Palmprint identification using isometric projection and linear discriminant analysis

Biometrics are unique, reliable and stable physical or behavioral characteristics that can be effectively used for personal identification. One of these robust biometrics is palmprint. In personal identification systems, feature extraction is an important issue. In this paper, we propose an algorithm that selects proper features in two stages. At first isometric projection (IsoP) and then linear discriminant analysis (LDA) is used to remove un-necessary features and extract proper features. Efficient extracted features are classified by K-nearest neighborhood (KNN) to identify person. Hong Kong Polytechnic University (PolyU) palmprint database is used to evaluate the performance of the proposed algorithm. Experimental results demonstrate that proposed method has better efficiency in comparison with recently proposed algorithms for palmprint identification.