Offline Signature Verification through Probabilistic Neural Network

Date issued

2010

Journal Title

Journal ISSN

Volume Title

Publisher

Václav Skala - UNION Agency

Abstract

In this paper, we show the positive potential of verifying the offline handwritten signatures through discrete Radon transform (DRT), principle component analysis (PCA) and probabilistic neural network (PNN). Satisfactory results are obtained with 1.51%, 3.23%, and 13.07% equal error rate (EER) for random, casual, and skilled forgeries respectively on our independent database.

Description

Subject(s)

offline verifikace podpisu, diskrétní Radonova transformace, analýza hlavních komponent, pravděpodobnostní neuronové sítě

Citation

WSCG 2010: Communication Papers Proceedings: 18th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision in co-operation with EUROGRAPHICS, p. 31-38.
OPEN License Selector