The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the
information contained therein. All document contained therein cannot be copied, reproduced or modified in the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Visual Surveillance Technologies for Enhancing ABC Secure Zones
Project: FastPassCsaba Beleznai, Stephan Veigl, Michael RauterDavid Schreiber, Andreas Kriechbaum
Video- and Security TechnologySafety & Security DepartmentAIT Austrian Institute of Technology GmbH
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
2
Motivation
Significantly increasing passenger flows• from 2012 to 2016 +800 million
Border guards face big challenges• in-depth document checks• reliable identity checks• check of entry conditions• discover possible threats
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
3
Motivation
Issues to be addressed Exactly one person per passport
Single person detection
Clean secure zone Left object detection
Situation overview Queue length estimation
Video Surveillance in Gate Reliable detections with 3D
stereo video cameraDemo eGate: Vienna Airport(Terminal 2, Non-Schengen-Arrivals)
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
ABC Gates
4
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
5
The Sensor
3D stereo-camera system developed by AIT Area-based, local-optimizing, correlation-
based stereo matching algorithm Specialized variant of the Census Transform Video image
• Spatial Resolution: 752x480 rectified to 608x328
• RGB Depth information
• 16 bit depth image USB 2 interface
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
6
Camera Set-Up (inside the gate)
Top-view of eGate interior ~15 frames per second Mounting height: 279 cm depth resolution: ~2 cm Ground floor 180x60 cm spatial resolution: ~1 cm
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
7
Single Person Detection
Motivation / Challenges Ensure only one person is inside the eGate
• reliable detection and counting of persons• multiple persons must not pass!
Real-time processing• low latency
Better than existing system• error-prone to tailgating / piggybacking• number of false alarms
Developed for secure zones
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
8
Single Person Detection
Method1. Head candidates
• hypothesis generation test positions
2. Classification with head-shoulder model• support-vector-machine
filter false hypothesis
3. Tracking of detections use door signals to increase
robustness
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Single Person Detection
Examples - single person
9video 1 - single person.avi
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Single Person Detection
Examples - multiple persons
10
video 2 - multiple persons.avi
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Single Person Detection
Examples - closed doors
11
video 3 - closed doors.avi
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
12
Left Object Detection
Motivation / Challenges Detect left objects in eGate
• eGate has to be empty after person left• visualize left object
Real-time processing• low latency
Difficult object size and / or apperance• small size (e.g. passport)• low contrast (e.g. empty bottle)
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
13
Left Object Detection
Method 3D Images
• color image• depth information• data fusion
Motion suppression
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Left Object Detection
Examples - Trolley
14
video 4 - trolley.avi
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Left Object Detection
Examples - Mobile Phone
15
video 5 – mobile phone.avi
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Left Object Detection
Examples - Trolley
16
video 6 – bottle.avi
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
17
Queue Length Estimation
Motivation / Challenges Enhanced queue management
• number of persons per queue
Visual tracking of queue dynamics• estimated waiting time
Overcome occlusion problems• eliminate top-view requirement
(for low ceiling height environments)• use 3D information
4 Min.
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
18
Queue Length Estimation
Outlook (Method) Stereo camera (e.g. mounted on eGate) Detection distance up to 12 meters
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Conclusion
Stereo Video Surviallance New dimension of information
Robust and reliable detection under variable situations
Decrease false alarms for• single person detection• left object detection
New options for queue length estimation
19
The work has been supported by the FastPass project. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 312583. This publication only reflects the author’s view and the European Union is not liable for any use that may be made of the information contained therein. All document contained therein cannot be copied, reproduced or modified in
the whole or in the part for any purpose without written permission from the FastPass Coordinator with acceptance of the Project Consortium.
Stephan VeiglVideo- and Security TechnologySafety & Security DepartmentAIT Austrian Institute of Technology [email protected]
20
AIT Austrian Institute of Technologyyour ingenious partner