Indian Counterfeit Banknote Detection Using Support Vector Machine
5 Pages Posted: 7 Apr 2020
Date Written: April 8, 2020
Abstract
Currency counterfeiting is a serious crime that affects a country's finances. The proposed system will be useful to detect counterfeit banknotes in banking systems. India is facing more serious problems due to the increase in fake notes in the market. To get rid of this problem, various fake note detection methods are available worldwide, but most of them are hardware based and costly. The proposed system focuses on having access to the public to identify counterfeit banknotes. The proposed system can identify the legitimacy of a banknote by checking for specific security features such as watermarks, latent images, security threads, etc. Identification of counterfeit banknotes is done using machine learning techniques. The methodology involves extracting and encoding these security features. Security features are extracted from the input image, feature detection and classification are performed using a support vector machine (SVM). Experimental results shown in the paper shows promising results.
Keywords: Counterfeit currency, Artificial Neural Network (ANN), Support Vector Machine (SVM), Region of Interest (ROI), Edge Detection, Artificial Intelligence (AI), Histogram of Oriented Gradient (HOG), Image Processing, Machine Learning (ML), Deep Learning (DL)
Suggested Citation: Suggested Citation