Departamento
de Informática da Universidade da Beira Interior |
SOCIA Lab. – Soft Computing and
Image Analysis Group Department of Computer Science, University of Beira
Interior, 6201-001 Covilhã, Portugal |
NICE:II Evaluation Let P denote the submitted application, which gives the dissimilarity
between segmented iris images. Let I={I1,…,In} be the data set containing the
input iris images an let M={M1,…,Mn}be the corresponding binary maps that
give the segmentation of the noise-free iris region. P receives two iris images (and the
corresponding binary maps) and outputs the dissimilarity value between the
corresponding irises: P(Ii,
Mi Ij, Mj) - - > D. D should be a
real positive value. Performing a “one-against-all”
comparison scheme for each image of I gives
a set of intra-class dissimilarity values DI={DI1,…,DIk}
and a set of inter-class dissimilarity values DE={DE1,…,DEm},
whether the captured images are from the same or from different irises. The decidability value d’(DI1,…,DIk,DE1,…,DEm
)- - > [0, ∞[ will be used as evaluation measure: d’=| avg(DI) - avg(DE)|
/ sqrt ( 0.5* ( std(DI)2 + std(DE)2)
) where avg(DI) and
avg(DE) denote the average values of the intra-class and
inter-class comparisons and std(DI)
and std(DE) the
corresponding standard deviation values. Participants of the NICE:II contest
will be ranked from the highest (best) to the lowest (worst) decidability
values. |
|
|
fghfghfghfgh
DI-UBI
Bloco VI Rua Marquês de Ávila e Bolama P- 6201-001 Covilhã PORTUGAL