Score level fusion of Iris and Fingerprint using wavelet features
Shailendra Tiwari, Sudhakar Tripathi, K. V. Arya · 2016
Unimodal biometric systems have been serving the security demands of real world applications to a great level but these systems show vulnerabilities to certain aspects like noisy inputs, non-universality, intra-class variability and spoofing. To overcome these limitations multimodal biometric systems were developed which use more than one biometric trait for recognition. Iris and Fingerprint were considered as biometric modalities in this work because of their high compatibility in real world applications. A combination of 2-level Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) are used to obtain features of iris. Similarly, a combination of 2-level DWT and Fast Fourier Transform (FFT) are used to obtain features of Fingerprint. Feature matching was performed using Euclidean distance algorithm. Fusion is done using linear summation of scores obtained from individual modalities. Verification and identification tests were conducted on proposed multimodal biometric systems of iris and fingerprint. The proposed system has shown better performance than the existing system.