Detection of liveness by fusing ECG and fingerprint

Fingerprint based biometric authentication is generally used for security purposes. Fingerprint scans aren’t highly reliable method of authorization, because it can be faked by obtaining a sample of fingerprint of person. In order to decrease prevalence of spoofing of biometric system, few spoof detection techniques available which must be primarily dependent on binary classification technique, wherein, two samples of fingerprint images, namely authorized and unauthorized images used to train classifier into detecting occurrence of any fake fingerprint. However, this method of training is not efficient as biometric system fails to detect fake fingerprints made using new techniques of spoofing. This can be avoided by including another method for efficient liveness detection using fusion of ECG signals captured from fingertips and fingerprint data. ECG signal will ensure the detection of real fingerprint samples from fake ones. In this paper, the proposed method is to combine ECG signal with fingerprint thus to reduce spoofing and increase system security and to achieve this, a liveness detection algorithm for fingerprint has been developed. Moreover, fingerprint recognition rate is also improved using this technique.