Abstract
Automated fingerprint matching is considered as the most challenging phase of fingerprint recognition since it can be affected by a variety of factors such as noise, skin condition, rotation, distortions and displacements. When there is a large database, the search time to get a matching fingerprint will be relatively high. To reduce the searching time in the database, we have proposed a minutiae tree based indexing method in this paper. The database is represented in the form of a minutiae tree and the fingerprint matching is done by visiting the nodes in the tree where the local configuration of the minutiae is stored. By using this tree structure it is found that the search time or matching time can be considerably reduced and the matching time is independent of the number of fingerprints enrolled in the database. Our framework is scalable and the experiments conducted explores its ability to find correct matches with minimal search time.
Highlights
Fingerprint verification and fingerprint identification are the two different modes of operations of a fingerprint based biometric system
The translation and rotation of the fingerprint images, application of a poor feature extraction algorithm, displaced, false and missing minutiae and the non linear deformations of the images are some of the reasons
It should return the correct result while comparing fingerprints from the same finger even the feature extractor has missed some features or even when the fingerprint images are affected by non linear distortions
Summary
Fingerprint verification and fingerprint identification are the two different modes of operations of a fingerprint based biometric system. The translation and rotation of the fingerprint images, application of a poor feature extraction algorithm, displaced, false and missing minutiae and the non linear deformations of the images are some of the reasons. To overcome these variations a powerful matching algorithm is needed. The matching algorithm should be invariant to translation and rotation of the images It should return the correct result while comparing fingerprints from the same finger even the feature extractor has missed some features or even when the fingerprint images are affected by non linear distortions
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have