• The Course
• The Lecturers
• School venue
• Travel and accomodation
• Open projects
• Final tests

Download:

• Application form 2020
• Application for Scholarships 2020

 

• Course Schedule 2019 - PDF
• Course Schedule 2018 - PDF
• Course Schedule 2017 - PDF
• Course Schedule 2016 - PDF
• Course Schedule 2015 - PDF

   

• Student presentations 2019 - PDF

Evaluation metrics for forensic biometrics

Prof. David Van Leeuwen
Radboud University, Netherlands

An important asset for development and deployment of biometric systems is a way for measuring the performance of such a system. In order to measure the performance, a number of ingredients are required.

Firstly, there is a need for a well defined methodology, upon which all stake holders (developer, technology provider, client) agree.

Secondly we need a performance metric, typically a number that indicates how well the system does what it is supposed to do, and that can be used to compare different systems, or different configurations of the same system.

Finally, there needs to be a data set that is used to test the quality of the biometric system.

As you will have noticed from previous lectures, the terminology and the actual performance metric used seems to be different for each of the biometric technologies. In this lecture we will set up a framework in which all of these evaluation metrics have a place, which will attempt to do using a common terminology. The difference between discrimination and calibration ability of the biometric technology is discussed, and we will also look at the different application scenarios of biometric technologies, and what the right evaluation measure for these is.