Title :
Feature Selection Based on Genetic Algorithms for On-Line Signature Verification
Author :
Galbally, Javier ; Fierrez, Julian ; Freire, Manuel R. ; Ortega-Garcia, Javier
Author_Institution :
Univ. Autonoma de Madrid, Madrid
Abstract :
Two different genetic algorithm (GA) architectures are applied to a feature selection problem in on-line signature verification. The standard GA with binary coding is first used to find a suboptimal subset of features that minimizes the verification error rate of the system. The curse of dimensionality phenomenon is further investigated using a GA with integer coding. Results are given on the MCYT signature database comprising 330 users (16500 signatures). Signatures are represented by means of a set of 100 features which can be divided into four different groups according to the signature information they contain, namely: i) time, ii) speed and acceleration, iii) direction, and iv) geometry. The GA indicates that features from subsets i and iv are the most discriminative when dealing with random forgeries, while parameters from subsets ii and iv are the most appropriate to maximize the recognition rate with skilled forgeries.
Keywords :
binary codes; feature extraction; genetic algorithms; handwriting recognition; image recognition; binary coding; feature selection; genetic algorithm; integer coding; online signature verification; random forgery; skilled forgery; Acceleration; Biometrics; Convergence; Error analysis; Feature extraction; Forgery; Genetic algorithms; Handwriting recognition; Information geometry; Spatial databases;
Conference_Titel :
Automatic Identification Advanced Technologies, 2007 IEEE Workshop on
Conference_Location :
Alghero
Print_ISBN :
1-4244-1300-1
Electronic_ISBN :
1-4244-1300-1
DOI :
10.1109/AUTOID.2007.380619