A New Proposal for Score Normalization in Biometric Signature Recognition Based on Client Threshold Prediction

Vivaracho-Pascual, Carlos; Simon-Hurtado, Arancha; Manso-Martinez, Esperanza; Pascual-Gaspar, J. M.

doi:10.1109/icdm.2012.50

Cited by 3 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A new proposal for score normalization in biometric signature recognition based on client threshold prediction was proposed by Pascual et al [51]. The use of score normalization in biometric based recognition system is a very important part, particularly in those based on behavioural traits, such as written signature.…”

Section: Single-script Signature-based Biometric Authenticationmentioning

confidence: 99%

Signature-Based Biometric Authentication

Pal

Blumenstein

2014

Studies in Computational Intelligence

View full text Add to dashboard Cite

In a modern, civilized and advanced society, reliable authentication and authorization of individuals are becoming more essential tasks in several aspects of daily activities and as well as many different important applications such as in financial transactions, access control, travel and immigration, healthcare etc. In some situations, when individual equipment is required for confirmation of one's identity to other groups of people in order to make use of services or to achieve access to physical places, it is always necessary to declare self-identity and to prove the claim. Traditional authentication methods, which are based on knowledge (password-based authentication) or the utility of a token (photo ID cards, magnetic strip cards and keybased authentication), are less reliable because of loss, forgetfulness and theft. These issues direct substantial attention towards biometrics as an alternative method for person authentication and identification. The word 'biometric' has been derived from the Greek words "Bio-metriks", "Bio" which means life and "metriks" which means measures. Therefore a biometric is the measurement and statistical analysis of unchanging biological characteristics. Biometrics evaluate a person's unique physical or behavioural traits to authenticate their identity. As biometric identifiers are unique to persons, they are more reliable in verifying identity than token-based and knowledge-based methods. In the last few years, substantial efforts have been devoted to the development of biometric-based authentication systems. Biometrics provide an expected and successful solution to the authentication problem, as it offers the construction of systems that can identify individuals by the analysis of their physiological or behavioural characteristics [1]. In fact, the field of biometrics is the science of using digital technologies and the intention of biometric systems is to perform the recognition or authentication of people based on some biological characteristics that are intrinsically unique for each individual. The effectiveness of a biometric system is measured mainly by the distinguishing attributes that are used to verify the identity. A large number of biometric traits have been investigated and some of them are nowadays used in several applications. Common physical traits include fingerprints, ear, hand or palm geometry, vein, retina, iris and facial characteristics [2]. Behavioural traits include voice, signature, keystroke pattern and gait.

show abstract

Section: Single-script Signature-based Biometric Authenticationmentioning

confidence: 99%