Identification of fake currency using soft computing

Abstract

Advancements in color printing techniques have led to a surge in the creation and duplication of counterfeit currency notes, which has become prevalent in India. Although it was previously difficult to carry out this type of operation, anyone with a laser printer now has the capability to do it. This issue is considered a major problem due to the country's various problems, such as frauds, corruption and black money. It is needed to develop the system for the detection of fake currency. A system that can quickly identify counterfeit money is being developed. The proposed solution will utilize image processing to authenticate Indian currency notes. The proposed system will involve extracting different features from the notes, such as the security thread, Bleed Line, edges, shapes, textures, colors and Fluorescence. It is being developed using the software known as MATLAB. The supervised learning model of support vector machine for the extraction of the notes' features is also being used. This is utilized for analyzing the data related to classification and regression. Once the statistical features have been extracted, they were matched with the features stored in MAT file using SVM (Support Vector Machine) classifier. We have also conducted a test of other methods, such as the black box algorithm. The proposed solution is very fast and simple to implement, allowing it to identify if the notes are fake or real. The outcome of proposed system is the remarkable working of proposed system for the detection of fake currency and accuracy has been measured with the help of confusion matrix. The accuracy, precision & F- score are best by using SVM as compare to SVC (support Vector Classifier) & GBC (Genetic Bee Colony) algorithm.

Counterfeit Currency Detection Using Machine Learning

“…Employing image segmentation allows for the isolation of critical features, while quantitative metrics such as Mean Squared Error (MSE) and Structural Similarity Index (SSI) are utilized to evaluate system performance, offering valuable insights for currency management [5].…”

Soft computing approach for feature extraction of palm biometric

“…In the past few years, the field of automated identification systems has expanded significantly. These include various functions such as validating personal identity, preventing identity fraud, and security checking (Shinde et al 2023a). Biometric identification systems, which are becoming more prevalent due to advancements in biotechnology, are expected to be userfriendly and accurate.…”