Continuous versus exclusive classification for fingerprint retrieval
Pattern Recognition Letters
Fingerprint Matching Using Transformation Parameter Clustering
IEEE Computational Science & Engineering
Fingerprint Indexing Based on Novel Features of Minutiae Triplets
IEEE Transactions on Pattern Analysis and Machine Intelligence
Fingerprint Identification Using Delaunay Triangulation
ICIIS '99 Proceedings of the 1999 International Conference on Information Intelligence and Systems
Error Control Coding, Second Edition
Error Control Coding, Second Edition
Fingerprint Indexing Using Ridge Invariants
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 04
Handbook of Fingerprint Recognition
Handbook of Fingerprint Recognition
A fingerprint retrieval system based on level-1 and level-2 features
Expert Systems with Applications: An International Journal
Fingerprint indexing with bad quality areas
Expert Systems with Applications: An International Journal
Indexing and retrieving in fingerprint databases under structural distortions
Expert Systems with Applications: An International Journal
Dynamic discriminant functions with missing feature values
Pattern Recognition Letters
Hi-index | 0.00 |
Indexing is the process of assigning a numerical value to a database entry in order to facilitate its rapid retrieval. Indexing a fingerprint database can reduce the search space and improve the response time of an identification system. We discuss a novel method for generating index codes for fingerprint images by using a small set of pre-determined reference fingerprints. In the proposed method, the match scores generated by comparing an input fingerprint with the reference fingerprints are subjected to a discretization function, which converts them into an index code. A search mechanism based on the Hamming distance identifies those index codes in the database that are similar to the code of the input image. The proposed technique has several advantages: it obviates the need to extract complex features from the fingerprint image; it utilizes the matcher that is already associated with a particular application; and it can be used to index any biometric database irrespective of the trait or matcher being used. Experimental results on two fingerprint databases (NIST-4 and WVU) indicate that the proposed encoding scheme generates index codes that are well-scattered thereby allowing noisy query images to be indexed correctly.