Departamento de Informática da Universidade da Beira Interior

 

 

 

 Home

 Protocol

 Evaluation 

 Publication

 Important Dates

 F.A.Q.

 Registration

 Registered Participants

 Forum

 About

 

 

 

 

 

 

SOCIA Lab. – Soft Computing and Image Analysis Group 

Department of Computer Science, University of Beira Interior, 6201-001 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={I1,…,In} be the data set containing the input close-up iris images.

Let O={O1,…,On}  be the output images correspondent to the above described inputs, such that Alg(Ii)=Oi.

Let C={C1,…,Cn} be the manually classified binary iris images, given by the NICE.I Organizing Committee. It must be assumed that each Ci contains the perfect iris segmentation and noise detection result for the input image Ii.

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 (E1) of the Alg participation on the input image Ii (Ei) is given by the proportion of correspondent disagreeing pixels (through the logical exclusive-or 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 (E1) of the Alg participation is given by the average of the errors on the input images Ei:

The value of (E1) 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 “non-iris pixels in the images. The type-I and type-II error rate (E2) of the image Ei is given by the average between the false-positives (FPR) and false-negatives (FNR) rates:

 

Ei = 0.5 * FPR + 0.5 FNR

Similarly to the E1 error rate, the final E2 error rate is given by the average of the errors (Ei) on the input images.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 











 

 

 

fghfghfghfgh 


DI-UBI Bloco VI Rua Marquês de Ávila e Bolama P- 6201-001 Covilhã PORTUGAL