Biometric authenticator algorithm based on multiresolution analysis
Abstract
In this paper, we propose a feature extraction method for two-dimensional imageauthentication algorithm using curvelet transform and principal component analysis(PCA). Since wavelet transform can not adequately describe facial curves features,Theproposed approach involves image denoising applying a 2D-Curvelet transform toachieve compact representations of curves singularities. To assess the performanceof the presented method, we have employed three classifification techniques: Neuralnetworks (NN), K-Nearest Neighbor (KNN) and Support Vector machines (SVM).Extensive experimental results and comparison with the existing methods show the effectiveness of the proposed recognition method in the ORL face database and CASIAiris database.
Keywords
Biometric authentication; Classifification algorithmes; Curvelet transform; Feature extraction; Multiresolution analysis
Full Text:
PDFDOI: http://doi.org/10.11591/ijeecs.v20.i3.pp1332-1341
Refbacks
- There are currently no refbacks.