Effect of Euler’s Number on Signature Verification System

Recently, renewed interest has grown in handwritten signature verification domain. In order to make personal verification, handwritten signature verification is the most widely used method. Such automated signature verification frameworks have wide range of applications. This work presents an innovative approach for classification between real and forged signatures using machine learning classification models. Two individual image datasets of real and forge signatures are used in this current study. Six features (mean, standard deviation, skewness, kurtosis, entropy and Euler’s Number) of the two datasets were calculated. The current work used classification algorithms such as K-Nearest Neighbors (KNN), Logistic Regression, Random Forest (RF) Classifier. The study was done to investigate the effect of Euler number in signature verification system, hence the framework recorded results with and without Euler’s Number. The classification report was further analyzed to find out the superior framework out of the three frameworks chosen and also to identify the importance of Euler’s name in signature verification system.