DeViTC: Deep-Vision Transformer to Recognize Originality of Currency
This paper presents a deep learning-based banknote recognition system using the Vision Transformer (ViT) architecture. Evaluated on Indian currency and four datasets, the ViT’s attention mechanism enhances accuracy with data augmentation. Comparisons with ResNet, VGG, GoogleNet, and EfficientNet demonstrate its robustness, aiding financial inclusion.
Authors
- Nikita Bhatt (ORCID: https://orcid.org/0000-0002-3243-5901)
- Aayushi Chaudhari
- Deep Kothadiya (ORCID: https://orcid.org/0000-0002-5997-1720)
- Dulari Gajjar
- Prisha Faldu
Institutions
- Charotar University of Science and Technology (IN)
Publication Details
- Journal
- Computer
- Published
- 2025-04-18
- DOI
- https://doi.org/10.1109/mc.2024.3514151
- Citations
- 2
- Primary Topic
- Currency Recognition and Detection
- Type
- article
- Field-Weighted Citation Impact
- 2.55