The measurement of biometric feature information is important for biometric technology, as for it can determine the uniqueness of biometric features, compare the performance of several feature extraction methods and quantify whether combination of features or biometric fusion offers any advantage. In this paper, we study the measurement of palmprint feature information using relative entropy between intra-person and inter-population. We compute the biometric feature information in which the feature extracted by three different methods, including: Principal Component Analysis (PCA), Linear Discriminant Analysis(LDA) and Locality Preserving Projections(LPP). The average biometric feature information is calculated to be approximately 280 bits for PCA, 246 bits for LDA features and 460 bits for LPP. © Springer International Publishing 2013.
CITATION STYLE
Shi, W., Sun, D., & Wang, S. (2013). Measuring biometric feature information in palmprint images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8232 LNCS, pp. 249–257). https://doi.org/10.1007/978-3-319-02961-0_31
Mendeley helps you to discover research relevant for your work.