Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
Smart Borders Pilot German Experiences
IBPC 2016
NIST, Gaithersburg 05.05.2016
Proposal: EU ENTRY / EXIT SYSTEM (EES)
Source: EU KOM , factsheet on EES, Brussels, 6.4.2016 Slide 2
Proposal: EU ENTRY / EXIT SYSTEM (EES)
Source: EU KOM , factsheet on EES, Brussels, 6.4.2016 Slide 3
Smart Borders Pilot in 2015
(March – September, 2015)
Focus on Technology state of play Which & “how many” biometrics? ABC & Self-Service Kiosks Operational & end user experiences
German Participation
EU Pilot Frankfurt Airport (FRA)
• FP Enrolment (4 / 8 / 10 FP) • Automatic Border Control (ABC at Exit)
Extended National Pilot Additional Location: Seaport Warnemünde Additional Biometrics: Facial Image + Iris Additional Test Cases (many) Unique within EU pilot:
• End-2-End Pilot (with backends) • Full national integration • Focus on processes
Slide 5
Smart Borders Pilot – National Integration
Slide 6 Slide 6
Equipment Enrolment Morpho: „Fingerprint on the Fly“ (FOTF)
Crossmatch: The „new“ Guardian
Iris ID - iCAM TD100
eGates at Exit
Allowing TCNs to use ABCs during Exit
Participating TCNs
ARE, ARG, AUS, AZE, CAN,
CHL, CHN, ISR, JPN, KAZ,
KOR, MDA, MKD, MYS,
NZL, QAT, RUS, SGP, SRB,
TGO, THA, TJK, TUR, TWN,
USA, VEN
Slide 8
Video
(5 min)
Various findings on Smart Borders
Lessons Learned on • Duration & Quality of data • Architecture & Organization • New Technologies & Biometrics
Eu-LISA Pilot Report
German Pilot Report • In addition to eu-LISA report • in english
Slide 10
Duration
Slide 11
Duration First Entry for Visa Exempt
34
167
120
89
0
20
40
60
80
100
120
140
160
180
Baseline EES 10FP+FI EES 8FP+FI EES 4FP+FI
Dura
tion,
med
ian
[s]
Duration of border control process for TCNVE, entry, different biometrics, different devices, different quality thresholds.
Slide 12
Duration Distribution
Seite 13 Slide 13
Duration: Example Subsequent Entry
120
72
0
20
40
60
80
100
120
140
8FP+FI 4FP
Initial Entry Subsequent entry
Dura
tion,
med
ian
[s]
Comparison of first and subsequent entries for visa-exempt travellers (TCNVE)
Slide 14
Duration: Distribution per Scanner
30
32
25
0
5
10
15
20
25
30
35
Existing contact scanner New contact scanner New contactless scanner
Dura
tion,
med
ian
[s]
Duration of fingerprint capture process depending on scanner technology (TCNVE, entry, 8 fingers, eu-LISA final threshold)
Slide 15
Fingerprints
Slide 16
Quality Assessment NFIQ
Seite 17
Quality Assessment NFIQ2 (beta)
Seite 18 Slide 18
Quality: Time vs. Threshold
“Initial”: Retry policy on NFIQ 2/2/2/2/3/2/2/2/2/3 “eu-LISA Final”: Retry policy on NFIQ 3/3/3/3/4/3/3/3/3/4
50,6%
37,4%
69,1% 77,0%
15,5%
11,1%
9,7% 6,2%
33,9%
51,5%
21,1% 16,8%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
New contact New contactless New contact New contactless
Initial eu-LISA final
Amou
nt o
f tot
al c
aptu
res [
%]
3
2
1
Distribution of the capture count depending on scanner type & quality threshold
Slide 19
Start with high quality threshold “Initial”. BUT number of retries were too high
ONLY used auto-capture by device for interim period. THAT was quick, but not comparable with rest of EU
Compromise: “eu-LISA final” Reduced threshold Number of retries “feasible”
Quality: Time vs. Threshold
“Initial”: Retry policy on NFIQ 2/2/2/2/3/2/2/2/2/3 “eu-LISA Final”: Retry policy on NFIQ 3/3/3/3/4/3/3/3/3/4
85
78
0
10
20
30
40
50
60
70
80
90
100
Initial eu-LISA Final
NFI
Q (1
=100
, 2=8
0, 3
=60,
4=4
0,5=
20)
Distribution of fingerprint quality depending on quality thresholds
Slide 20
Quality loss by reducing threshold is measurable BUT adequate
Recommendation: accept lower quality and optimize time wise
Face & Iris
Slide 21
Live Capturing of Facial Image
In general quality positive Sometimes not sharp, too
close etc.
Changes to infrastructure Handling of camera
Slide 22
Quality: „chip“ versus „live“
Slide 23
Iris
Positive Very robust to environment
conditions
Slide 24
Quality Assessment Iris
Slide 25
German Report: Findings & Requirements
Maximum of available biometrics (“10fp or 8fp in combination with other”)
Biometric-driven – not passport-driven “Crossover” – architecture (EES, VIS …) Long retention period (5 yeas like VIS) … De-Duplication for 1st line Biometric Enrolment:
Reduce threshold, but take “as much as possible”
….many more (in the report)
Facilitation
Security
Efficiency
Slide 26
National Plans for 2016
German Smart Borders Pilot –Part Two Integration of Self-Service-Systems Adapting to new EU COM Proposal
EU level: Negotiation / Design
… and move towards a better integration Slide 27
Thank you for your attention!
Fares Rahmun [email protected]
+49 228 99 358 1548
Federal Office of Administration (Bundesverwaltungsamt)
www.bunderverwaltungsamt.de/en
Smart Borders Pilot�German Experiences��IBPC 2016�NIST, Gaithersburg�05.05.2016��Proposal: EU ENTRY / EXIT SYSTEM (EES)Proposal: EU ENTRY / EXIT SYSTEM (EES)Smart Borders Pilot in 2015German ParticipationSmart Borders Pilot – National IntegrationEquipment EnrolmenteGates at ExitFoliennummer 9Various findings on Smart BordersFoliennummer 11Duration First Entry for Visa Exempt Duration DistributionDuration: Example Subsequent EntryDuration: Distribution per ScannerFoliennummer 16Quality Assessment NFIQQuality Assessment NFIQ2 (beta)Quality: Time vs. ThresholdQuality: Time vs. ThresholdFoliennummer 21Live Capturing of Facial ImageQuality: „chip“ versus „live“IrisQuality Assessment IrisGerman Report: Findings & RequirementsFoliennummer 27Foliennummer 28