Identity recognition based on palm vein feature using two-dimensional linear discriminant analysis

Research on biometrics continues to grow. Various studies conducted to increase the performance of personal identification based on physical and behavioral characteristics. Palm vein recognition became an interesting field lately. Palm vein feature covered underneath the skin so that it hard to forge and more resist to external factors than fingerprint and face features. In this research, recognition process consists of preprocessing, feature extraction and matching. Feature extraction has been done using Two-Dimensional Linear Discriminant Analysis. This method could reduce dimension by maximizing between-class scatter and minimizing within-class scatter. Two-Dimensional Linear Discriminant Analysis obtained a good performance for CASIA palm vein dataset. The configuration of parameters needs to be determined in order to increase the system performance. The best performance obtained 8% in term of EER and Recognition Rate 94,67% with threshold 0,4933.