A novel GAN-transformer framework for early Brahmi script generation and recognition
Engineering and Applied Science Research · 11.03.2026 · 53 (2) · 112-126
DOI: 10.64960/easr.2026.261416 ↗Kurz gefasst
Fachartikel in der Zeitschrift Engineering and Applied Science Research, online erschienen am 11.03.2026. Verfasst von Pabasara Surasinghe und 1 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Recognizing ancient scripts is crucial for understanding the historical, cultural, and linguistic context of past civilizations. However, the recognition of Early Brahmi letters from Sri Lankan inscriptions faces significant challenges due to scarce digitized data, degradation of inscriptions, and visual similarity among characters. This study introduces BrahmiGAN, a novel Generative Adversarial Network (GAN) designed to generate realistic synthetic Early Brahmi letters, addressing data limitations that hinder recognition model training. Using a benchmark dataset of 73 inscriptions, 21,195 high-quality synthetic images were generated from 888 real samples. These synthetic images demonstrated high fidelity, validated through feature-based, raster-based, and vector-based evaluations, and achieved a 92.15% approval rate from human experts. Furthermore, a vision Transformer-based ensemble model integrating Pyramid Vision Transformer and Swin Transformer is proposed for Early Brahmi letter recognition. A classification accuracy of 96.06% was attained by the ensemble model when trained on combined synthetic and real images, outperforming existing methods and surpassing the same model trained exclusively on real images. The generated dataset is publicly available to support future research (https://zenodo.org/records/14961074).
Zitieren
APA
Surasinghe, P., & Thanikasalam, K. (2026). A novel GAN-transformer framework for early Brahmi script generation and recognition. Engineering and Applied Science Research, 53(2), 112-126. https://doi.org/10.64960/easr.2026.261416
Vancouver
Surasinghe P, Thanikasalam K. A novel GAN-transformer framework for early Brahmi script generation and recognition. Engineering and Applied Science Research. 2026;53(2):112-126. doi:10.64960/easr.2026.261416
BibTeX
@article{Surasinghe2026,
author = {Surasinghe, Pabasara and Thanikasalam, Kokul},
title = {A novel GAN-transformer framework for early Brahmi script generation and recognition},
journal = {Engineering and Applied Science Research},
year = {2026},
volume = {53},
number = {2},
pages = {112--126},
doi = {10.64960/easr.2026.261416}
}
EUVIMED