Improving palmprint identification by combining multiple classifiers and using gabor filter

Biometrics technology has been attracting extensive attention from researchers of personal authentication due to the ever growing demand on access control, public security, forensics and e-banking. Diverse biometric systems have been now widely used in various applications. The palmprint is most popular modality that has the largest shares in the biometrics market. In this paper, an efficient online personal identification system based on palmprint is proposed. Firstly, the palmprint is filtered by the Gabor filter. The real and imaginary responses of the filtering are used to create three different feature vectors. Subsequently, the Hidden Markov Model (HMM), the Gaussian Mixture Model (GMM) and the Radial Basis Function (RBF) are used for modeling and so for classifying the feature vectors. The results of the three classifiers are combined using matching score level fusion strategy. The proposed system is tested and evaluated using PolyU 2D and 3D palmprint database of 250 persons. The obtained experimental results show that the system yields the best performance for identifying palmprint and it is able to provide the highest degree of biometrics-based system security.