Departamento
de Informática da Universidade da Beira Interior 
SOCIA Lab. – Soft Computing and Image
Analysis Group Department of Computer Science, University of Beira
Interior, 6201001 Covilhã, Portugal 
NICE.I Evaluation Let Alg denote the submitted executable, which performs the
segmentation of the noise free regions of the iris. Let I={I_{1},…,I_{n}} be the data set containing the
input closeup iris images. Let O={O_{1},…,O_{n}} be the output images correspondent to the above
described inputs, such that Alg(I_{i})=O_{i}.
Let C={C_{1},…,C_{n}} be the manually classified binary iris images, given by the NICE.I
Organizing Committee. It must be assumed that each C_{i} contains the perfect
iris segmentation and noise detection result for the input image I_{i}. All the images of I, O and C have the same dimensions: c columns and r rows. Two measures of evaluation will be
used: The classification error rate (E^{1})
of the Alg participation on the
input image I_{i} (E_{i})
is given by the proportion of correspondent disagreeing pixels (through the logical
exclusiveor operator) over all the image: where O(c’,r’) and C(c’,r’)
are, respectively, pixels of the output and class images. The classification error rate (E^{1}) of the Alg participation is given by the
average of the errors on the input images E_{i}: The value of (E^{1}) is closed in the [0, 1] interval and will be the measure of evaluation and
classification of the NICE.I participations. In this context, “1” and “0”
will be respectively the worst and optimal values. The second error measure aims to
compensate the disproportion between the apriori probabilities of “iris” and “noniris
pixels in the images. The typeI and
typeII error rate (E^{2}) of the image E_{i} is given by the average between the falsepositives
(FPR) and falsenegatives (FNR) rates: E_{i} = 0.5 * FPR + 0.5 FNR Similarly to the E^{1} error
rate, the final E^{2} error rate is given by the average of the
errors (E_{i}) on the input images. 


fghfghfghfgh
DIUBI
Bloco VI Rua Marquês de Ávila e Bolama P 6201001 Covilhã PORTUGAL