Local feature based off-line signature verification using neural network classifiers

Authors:
Bence Kovari;Adam Horvath;Benedek Toth;Hassan Charaf
Affiliations:
Department of Automation and Applied Informatics, Budapest University of Technology and Economics, Budapest, Hungary;Department of Automation and Applied Informatics, Budapest University of Technology and Economics, Budapest, Hungary;Department of Automation and Applied Informatics, Budapest University of Technology and Economics, Budapest, Hungary;Department of Automation and Applied Informatics, Budapest University of Technology and Economics, Budapest, Hungary
Venue:
MAMECTIS'09 Proceedings of the 11th WSEAS international conference on Mathematical methods, computational techniques and intelligent systems
Year:
2009

Citing 5
Cited 0

Learning Strategies and Classification Methods for Off-Line Signature Verification

IWFHR '04 Proceedings of the Ninth International Workshop on Frontiers in Handwriting Recognition
On the Power of Feature Analyzer for Signature Verification

DICTA '05 Proceedings of the Digital Image Computing on Techniques and Applications
The Image Processing Handbook, Fifth Edition (Image Processing Handbook)

The Image Processing Handbook, Fifth Edition (Image Processing Handbook)
Offline signature verification using the discrete radon transform and a hidden Markov model

EURASIP Journal on Applied Signal Processing
Time-efficient stroke extraction method for handwritten signatures

ACS'07 Proceedings of the 7th Conference on 7th WSEAS International Conference on Applied Computer Science - Volume 7

Quantified Score

Hi-index	0.00

Visualization

Abstract

Signature recognition is probably the oldest biometrical identification method with a high legal acceptance. Although automated signature verification has been studied for more than 30 years this field still lacks the necessary formalization to evaluate and compare different signature verification systems. Our research aims separating the steps of signature verification and dissecting the monolith verification systems into smaller benchmarkable parts. This paper focuses on classification, the last phase of signature verification. In contrast with typical applications, our solution is able to take both global and local features of the signatures into consideration. This also introduces some questions which are all addressed in the paper. Several local features are introduced, and evaluated by using a neural network classifier, with a special emphasis on the usability of shape descriptors.