Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{181834,
        author = {PARUCHURI VENKATA SUDHEER and T. Y. S. SRI HARI and B. JAYAKRISHNA},
        title = {Counterfeit (Fake) Currency Detection using Mobilenet and Resnet Models},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {2},
        pages = {308-317},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=181834},
        abstract = {The global economy is seriously threatened by the growing problem of counterfeit currency circulation which erodes confidence in monetary systems and causes financial instability. Real-time fake currency detection is a difficult challenge that requires creative solutions especially given the variety of counterfeit currency notes. Physical examination is a major component of traditional counterfeit detection techniques but it can be laborious, error-prone and non-scalable. In order to solve this issue, this work builds an automated system for detecting counterfeit currencies using MobileNetV2 and ResNet designs and sophisticated deep learning techniques. To increase the robustness of the model the dataset which was gathered from several internet sources is enhanced using a variety of data augmentation approaches. The main objective of the model is to correctly identify genuine or counterfeit money notes. When identifying real or fake currencies using CNN-based structures MobileNetV2 obtained an astonishing 99.03% accuracy whereas ResNet achieved 74.03%. Additionally, the system may identify the denomination of notes of currency such as 10, 20, 50, 100, 200, 500 and non-currency Images. Using Image Augmentation and preprocessing methods for improved model performance allows for this. Furthermore, the system makes use of streamlit as the real-time currency detection user interface providing a user-friendly and easily accessible platform that enables users to instantaneously verify the legitimacy of currency notes. The method is an efficient way to detect counterfeit currency because of its 94% overall accuracy in identifying both forged notes and their corresponding denominations.},
        keywords = {Counterfeit Detection, MobileNetV2, ResNet-50, Currency Classification, Image Processing and Streamlit UI.},
        month = {July},
        }