Counterfeit Buster: Indian Currency Detection Using Generative Adversarial Networks

Abstract

Counterfeit currency undermines financial trust and economic stability, particularly in cash- dependent regions. As counterfeiters adopt increasingly sophisticated techniques, forged notes can evade visual inspection, rendering traditional detection methods unreliable. Furthermore, existing verification techniques are typically hardware-dependent, expensive, or too complex for general users, making them impractical and inaccessible in real-world applications. To overcome these limitations, we introduce CounterfeitBuster, an Android application that leverages Generative Adversarial Networks (GANs) for real-time detection of counterfeit Indian currency. GANs operate by training two neural networks—a generator that produces synthetic data and a discriminator that learns to distinguish it from real data—thereby enhancing detection accuracy through continuous learning. The app allows users to scan currency using their smartphone camera and receive instant results, offering a portable, affordable, and user-friendly solution. Additional features such as denomination recognition, real-time currency conversion, and voice feedback further improve accessibility and usability, empowering users to effectively combat currency fraud.

Introduction

Counterfeit Currency Detection: Overview

Counterfeit currency remains a major global threat, causing economic losses and undermining trust. Advances in printing have made forged notes hard to detect using traditional hardware-based methods like UV scanning and watermark checks, which require specialized equipment and personnel, limiting accessibility and scalability.

Need for Intelligent Solutions

To overcome these limitations, the integration of deep learning techniques, particularly Generative Adversarial Networks (GANs) and Convolutional Neural Networks (CNNs), offers a powerful alternative. GANs generate synthetic counterfeit images to enrich training data, while CNNs excel at detecting subtle visual differences, enabling accurate counterfeit detection.

Literature Insights

Research shows CNNs outperform traditional methods in image classification tasks, with hybrid GAN-CNN models further improving detection accuracy and dataset diversity. Challenges include image quality, lighting variations, and dataset imbalance, but ongoing advances promise more robust, scalable solutions.

Limitations of Existing Systems

Conventional detection systems are expensive, stationary, and mostly restricted to banks or large retailers, inaccessible to everyday users like shopkeepers or rural populations. They lack mobile integration, hindering on-the-go validation.

Proposed System: Counterfeit Buster App

An Android app called Counterfeit Buster is proposed to enable easy, real-time verification of Indian currency notes (?100, ?200, ?500) via smartphone cameras. Key features include:

• Automated counterfeit detection with instant visual and voice feedback.

• Automatic denomination recognition.

• Currency conversion based on live exchange rates.

• Accessibility features like voice output for visually impaired users.

GAN Architecture

The system uses a GAN where the Generator creates fake currency images and the Discriminator learns to differentiate them from genuine ones through an adversarial training process, improving the model’s ability to detect counterfeits.

Technology Stack

• Python and TensorFlow for model development.

• NumPy and Matplotlib for data handling and visualization.

• Android Studio and Java for app development.

• Google Colab for cloud-based training.

• TensorFlowLite to run models efficiently on mobile devices.

Results

The app demonstrates effective detection of real and counterfeit notes, recognition of denominations, and user-friendly interface features suitable for widespread public use.

Conclusion

Counterfeit Buster delivers a practical and inclusive solution for detecting counterfeit Indian currency by leveraging the power of Generative Adversarial Networks alongside deep learning techniques. Designed as an Android application, it ensures wide accessibility for the general public while offering reliable currency validation through a combination of visual recognition and audio feedback. Its added capabilities of recognizing denominations and converting currency values enhance its utility beyond simple detection. With further refinements, such as support for additional denominations and real-time exchange updates, Counterfeit Buster has the potential to become an essential tool for individuals and small businesses in safeguarding against currency fraud.

References

[1] K. Bhushanm, M. Asritha, P. R. Sultana, P. Anil Kumar, and S. Mahesh Babu, “Fake Currency Detection using Deep Learning,” International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, pp. 261–269, Mar. 2024. [2] V. Shokeen, S. Kumar, A. Sharma, N. Singh, A. Prasad, and Y. Sachdeva,“Detection of Counterfeit Currency Notes Through Machine Learning Algorithms and Image Processing,” International Journal of Engineering Research and Technology, vol. 12, no. 5, pp. 1023–1030, May 2023. [3] Y. Wang, “Currency Recognition Based on Deep Learning,” Journal of Computer Vision and Pattern Recognition, vol. 8, no. 2, pp. 145–152, Apr. 2023. [4] S. Desai, A. Rajadhyaksha, A. Shetty, and S. Gharat, “Counterfeit Indian Currency Recognition Using Generative Adversarial Networks,” Proceedings of the International Conference on Machine Learning and Data Engineering, pp. 210–215, Dec. 2021. [5] S. N. Kumar, G. Singal, S. Sirikonda, and R. Nethravathi, “A Novel Approach for Detection of Counterfeit Indian Currency Notes Using Deep Convolutional Neural Network,” International Journal of Recent Technology and Engineering, vol. 9, no. 1, pp. 456–462, May 2020. [6] R. Sharan and P. Kaur, “Detection of Fake Currency Using Image Processing in MATLAB,” International Journal of Engineering and Techniques, vol. 5, no. 1, pp. 12–17, 2019. [7] A. Bhatia, M. Kumar, V. Kedia, B. K. Shah, A. Shroff, and Aryan, “Fake Currency Detection with Machine Learning Algorithm and Image Processing,” International Journal of Scientific Research in Computer Science, Engineering and Information Technology, vol. 7, no. 2, pp. 98–104, Feb. 2021. [8] M. Laavanya and V. Vijayaraghavan, “Real Time Fake Currency Note Detection using Deep Learning,” International Journal of Engineering and Advanced Technology, vol. 9, no. 1S5, pp. 2249–8958, Dec. 2019. [9] R. K. Yadav, P. Valecha, and S. Paliwal, “Counterfeit Currency Detection Using Supervised Machine Learning Algorithms,” in Machine Learning for Predictive Analysis, A. Joshi, M. Khosravy, and N. Gupta, Eds. Springer, 2021, pp. 123–134. [10] A. Saxena, “Fake Currency Detection using Image Processing,” International Journal of Trend in Research and Development, vol. 4, no. 1, pp. 2394–9333, Jan. 2018.

Copyright

Copyright © 2025 Hrutali Dhawade, Ankita Sonawane, Prof. Kumud Wasnik. This 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.