Research Article
Counterfeit Currency Detection Leveraging MobileNet and ResNet Models
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357812, author={Naga Prabhakar Ejaru and Sai Pranitha P and Thanmai S and Saisasi G and Chennakesava P}, title={Counterfeit Currency Detection Leveraging MobileNet and ResNet Models }, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={counterfeit mobilenet resnet image processing and streamlit}, doi={10.4108/eai.28-4-2025.2357812} }- Naga Prabhakar Ejaru
Sai Pranitha P
Thanmai S
Saisasi G
Chennakesava P
Year: 2025
Counterfeit Currency Detection Leveraging MobileNet and ResNet Models
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357812
Abstract
Today the world economy is suffering from severe problems due to the growing fake currency notes circulation that reduces faith in monetary systems, and raises financial instability. Instant fake currency detection is being a challenging task for development of innovative solutions especially due to the diversified fake currency notes. Physical inspection is the predominant technique in traditional systems of counterfeit currency detection. Real or fake notes detection using CNN-based architectures Using CNN based technology, MobileNet achieved the highest accuracy of 96.03% and ResNet provided the lowest accuracy of 74.03%. Further, the system may recognize the circulation monetary note and currency image denomination. This can be achieved by applying image augmentation and preprocessing techniques to enhance model performance. In addition, the interface of the real-time currency detection is built on with Streamlit, providing human-friendly and convenient platform for people meeting the need of counterfeit checking on the spot by themselves.
