SlideShare a Scribd company logo
1 of 32
Download to read offline
© Fraunhofer
Smart Video Surveillance
and Privacy
 Dr.-Ing. Erik Krempel,
 Fraunhofer IOSB, Karlsruhe, Germany
© Fraunhofer IOSB
2
What can we do with video analytics…
SMART VIDEO SURVEILLANCE
© Fraunhofer IOSB
3
From Visual to Symbolic Video Surveillance
• The classic approach of video surveillance is visual video
monitoring by security staff.
• Today, video analytics are more and more in use as assistance systems. Mainly to
attract and sustain attention of the staff to possible points of interst.
• Current research is going towards useage of advanced video analytics for symbolic
situation awareness and „privacy-aware management by exception“
Classic Approach Intelligent Video Analytics
Privacy-aware Situation
Awareness Tools
© Fraunhofer IOSB
4
Video Analytics for Scene Understanding
Object Tracking
Activity Recognition
Semantic Scene
Understanding
Scene Context Recognition
Object DetectionImage Enhancement
© Fraunhofer IOSB
5
Current Research: Advanced Objekt Detection, Tracking
and Re-Identification
• Multi-camera multi-object detection and tracking of persons and vehicles
• indoor / outdoor applications with complex illumination
conditions
• self-calibrating cameras (automated geometric and color
calibration)
• New approaches on dynamic learning of objects‘ unique features
© Fraunhofer IOSB
6
Towards Activity Recognition for (Global) Scene
Understanding
• Classification of activities in videos.
© Fraunhofer IOSB
7
Person Retrieval
• Component that is looking for similar persons
in archive in a specific time interval
• Features of the person of interest
(color features and texture features) are compared to already computed
features of other tracks
• System generates a list
of tracks, sorted based
on their similarity to
the person of interest
• Operator can verify the
results visually
© Fraunhofer IOSB
8
Why do we need to certify smart video surveillance systems
CERTIFICATION
© Fraunhofer IOSB
9
Security
 What is the error rate of video processing algorithms?
 False positive rates
 False negative rates
 Is the system robust against environment conditions?
 Rain on the camera
 Light conditions
 Is the system safe against attacks?
 The systems keeps secrets for itself (Confidentiality)
 The systems stays operational in the presence of an attacker
(Availability)
 The systems detects the same results in the presence of an attacker
(Integrity)
© Fraunhofer IOSB
10
Trust / Transparency
© Fraunhofer IOSB
11
Trust
 How can you achieve real transparency in such complex systems?
 Is there a human preventing the system from making wrong decisions?
 Does the systems prevent discrimination?
© Axis Communications
© Fraunhofer IOSB
12
Efficiency dimension
 Is the system able to reduce crime / increase security?
 Is the system easy to use?
 Can you extend the system / use components of another manufacturer?
© Fraunhofer IOSB
13
Freedom Infringement
 Does the system prevent discrimination?
 Are video processing algorithms restricted to legal applications?
 Does the system prevent misuse by the operator?
 Security of the data processing…
 Secure against data theft…
 Purpose limitation…
 Errors by the systems…
© Fraunhofer IOSB
14
RESEARCH WORK
Privacy in smart video surveillance
© Fraunhofer IOSB
15
Situation-dependent video surveillance
 Alexander Roßnagel, Monika Desoi, and Gerrit Hornung (2011)
© Fraunhofer IOSB
16
Idea: Privacy-aware Surveillance Workflows
Default Mode: Optimized for privacy Assessment Mode:
Event-specific, privacy preserving
Event
detected
Event
confirmed
Event
resolved
Investigation Mode:
Event specific functions unlocked
© Fraunhofer IOSB
17
Default Mode: Optimized for Privacy
 Computer vision algorithms in background
 Abstract representation of observed environment visualized
 No access to video data
 No access to archived data
 Exposes as little information about observed persons as possible
© Fraunhofer IOSB
18
Assessment Mode: Event-specific, Privacy Preserving
Event
detected
 Assessment view according to event type
 Anonymized live video data released
 Possibly anonymized access to a limited buffer of recorded (video)
data
 Only selected cameras
© Fraunhofer IOSB
19
Investigation Mode: Event-specific Functions Unlocked
Event
confirmed
Event
resolved
 Allows additional privacy intrusions for investigation purposes
 I.e., retrieving the person who dropped a piece of luggage
 Restrict additional analyses to persons related to the event under
investigation
© Fraunhofer IOSB
20
IMPLEMENTATION
How to build systems with this concept…
© Fraunhofer IOSB
21
Security Use-Case
© Fraunhofer IOSB
22
Safety Use-Case: Prototype NurseEye
© Fraunhofer IOSB
23
Operator interaction
 Default mode:
 No Access
 Assessment mode:
 Access to
anonymized video
 Investigation mode:
 Full access to video
to help in
emergency handling
© Fraunhofer IOSB
24
Transparency
 All cameras come with displays showing the current mode and data
usage
 Displays become monitor for chat in the investigation mode
Default mode Assessment mode Investigation mode
© Fraunhofer IOSB
25
Video NurseEye
© Fraunhofer IOSB
26
TOOLS
© Fraunhofer IOSB
27
User Study on Anonymization Techniques: Introduction
 Which obfuscation technique should be used?
 Regarding privacy (identity leakage), utility and perceptual video
quality
ROI Anonymized person
Blurring Silhouette Edge detection PixelizationBlurring
(gray scale)
© Fraunhofer IOSB
28
Results: Subjective assessments
1 = silhouette; 2 = pixelization; 3,5 = edge detection; 4 gray scale blurring; 6-8 color blurring
Error bars represent one standard
deviation of the data sample
Perceived privacy protection and perceived image quality
© Fraunhofer IOSB
29
Video anonymisation
Edge detectionPixelization
Gaussian blurring Silhouette
© Fraunhofer IOSB
30
Adaptive „Privacy-Masking“ for Pan/Tilt/Zoom Camera
• Advanced Privacy Masking for semi-stationary PTZ-cameras
• Given position of camera over ground (or an elevation model of the site) optical distance to
objects in the scene is estimated
• Given a parameter for „privacy-preserving resolution“ video is pixelized or blurred
inhomoge-
nously depending on distance to object.
• Highly senstive areas (e.g. buildings / windows) can be blacked out of the stream by
dynamic adaptable polygons (depending on pan/tilt/zoom settings)
© Fraunhofer IOSB
31
Information Flow Tracking in NEST
Video &
Observation
archive
PEP
HMI
GeoViewer
Observations
PEP
Requests
HMI
StreamViewer
PEP
Image
exploitation
algorithm
Image
exploitation
algorithm
Image
exploitation
algorithm
PEP
IMG
Policy
Decision
Point
Policy
Information
Point
Policy
Observations
Control
PEP
Access
Control
OOWM
Association
PEP HLS-Alarms
PEP Classification
PEP Fusion
© Fraunhofer IOSB
32
Discussion
“Data is not an asset, it’s a liability”
-Marko Karppinen-
Contact Information:
Erik Krempel
Fraunhofer IOSB
Fraunhoferstr. 1
76131 Karlsruhe, Germany
erik.krempel@iosb.fraunhofer.de
+49-721-6091-292

