From Local Features to Global Context: Comparing CNN and Transformer for Sundanese Script Classification
Journal of Intelligent Systems Technology and Informatics · 2025
DOI: 10.64878/jistics.v1i2.38 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Systems Technology and Informatics, 2025. Verfasst von Yoga Agustiansyah und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The digital preservation of historical writing systems like Aksara Sunda is critical for cultural heritage, yet automated recognition is hindered by high character similarity and handwriting variability. This study systematically compares two dominant deep learning paradigms, Convolutional Neural Networks (CNNs) and Transformers, to evaluate the crucial trade-off between model accuracy and real-world robustness. Using a transfer learning approach, we trained five models (ResNet50, MobileNetV2, EfficientNetB0, ViT, and DeiT) on a balanced 30-class dataset of Sundanese script. Performance was assessed on a standard in-distribution test set and a challenging, independently collected Out-of-Distribution (OOD) dataset designed to simulate varied real-world conditions. The results reveal a significant performance inversion. While EfficientNetB0 achieved the highest accuracy of 96.9% on in-distribution data, its performance plummeted on the OOD set. Conversely, ResNet50, despite being lower in in-distribution accuracy, proved to be the most robust model, achieving the highest accuracy of 92.5% on the OOD data. This study concludes that for practical applications requiring reliable performance, the generalization capability demonstrated by ResNet50 is more valuable than the specialized accuracy of EfficientNetB0, offering a crucial insight for developing robust digital preservation tools for historical scripts.
Zitieren
APA
Agustiansyah, Y., & Fauzi, D. R. (2025). From Local Features to Global Context: Comparing CNN and Transformer for Sundanese Script Classification. Journal of Intelligent Systems Technology and Informatics. https://doi.org/10.64878/jistics.v1i2.38
Vancouver
Agustiansyah Y, Fauzi DR. From Local Features to Global Context: Comparing CNN and Transformer for Sundanese Script Classification. Journal of Intelligent Systems Technology and Informatics. 2025. doi:10.64878/jistics.v1i2.38
BibTeX
@article{Agustiansyah2025,
author = {Agustiansyah, Yoga and Fauzi, Dhika Restu},
title = {From Local Features to Global Context: Comparing CNN and Transformer for Sundanese Script Classification},
journal = {Journal of Intelligent Systems Technology and Informatics},
year = {2025},
doi = {10.64878/jistics.v1i2.38}
}
EUVIMED