Fingerprint Indexing Based on Local Arrangements of Minutiae Neighborhoods

Abstract

This paper proposes a hash-based indexing method to speed up fingerprint identification in large databases. For each minutia, its local neighborhood information is computed with features defined based on the geometric arrangements of its neighboring minutiae points. The features used are provably invariant to translation, rotation, scale and shear. These features are used to create an affine invariant local descriptor, called an arrangement vector, for each minutia. To account for missing and spurious minutiae, we consider subsets of the neighboring minutiae and hashes of these structures are used in the indexing process. The primary goal of the work is to explore the effectiveness of affine invariant features for representing local minutiae structures. Experiments on FVC 2002 databases show that representation is quite effective even though the technique performs slightly below the state-of-the-art methods. One could use the representation in combination with other techniques to improve the overall performance.