Authenticity Verification of Indian Notes via CNN

MCA

Advances in color-printing technology have made large-scale duplication of banknotes increasingly feasible. Although digital payments are growing and paper-cash usage has dipped in recent years, currency notes remain widely circulated because they are reliable and simple to use. What once required specialized print shops can now be approximated at home with consumer laser printers, raising the prevalence of counterfeits in circulation. India, which already contends with issues such as corruption and unaccounted cash, faces a persistent challenge from fake notes. To address this, we propose a deep-learning solution for authenticating Indian banknotes. A MATLAB-based implementation is used to classify each note as genuine or counterfeit. Counterfeiting—the unauthorized reproduction of lawful currency—undermines trust and economic stability. In India, the Reserve Bank of India (RBI) is the sole authority for printing notes, yet once counterfeits enter circulation, they must be identified and removed continually. Rapid improvements in consumer printing and scanning have amplified the threat, devalued legitimate currency and stressed detection workflows. Traditional approaches that depend on hardware add-ons and handcrafted image-processing pipelines tend to be labor-intensive and less reliable at scale. To overcome these limitations, we propose an Xception-based convolutional neural network that analyzes currency images and learns discriminative features—such as security-thread characteristics—directly from data. The system targets ₹500 and ₹2000 denominations, delivering efficient, accurate screening of forged notes from captured images.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.