Online Handwritten Signature Verification: The State of the Art

Abstract

Handwritten signature is the most common method for biometric verification. The purpose of this research is to analyze the existing approaches to the implementation of the algorithm for verification of handwritten signatures. Existing researches use various technologies, such as neural network, hidden Markov model and SVM algorithm, to solve the task of signature verification, and they are constantly introducing new ideas, concepts and algorithms. Signature verification is a real challenge for researchers due to many difficulties that may arise in the process of creating such system. The most promising algorithm will form the basis of the developed authentication system based on a handwritten signature.