A novel method of applying directional filter bank (DFB) for Finger-Knuckle-Print (FKP) recognition

Abstract

Biometrics based personal authentication is an effective method for recognizing a person's identity. One of the latest biometric recognition methods, which are recently used for human identity authentication, is Finger-Knuckle-Print (FKP). In this paper, we propose a new method for personal identification. This paper includes a new method of using directional filter bank for feature extraction, combination of LDA algorithms for dimensionality reduction and Euclidean distance for matching and classification purpose. Also the information fusion is used with combination of fingers to improve the performance of recognition. This work is used at feature level. Poly-U Finger-Knuckle-Print database is used to evaluate the accuracy of the proposed method. The result of identification experiments in combining the features of four fingers obtained 99.29% accuracy, which means the proposed method is efficient and effective.