J.A. Carrasco-Ochoa et al. (Eds.): MCPR 2010, LNCS 6256, pp. 210–219, 2010 © Springer-Verlag Berlin Heidelberg 2010 On-line Signature Verification Based on Modified Dynamic Time Warping and Wavelet Sub-band Coding Juan Carlos Sánchez-Diaz 1 , Juan Manuel Ramírez-Cortes 1 , Rogerio Enriquez-Caldera 1 , and Pilar Gomez-Gil 2 1 Department of Electronics; National Institute of Astrophysics, Optics and Electronics. Luis Enrique Erro No. 1 Tonantzintla, Puebla. 72840. Mexico 2 Department of Computer Science, National Institute of Astrophysics, Optics and Electronics. Luis Enrique Erro No. 1 Tonantzintla, Puebla. 72840. Mexico karnaught@hotmail.com, jmram@inaoep.mx, rogerio@inaoep.mx, pgomez@inaoep.mx Abstract. This paper presents an on-line signature biometric system based on a modified Dynamic Time Warping (DTW) algorithm applied to the signature wavelet coefficients. The modification on DTW relies on the use of direct matching points information (DMP) to dynamically adapt the similarity meas- ure during the matching process, which is shown to increase the verification success rate. The wavelet analysis is done using a sub-band coding algorithm at global and local level. The use of wavelet coefficients showed a considerable reduction in processing time and an improvement in the equal error recognition rate (EER). The system was tested using a locally constructed database. A com- parison of the ROC curves obtained in each case is presented. Keywords: Signature, online, verification, dynamic time warping, wavelet. 1 Introduction Automatic personal identification and verification based on biometrics has received extensive attention in past years. Biometric identification refers to identifying an indi- vidual based on physiological or behavioral characteristics. It associates/disassociates an individual with a previously determined identity based on how one is or what one does. Identification can be in the form of verification, which entails authenticating a claimed identity, or recognition, which entails determining the identity of a given per- son from a database of persons known to the system [1]. A biometric system aims to provide automatic recognition of an individual based on features or characteristics unique to each human being. Biometric systems are based on several modalities, such as iris, face, ear shape, hand-shape, fingerprints, palm prints [2-5], or dynamical features like gait, on-line signature verification [6-7], or combination of them [8]. Requirements, strengths, and weaknesses of each modali- ty have been widely reported in the literature. Among the different existing forms of biometrics, signature-based verification has the advantage that signature analysis requires no invasive measurements and it is