EPISTEMOLOGY OF SMART MONEY CLASSIFIER & LOCATION TRACKER APPLICATION BASED ON CONVOLUTIONAL NEURAL NETWORK (CNN) FOR THE BLIND
JOURNAL OF SCIENCE AND APPLIED ENGINEERING · 12.05.2026 · 9 (1) · 1-8
DOI: 10.31328/jsae.v9i1.7378 ↗Kurz gefasst
Fachartikel in der Zeitschrift JOURNAL OF SCIENCE AND APPLIED ENGINEERING, online erschienen am 12.05.2026. Verfasst von Lis Diana Mustafa und 3 weiteren Autor:innen.
Abstract (Originalsprache)
The number of blind people in Indonesia continues to increase every year due to congenital and degenerative factors. This reality places Indonesia as the country with the second-highest number of blind people in the world after Ethiopia. Visual disabilities (blindness) cause significant functional obstacles in daily life activities, one of which is in the domain of financial transactions, especially the inability to independently and accurately distinguish the value of money, which leads to a loss of Financial Autonomy and makes one a target of fraud and transaction errors. This condition demands technological innovation that is not only functional but also grounded in strong scientific principles. This study examines the development of the Smart Money Classifier & Location Tracker application based on Convolutional Neural Network (CNN) from the perspective of how knowledge is generated, validated, and used in data-based systems. The method used to develop this application is to apply an artificial intelligence-based approach, a Convolutional Neural Network (CNN), to the process of classifying currency images, as well as to utilize GPS sensors for location tracking. The process begins by collecting and labelling currency image data, then by training a CNN to recognise various currency denominations accurately. The trained model is integrated into the Android application using a machine learning framework (such as TensorFlow Lite) to run efficiently on the user's device. Furthermore, GPS on Android provides real time location information, assisting the visually impaired with orientation and navigation. The test results show that this application can classify the nominal value of money with 90% accuracy, instilling confidence in users and their families, and can be installed on various Android phone brands, supporting independence. It also records user location tracking, accessible by family or companions, reinforcing safety and trust.
Zitieren
APA
Mustafa, L. D., Dali, S. W., Wahidah, K., & Patmanthara, S. (2026). EPISTEMOLOGY OF SMART MONEY CLASSIFIER & LOCATION TRACKER APPLICATION BASED ON CONVOLUTIONAL NEURAL NETWORK (CNN) FOR THE BLIND. JOURNAL OF SCIENCE AND APPLIED ENGINEERING, 9(1), 1-8. https://doi.org/10.31328/jsae.v9i1.7378
Vancouver
Mustafa LD, Dali SW, Wahidah K, Patmanthara S. EPISTEMOLOGY OF SMART MONEY CLASSIFIER & LOCATION TRACKER APPLICATION BASED ON CONVOLUTIONAL NEURAL NETWORK (CNN) FOR THE BLIND. JOURNAL OF SCIENCE AND APPLIED ENGINEERING. 2026;9(1):1-8. doi:10.31328/jsae.v9i1.7378
BibTeX
@article{Mustafa2026,
author = {Mustafa, Lis Diana and Dali, Sri Wahyuni and Wahidah, Khoirunnisa and Patmanthara, Syaad},
title = {EPISTEMOLOGY OF SMART MONEY CLASSIFIER & LOCATION TRACKER APPLICATION BASED ON CONVOLUTIONAL NEURAL NETWORK (CNN) FOR THE BLIND},
journal = {JOURNAL OF SCIENCE AND APPLIED ENGINEERING},
year = {2026},
volume = {9},
number = {1},
pages = {1--8},
doi = {10.31328/jsae.v9i1.7378}
}
EUVIMED