Algorithm Performance Evaluation

The parameters adopted for the performance evaluation will be the following:

Evaluation per dataset

• Frej_n: Rate of failure to process for the sub-set n
• Fcorrlive_n: Rate of correctly classified live iris for sub-set n
• Fcorrfake_n: Rate of correctly classified fake iris for sub-set n
• Ferrlive_n: Rate of misclassified live iris for sub-set n
• Ferrfake_n: Rate of misclassified fake iris for sub-set n
• ET: Average processing time per image

Rates are based on the assumption of a threshold of 50.

Overall evaluation based on average across the three datasets

• Frej: Rate of failure to process
• Fcorrlive: Rate of correctly classified live iris
• Fcorrfake_n: Rate of correctly classified fake iris
• Ferrlive_n: Rate of misclassified live iris
• Ferrfake_n: Rate of misclassified fake iris

Declaration of the Winner

• The winner will be awarded by minimum of the average of the overall classification errors on the three datasets. Only one winner will be awarded.