Title :
Dynamic stochastic segmentation of on-line signature contours
Author :
Paulik, Mark J. ; Mohankrishnan, N. ; Nikiforuk, Micheal
Author_Institution :
Dept. of Electr. Eng., Detroit Univ., MI, USA
Abstract :
An on-line signature contour is represented as an ordering of unique “curve types”, and modeled as a jump-nonstationary autoregressive (AR) process. We develop a stochastic segmentation algorithm which divides the signature contour in an effort to get parsimonious segmental AR models. This segmental model can then be utilized to provide effective identification and authentication results
Keywords :
autoregressive processes; handwriting recognition; identification; image segmentation; pattern recognition; authentication; autoregressive process; dynamic stochastic segmentation; identification; jump-nonstationary AR process; online signature contours; parsimonious segmental AR models; signature database; Authentication; Classification algorithms; Credit cards; Databases; Fingerprint recognition; Handwriting recognition; Hardware; Muscles; Radio access networks; Stochastic processes;
Conference_Titel :
Circuits and Systems, 1996., IEEE 39th Midwest symposium on
Conference_Location :
Ames, IA
Print_ISBN :
0-7803-3636-4
DOI :
10.1109/MWSCAS.1996.587798
