Fingerprint matching by neighbourhood characteristics

This paper presents the formulation and implementation of a new fingerprint pattern-matching algorithm. The algorithm uses the spatial relation between the junction points within the 11 × 11 neighbourhood of the core points to generate the pattern matching scores. The junction points were the points of intersections of straight lines connecting the feature points within the neighbourhood. Experiments were conducted using FVC2002 fingerprint database comprising four datasets of images of different sources and qualities. Three statistics; namely False Match Rate (FMR), False Non-Match Rate (FNMR) and the Average Matching Time (AMT) were generated for measuring the performance of the algorithm on the images. The results obtained demonstrated the effectiveness of the algorithm in distinguishing fingerprint images from different fingers. The results also revealed the failure rate of the algorithm when subjected to images with regions of significant degradations. Finally, findings from comparative analysis of the generated results with what obtained for some recently formulated fingerprint pattern matching algorithms revealed best performance for the proposed algorithm.