BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Europe/Rome
BEGIN:DAYLIGHT
DTSTART:20180325T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20181028T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20180928T162211Z
UID:6F49CF04-B132-4EDA-9192-DC8F21F9E81A
DTSTART;TZID=Europe/Rome:20181016T100000
DTEND;TZID=Europe/Rome:20181016T110000
DESCRIPTION:The likelihood-ratio is a valuable concept in pattern recogniti
 on in general as well as in biometrics. One reason for this is that it is 
 optimal in Neyman-Pearson sense\, guaranteeing the best receiver operating
  characteristic. In this presentation\, 2 examples of how the use of likel
 ihood-ratio can improve biometric recognition performance.\n\nExample 1: A
  likelihood-ratio classifier for histogram features\n\nIn a number of clas
 sification problems\, the features are represented by histograms. Traditio
 nally\, histograms are compared by simple distance measures such as the ch
 i-square\, the Kullback-Leibler\, or the Euclidian distance. This paper pr
 oposes a likelihood ratio classifier for histogram features. It assumes th
 at the bin probabilities of the histograms can be modeled by a Dirichlet d
 istribution. A simple method to estimate the Dirichlet parameters is inclu
 ded. Feature selection prior to classification improves the classification
  performance. It will be demonstrated that the proposed classifier outperf
 orms the chi-square distance measure.\n\nExample 2: Integration of homomor
 phic encryption and optimal likelihood ratio classifiers\n\nHomomorphic en
 cryption can be used to perform biometric verification\, in particular the
  comparison of biometric features\, in an encrypted domain. Because of its
  complexity in terms of numbers of required operations and the communicati
 on protocols involved\, proposed implementations involve simple classifier
 s\, such as Euclidian or cosine distances. This may lead to a lower recogn
 ition performance than could be achieved with a better classifier. We prop
 ose a two-party verification protocol that integrates the likelihood-ratio
  classifier with El-Gamal homomorphic encryption in such a way that the bi
 ometric recognition performance does not decrease with low computational c
 omplexity.\n\nSpeaker(s): R.N.J. Veldhuis (Raymond)\, \n\nRoom: Multimedia
  room\, Bldg: B\, Roma Tre University\, Dept. of Engineering\, Section of 
 Applied Electronics\, Via Vito Volterra 62\, Roma\, Lazio\, Italy\, 00146
LOCATION:Room: Multimedia room\, Bldg: B\, Roma Tre University\, Dept. of E
 ngineering\, Section of Applied Electronics\, Via Vito Volterra 62\, Roma\
 , Lazio\, Italy\, 00146
ORGANIZER:patrizio.campisi@uniroma3.it
SEQUENCE:5
SUMMARY:Likelihood-ratio classifiers in biometrics
URL;VALUE=URI:https://events.vtools.ieee.org/m/178151
X-ALT-DESC:Description: <br /><p>The likelihood-ratio is a valuable concept
  in pattern recognition in general as well as in biometrics. One reason fo
 r this is that it is optimal in Neyman-Pearson sense\, guaranteeing the be
 st receiver operating characteristic. In this presentation\, 2 examples of
  how the use of likelihood-ratio can improve biometric recognition perform
 ance.</p>\n<p>Example 1: A likelihood-ratio classifier for histogram featu
 res</p>\n<p>In a number of classification problems\, the features are repr
 esented by histograms. Traditionally\, histograms are compared by simple d
 istance measures such as the chi-square\, the Kullback-Leibler\, or the Eu
 clidian distance. This paper proposes a likelihood ratio classifier for hi
 stogram features. It assumes that the bin probabilities of the histograms 
 can be modeled by a Dirichlet distribution. A simple method to estimate th
 e Dirichlet parameters is included. Feature selection prior to classificat
 ion improves the classification performance. It will be demonstrated that 
 the proposed classifier outperforms the chi-square distance measure.</p>\n
 <p>Example 2: Integration of homomorphic encryption and optimal likelihood
  ratio classifiers</p>\n<p>Homomorphic encryption can be used to perform b
 iometric verification\, in particular the comparison of biometric features
 \, in an encrypted domain. Because of its complexity in terms of numbers o
 f required operations and the communication protocols involved\, proposed 
 implementations involve simple classifiers\, such as Euclidian or cosine d
 istances. This may lead to a lower recognition performance than could be a
 chieved with a better classifier. We propose a two-party verification prot
 ocol that integrates the likelihood-ratio classifier with El-Gamal homomor
 phic encryption in such a way that the biometric recognition performance d
 oes not decrease with low computational complexity.&nbsp\;</p>
END:VEVENT
END:VCALENDAR