You are on page 1of 4

LivDet competitions

Pattern Recognition
and Applications Lab

LivDet 2013
Fingerprint Liveness Detection Competition
Gian Luca Marcialis, Stephanie Schuckers, Fabio Roli
Organizing Committee

di Cagliari, Italia

ICB-2013 - Madrid, Spain
Dipartimento di
Ingegneria Elettrica
ed Elettronica

Competition Overview

Fake creation (cooperative method)

• Algorithms
– Software-based systems
– Four data sets (sensors)
• 2,000 live and fake fingers for training
• 2,000 live and fake fingers for test
• Four basic materials (gelatine, silicone, plasticine, latex)

– Three optical sensor (Biometrika, Italdata, Crossmatch) and one swipe sensor
– Biometrika and Italdata
• Test on non-consensual (latent) fake fingerprints

– Crossmatch and Swipe
• Test on consensual fake fingeprints

– Due to «problems» with CMT data set we only present Biometrika, Italdata and
Swipe results

• Systems
– Hybrid-based systems (even hardware-based fingerprint liveness detection)
– 1,000 live and spoof (fake) attempts
– additional attempts by 50 fake fingerprints per material brushed on saline

http://pralab.diee. • [outputfile]: text file with the output of each processed Fake creation (non-cooperative method) Fake creation (non-cooperative method) http://pralab.unica.unica.Protocol A Win32 console application: LIVENESS http://pralab.exe [ndataset] [inputfile] [outputfile] • [ndataset]: id number of the data set to .diee. • [inputfile]: text file with the list of images to analyze. in the same order of [inputfile].Fake creation (cooperative method) Part 1: Algorithms .diee.

A.A.15% 85.30% 99.Protocol Part 1: Algorithms .diee.unica.72% UniNap3 93.exe or similar executable file.75% 95.unica. CAoS 95. • The use of either a USB or Firewire connection. 0 spoof) .00 20. • An .unica.diee.99% Anonym1 98.45% 90.81% 79.00 50.30% 96.05% 86.55% 73.diee.19% 97.55% 66.20% 99.85% N. N.A.58% UniNap2 93.Part 1: Algorithms .40% 54.26% UniNap1 95.50% 85.00 10.A.45% 96.57% Itautec 91.00 30.42% HZ-JLW 67.A. ATVS 94.00 70.10% N.00% 53.50% 85.96% Anonym3 94. three of them preferred to be anonymous • Eleven different algorithms Non-consensual (LivDet 2013) Average standard deviation http://pralab.40% 94.00 40.47% 97.00 90.85% 84. N.Results Overall Accuracy rate • Four data sets created with four different sensors • 4000 or more images for each of the data set • Nine participants (academic and industrial institutions).93% 91.17% Anonym2 98. • To output the collected image if the image is considered a live image and a liveness score in the range of 0 and 100 (100 is the maximum degree of liveness.00% 98.30% 97.00 0.20% 94.00 Consensual (LivDet 2011) Average accuracy http://pralab. Part 1: Algorithms – Results Part 2: Systems .93% 92.00 60.10% N.00 System requirements: • Running on either a Windows XP 32-bit or 64-Bit system.unica.20% 96.Protocol 100.A.diee.00% 96.95% 50. Biometrika Italdata Swipe Average Dermalog 98.

a high FerrLive value. from both academic and industrial institutions.6% 0.unica.8% 1. leading to a noticeable error rate. Gelatin and Ecoflex). • In Part 2.diee. is growing with respect to previous FerrFake Unknown 1.4% FerrFake Known 0.0% 0.Protocol Part 2: Systems .unica.0% FerrLive 11. • Part 1 showed that both consensual and nonconsensual fake fingerprint were difficult to recognize.diee.diee.unica.0% Many thanks to… • The number of participants.3% 0.unica. • In addition the systems were tested with two unspecified methods (the materials were Modasil and Latex). Submitted Systems Dermalog Morpho FerrFake 0.Results FerrLive and FerrFake for submitted systems • Participants were asked to ship a fingerprint system which captures a fingerprint image as well as outputs a liveness detection score. could still be a problem when integrating fingerprint liveness detection methods on verification systems http://pralab. Stephanie Bauer Luca Ghiani Valerio Mura Simona Tocco David Yambay http://pralab.0% • Three spoof recipes and methods were made available to the participants (the materials were Conclusions http://pralab.Part 2: Systems . and thus a high rate of live fingerprints misclassified as fakes.