More Related Content

Viewers also liked

Viewers also liked (12)

Crisp kaleidoscope presentation 13112015
Crisp kaleidoscope presentation 13112015Crisp kaleidoscope presentation 13112015
Crisp kaleidoscope presentation 13112015
 
Policy and Standardisation perspective - CRISP Final Conference
Policy and Standardisation perspective - CRISP Final ConferencePolicy and Standardisation perspective - CRISP Final Conference
Policy and Standardisation perspective - CRISP Final Conference
 
CRISP and HECTOS projects - key findings
CRISP and HECTOS projects - key findings CRISP and HECTOS projects - key findings
CRISP and HECTOS projects - key findings
 
Tips - How to interview a candidate
Tips - How to interview a candidateTips - How to interview a candidate
Tips - How to interview a candidate
 
شركات انشئت لتيقى
شركات انشئت لتيقىشركات انشئت لتيقى
شركات انشئت لتيقى
 
Napoles
NapolesNapoles
Napoles
 
Democratization and political_parties_in_nepal-_political__science[1]
Democratization and political_parties_in_nepal-_political__science[1]Democratization and political_parties_in_nepal-_political__science[1]
Democratization and political_parties_in_nepal-_political__science[1]
 
Standardisation and certification basics
Standardisation and certification basicsStandardisation and certification basics
Standardisation and certification basics
 
[서울예술대학교 201701] 디지털아트전공실기 III - 4주차
[서울예술대학교 201701] 디지털아트전공실기 III - 4주차[서울예술대학교 201701] 디지털아트전공실기 III - 4주차
[서울예술대학교 201701] 디지털아트전공실기 III - 4주차
 
Storecove sending invoices - SimplerInvoicing
Storecove sending invoices - SimplerInvoicingStorecove sending invoices - SimplerInvoicing
Storecove sending invoices - SimplerInvoicing
 
Port amenities
Port amenitiesPort amenities
Port amenities
 
Circuitos
CircuitosCircuitos
Circuitos
 

Similar to Smart Video Surveillance and Privacy - CRISP Final Conference

iSentry Brochure
iSentry BrochureiSentry Brochure
iSentry Brochure
Zalmi Katz
 
Ivisiotech A2 Tech
Ivisiotech A2 TechIvisiotech A2 Tech
Ivisiotech A2 Tech
a2tech
 
Reacting to Advanced, Unknown Attacks in Real-Time with Lastline
Reacting to Advanced, Unknown Attacks in Real-Time with LastlineReacting to Advanced, Unknown Attacks in Real-Time with Lastline
Reacting to Advanced, Unknown Attacks in Real-Time with Lastline
Lastline, Inc.
 
