Fingerprint verification based on dual transformation technique

The Fingerprint is used to authenticate a person for multiple applications. In this paper, we propose Fingerprint Verification Based on Dual Transformation (FVDT). The Fingerprint image size of 300*480 is segmented into four cells of each size 150*240. The DCT is applied on each cell to convert from spatial domain to frequency domain. One level DWT is applied on DCT coefficients to derive four sub bands such as LL, LH, HL and HH. The directional information features and the centre area features on LL sub bands are computed. The final feature vector set is formed by concatenating directional information features and centre area features. The matching between database fingerprint and test fingerprint is based on ED, SVM and RF. It is observed that TSR and FRR values are improved in the case of proposed algorithm compared to the existing algorithm.