Wyniki wyszukiwania - Biblioteka Nauki

Ten serwis zostanie wyłączony 2025-02-11.

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

Znaleziono wyników: 3

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: signature verification

Sortuj według:

Ogranicz wyniki do:

Hidden Signature for DTW Signature Verification in Authorizing Payment Transactions

100%

Putz-Leszczyńska J. , Kudelski M.

tom nr 4

59-67

Traditional use of dynamic time warping for signature verification consists of forming some dissimilarity measure between the signature in question and a set of "template signatures". In this paper, we propose to replace this set with the hidden signature and use it to calculate the normalized errors of signature under verification. The approach was tested on the MCYT database, using both genuine signatures and skilled forgeries. Moreover, we present the real-world application of the proposed algorithm, namely the complete biometric system for authorizing payment transactions. The authorization is performed directly at a point of sale by the automatic signature verification system based on the hidden signature.

Signature verification: A comprehensive study of the hidden signature method

75%

Putz-Leszczyńska J.

tom Vol. 25, no. 3

659--674

Many handwritten signature verification algorithms have been developed in order to distinguish between genuine signatures and forgeries. An important group of these methods is based on dynamic time warping (DTW). Traditional use of DTW for signature verification consists in forming a misalignment score between the verified signature and a set of template signatures. The right selection of template signatures has a big impact on that verification. In this article, we describe our proposition for replacing the template signatures with the hidden signature—an artificial signature which is created by minimizing the mean misalignment between itself and the signatures from the enrollment set. We present a few hidden signature estimation methods together with their comprehensive comparison. The hidden signature opens a number of new possibilities for signature analysis. We apply statistical properties of the hidden signature to normalize the error signal of the verified signature and to use the misalignment on the normalized errors as a verification basis. A result, we achieve satisfying error rates that allow creating an on-line system, ready for operating in a real-world environment.

On-line signature biometric system with employment of single-output multilayer perceptron

63%

Szklarzewski A. , Derlatka M.

2006

tom Vol. 26, no. 4

91-102

This paper presents a practical approach to handwritten signature recognition. The biometric system was designed by means of the single-output multilayer perceptrons (MLP). The signatures were acquired using a digitizing tablet which captures information about the signing process, like the pen tip position (x,y), pen tip pressure and pen azimuth, and altitude angles. Both the identification and signature verification problems were considered. The system recognizes forged and original signatures in case of identification and verification problems and, based on it, accepts or denies access for a given user. Working of the designed system allowed to suppose that the presented realization of the signature biometric system can be used in various security systems which have a limited number of registered users.