A Comparative Analysis of Signature Recognition Methods

Introduction

Biometric systems use the characteristics of each individual that are unique for every individual. Unlike past verification systems which used keys and passwords were prone to get stolen or forgotten, the physical characteristics that are used in biometrics systems cannot be easily stolen, forgotten or transferred from one person to another (Hezil et al., 2018; Khanday et al., 2018).The objective of such systems is to recognize individuals based on physiological or behavioural traits. Physiological traits are based on measurements of behavioral traits, such as the gestures, signature etc and Behavioral traits are based on measurements of biological traits, such as the fingerprint, face, iris, etc. There are two steps that are done in biometric systems: verification and identification. In the verification phase, a user claims its identity to the system by giving a sample of its biometric identity to the system. In the verification system, the main purpose is to check whether the user is an authorized user or a forger. In the identification phase, the biometric sample is checked among all the registered users in the system (Malik & Arora, 2015). To verify a person’s identity in legal, financial and administrative areas, the most important biometric trait used mostly in documentation verification is the handwritten signature.  In our daily life, people are very familiar with the signature recognition systems because the process of collection of handwritten signatures is non-invasive. Signature verification systems are used to automatically discriminate if the biometric sample of the user is indeed of a claimed individual or a forger. In other words, the verification is done through these systems. Numerous growing areas of research is the main issue that comes along with personal verification and identification. The different personal characteristics of the individual are used by the system such as face, odor, gait, iris, voice, lip movements, hand geometry, retina. Most commonly used method in personal identification and verification is done through psychological or behavioral characteristics of the individual. There were many traditional authentication techniques which included  passwords, PIN numbers, smartcards etc which were replaced by the biometric systems due to the reason that biometric characteristics of the individual cannot be copied by anyone, and cannot be lost , looted or damaged and cannot be easily transferred to anyone and are uniquely identified for every person. There are several factors on which the biometric system depends: User acceptance, Required security, Precision, Implementation and cost time 

  There are different ways to check the validity of one’s personal data, either using signature or fingerprint. Signature refers to the symbol or a sign of the name written by the hand .In our daily lives such as schools, banks, corporations, hospitals, and government departments etc, signatures are often used in verification. There are some fraudulent parties who want to manipulate the signatures of others for illegal use, Due to the importance of signature acceptance. Duplicate signatures can be detrimental and included in the criminal realm (Kekre & Bharadi, 2010). Fingerprints scanning and Retinal vascular pattern screening were some of the electronic identification methods in comparison with Signature verification  methods and signature verification methods is found as one of the most accepted ways to identify the person’s identity. As the primary means of identifying (verifying and authenticating) using the signature of a person , the user that provides the written signature  is based on the assumption that the signature changes slowly and cannot be altered , forged, copied without proper detection. Using the signature verification systems, it was easier for people to migrate from popular pen paper signature systems to the system where handwritten signatures are captured and verified electronically. Signature recognition is categorized as a behavioral tool which is used in banking, credit card validation, security systems, offices etc. In nutshell, handwritten signature verification can be categorized into two broad categories – on–line verification and off–line verification.