darkness,” Int. J. Comput. Vis., vol. 131, no. 1, pp. 48–66, 2023, doi: 10.1007/s11263-022-01667-9.

W. Wang, X. Wu, X. Yuan, and Z. Gao, “An Experiment-Based Review of Low-Light Image Enhancement Methods,” IEEE Access, vol. 8, pp. 87884–87917, 2020, doi: 10.1109/ACCESS.2020.2992749.

Z. Zhong, L. Zheng, G. Kang, S. Li, and Y. Yang, “2017(cites:1625)Random Erasing Data Augmentation,” 2017.

A. Wikarta, B. Suryo, and M. Khoirul Effendi, “Analisa Pengaruh Ukuran Testing Data dan Data Augmentation pada Tingkat Akurasi Deteksi Pemakaian Masker oleh Pengemudi Kendaraan menggunakan Deep Learning Influence,” Sent. 2020, no. November 2020, pp. 20–24, 2020.

J. Sanjaya and M. Ayub, “Augmentasi Data Pengenalan Citra Mobil Menggunakan Pendekatan Random Crop, Rotate, dan Mixup,” J. Tek. Inform. dan Sist. Inf., vol. 6, no. 2, pp. 311–323, 2020, doi: 10.28932/jutisi.v6i2.2688.

S. F. Tumewu, D. H. Setiabud, and I. Sugiarto, “Klasifikasi Motif Batik menggunakan metode Deep Convolutional Neural Network dengan Data Augmentation,” J. Infra, vol. 8, no. 2, pp. 189–194, 2020.

R. Z. Fadillah, A. Irawan, M. Susanty, and I. Artikel, “Data Augmentasi Untuk Mengatasi Keterbatasan Data Pada Model Penerjemah Bahasa Isyarat Indonesia (BISINDO),” J. Inform., vol. 8, no. 2, pp. 208–214, 2021, [Online]. Available: https://ejournal.bsi.ac.id/ejurnal/index.php/ji/article/view/10768

Y. N. Kunang, “Komering_(Ulu)_Characters_Datases.” Intelligent System Research Group Bina Darma Dataset, 2022. [Online]. Available: https://www.kaggle.com/datasets/ykunang/aksara-komering

“tf.keras.preprocessing.image.ImageDataGenerator.” Keras, 2022. [Online]. Available: https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator

Downloads

Published

2023-12-28

Issue

Section

Article

Citation Check