Chapter 19: Hybrid Fingerprint Verification System Based on Fusion of Feature Extraction and Minutiae Detection Strategy

Abstract

In this Paper we have presented a Fingerprint verification system based on hybrid approach which combines both the Feature extraction by applying a set of different wavelet families (functions) and minutiae information available in the fingerprint image by fusion rule method. The Core Point (CP) of the input Fingerprint is detected. Keeping the CP in the center,the image of size w x w is cropped and further processing is done with the cropped portion of the image. The cropped portion of the image is divided into four equal parts and each part is decomposed into frequency sub-bands for feature extraction. Minutiae are also detected from the cropped portion of the image. To evaluate the systempsilas rate of FAR and FRR, the DET curve has been implemented which will automatically plot the graph between FAR and FRR of the proposed system. Extensive experiments have been conducted on FVC2002 and Biometric System Lab fingerprint database. The results of this hybrid technique have also been compared by replacing the proposed approach with Gabor filter convolution strategy for feature extraction with 16 Gabor filters. The performance output shows the tremendous efficacy of the proposed system.