Title :
Performance of parametric and reference pattern based features in static signature verification: a comparative study
Author_Institution :
Fac. of Mech. & Electr. Eng., Damascus Univ., Syria
Abstract :
The performances of parametric and reference-pattern-based features (RPBFs) in static signature verification are compared. It is shown that RPBFs significantly improve the results and result in a verification rate of about 90% on skilled forgeries using only shape features. The proposed extension of the concept of RPBFs to the two-dimensional form has led to an important achievement in signature verification, where approximately 100% elimination of skilled forgeries in offline systems has been made possible using only shape features
Keywords :
character recognition; character recognition; parametric features; reference pattern based features; shape features; skilled forgeries; static signature verification; two-dimensional form; Data mining; Feature extraction; Forgery; Hafnium; Handwriting recognition; Humans; Performance analysis; Pixel; Position measurement; Shape;
Conference_Titel :
Pattern Recognition, 1990. Proceedings., 10th International Conference on
Conference_Location :
Atlantic City, NJ
Print_ISBN :
0-8186-2062-5
DOI :
10.1109/ICPR.1990.118180