A Visually Impaired Mobile Application for Currency Recognition using MobileNetV2 CNN Architecture
Journal of Computing and Social Informatics · 10.03.2025 · 4 (1) · 26-42
DOI: 10.33736/jcsi.8562.2025 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Computing and Social Informatics, online erschienen am 10.03.2025. Verfasst von Abraham Eseoghene Evwiekpaefe und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
There are at least 2.2 billion people with visual impairment globally of which almost half of these cases could have been addressed or prevented. Visual impairment has both personal and economic impacts on individuals. It impacts negatively on the quality of life, especially among adults. Many visually impaired persons in our society today face a lot of challenges and one of these challenges is object recognition. The visually impaired persons need assistance so they can perform monetary transactions without being cheated. This work is aimed at developing a model for the recognition of Nigerian naira notes for visually impaired persons. The model was trained using the concept of transfer learning with a trainable layer built on the MobileNetV2 convolutional neural network architecture pre-trained model using python programming language on Spyder anaconda IDE. The model was saved and converted to TensorFlow lite format which was deployed into a mobile application coded in the java programming language in android studio. A total of 3615 image datasets were collected, including N5, N10, N20, N50, N100, N200, N500, and N1000 denominations and some random images of objects that constitute the non-currency class for the training of the model. The collected data was divided into 80% for training and 20% for testing. The model achieved an accuracy of 98%.
Zitieren
APA
Evwiekpaefe, A. E., & Isacha, H. (2025). A Visually Impaired Mobile Application for Currency Recognition using MobileNetV2 CNN Architecture. Journal of Computing and Social Informatics, 4(1), 26-42. https://doi.org/10.33736/jcsi.8562.2025
Vancouver
Evwiekpaefe AE, Isacha H. A Visually Impaired Mobile Application for Currency Recognition using MobileNetV2 CNN Architecture. Journal of Computing and Social Informatics. 2025;4(1):26-42. doi:10.33736/jcsi.8562.2025
BibTeX
@article{Evwiekpaefe2025,
author = {Evwiekpaefe, Abraham Eseoghene and Isacha, Habila},
title = {A Visually Impaired Mobile Application for Currency Recognition using MobileNetV2 CNN Architecture},
journal = {Journal of Computing and Social Informatics},
year = {2025},
volume = {4},
number = {1},
pages = {26--42},
doi = {10.33736/jcsi.8562.2025}
}
EUVIMED