BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Fingerprint Analysis: Parametric and Nonparametric Models - Stepha
 n Huckemann (Georgia Augusta University Göttingen)
DTSTART:20140120T140000Z
DTEND:20140120T150000Z
UID:TALK49901@talks.cam.ac.uk
CONTACT:Carola-Bibiane Schoenlieb
DESCRIPTION:Fingerprint analysis for forensic and security related applica
 tions relies on global features such as the orientation field and local fe
 atures such as minutiae patterns. Here the ultimate goal would be to extra
 ct from a high-dimensional finger's imprint image bitmap a low-dimensional
  feature vector that is robust against Euclidean motions and noise such as
  local distortions. For the orientation field we propose a parametric mode
 l based on quadratic differentials and for minutiae patterns a non-paramet
 ric model based on histogram counts.
LOCATION:MR 15\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR