Papers by michael hantschel
We study the effect of growth on the fingerprints of adolescents, based on which we suggest as im... more We study the effect of growth on the fingerprints of adolescents, based on which we suggest as imple method to adjust for growth when trying to retrieve an adolescent'sfi ngerprint in ad atabase years later.H ere, we focus on the statistical analyses used to determine howfingerprints grow: Procrustes analysis allows us to establish that fingerprints growisotropically,anappropriate mixed effects model shows that fingerprints essentially growproportionally to body height. The resulting growth model is validated by showing that it brings points of interest as close as if both fingerprints were taken from an adult. Further details on this study,inparticular results when applying our growth model in verification and identification tests, can be found in C. Gottschlich, T. Hotz, R. Lorenz, S. Bernhardt, M. Hantschel and A. Munk: Modeling the Growth of Fingerprints ImprovesMatching for Adolescents,IEEE Transations on Information Fo rensics and Security,2011 (to appear). 1I ntroduction Consider thefollowing scenario: an adolescent at age 11, say,gives his fingerprints which are entered into an automatic fingerprint identification system (AFIS); later,atage 30, his fingerprints are again taken, and run against the AFIS database. To find the adolescent fingerprint matching the adult one is made difficult by the fact that the adolescent has grown into an adult-a sh aveh is fingerprints, compare Figure 1. As these systems are usually engineered for adults, growth effects are not taken into account appropriately,and the AFIS will decide that the adolescent print matches the adult print poorly; indeed, the points of interest (POI), i.e. minutiae and singular points, cannot be brought close by merely rotating and translating the imprints, see Figure 1(right).
Uploads
Papers by michael hantschel