Broucher Soft copy
Broucher Soft copyBroucher Soft copy
Broucher Soft copy
evsrao
 
Ruleml2012 - Rule-based high-level situation recognition from incomplete tra...
Ruleml2012  - Rule-based high-level situation recognition from incomplete tra...Ruleml2012  - Rule-based high-level situation recognition from incomplete tra...
Ruleml2012 - Rule-based high-level situation recognition from incomplete tra...
RuleML
 
Embedded Image and Vision Processing 2017 report by Yole Developpement
Embedded Image and Vision Processing 2017 report by Yole DeveloppementEmbedded Image and Vision Processing 2017 report by Yole Developpement
Embedded Image and Vision Processing 2017 report by Yole Developpement
Yole Developpement
 
My documentation
My documentationMy documentation
My documentation
cas123
 
My documentation
My documentationMy documentation
My documentation
cas123
 

Similar to Smart Video Surveillance and Privacy - CRISP Final Conference (20)

ParanaVision Presentation (English)
ParanaVision Presentation (English)ParanaVision Presentation (English)
ParanaVision Presentation (English)
 
CNL Software - PSIM for Energy, Oil & Gas
CNL Software - PSIM for Energy, Oil & GasCNL Software - PSIM for Energy, Oil & Gas
CNL Software - PSIM for Energy, Oil & Gas
 
iSentry Brochure
iSentry BrochureiSentry Brochure
iSentry Brochure
 
Smart surveillance
Smart surveillanceSmart surveillance
Smart surveillance
 
Private residence complex security. Technical solutions
Private residence complex security.  Technical solutionsPrivate residence complex security.  Technical solutions
Private residence complex security. Technical solutions
 
Advance Intelligent Video Surveillance System Using OpenCV
Advance Intelligent Video Surveillance System Using OpenCVAdvance Intelligent Video Surveillance System Using OpenCV
Advance Intelligent Video Surveillance System Using OpenCV
 
Indigo Vision - Complete Solution Portfolio
Indigo Vision -  Complete Solution PortfolioIndigo Vision -  Complete Solution Portfolio
Indigo Vision - Complete Solution Portfolio
 
Ivisiotech A2 Tech
Ivisiotech A2 TechIvisiotech A2 Tech
Ivisiotech A2 Tech
 
Advanced Intelligent Video Surveillance System In Elevators By Using OpenCV
Advanced Intelligent Video Surveillance System In Elevators By Using OpenCVAdvanced Intelligent Video Surveillance System In Elevators By Using OpenCV
Advanced Intelligent Video Surveillance System In Elevators By Using OpenCV
 
Event-Handling Based Smart Video Surveillance System
Event-Handling Based Smart Video Surveillance SystemEvent-Handling Based Smart Video Surveillance System
Event-Handling Based Smart Video Surveillance System
 
Reacting to Advanced, Unknown Attacks in Real-Time with Lastline
Reacting to Advanced, Unknown Attacks in Real-Time with LastlineReacting to Advanced, Unknown Attacks in Real-Time with Lastline
Reacting to Advanced, Unknown Attacks in Real-Time with Lastline
 
Nodeflux ai for covid19 adhiguna mahendra phd
Nodeflux ai for covid19 adhiguna mahendra phdNodeflux ai for covid19 adhiguna mahendra phd
Nodeflux ai for covid19 adhiguna mahendra phd
 
VORTEX
VORTEXVORTEX
VORTEX
 
DEEP LEARNING APPROACH FOR SUSPICIOUS ACTIVITY DETECTION FROM SURVEILLANCE VIDEO
DEEP LEARNING APPROACH FOR SUSPICIOUS ACTIVITY DETECTION FROM SURVEILLANCE VIDEODEEP LEARNING APPROACH FOR SUSPICIOUS ACTIVITY DETECTION FROM SURVEILLANCE VIDEO
DEEP LEARNING APPROACH FOR SUSPICIOUS ACTIVITY DETECTION FROM SURVEILLANCE VIDEO
 
Broucher Soft copy
Broucher Soft copyBroucher Soft copy
Broucher Soft copy
 
Ruleml2012 - Rule-based high-level situation recognition from incomplete tra...
Ruleml2012  - Rule-based high-level situation recognition from incomplete tra...Ruleml2012  - Rule-based high-level situation recognition from incomplete tra...
Ruleml2012 - Rule-based high-level situation recognition from incomplete tra...
 
Embedded Image and Vision Processing 2017 report by Yole Developpement
Embedded Image and Vision Processing 2017 report by Yole DeveloppementEmbedded Image and Vision Processing 2017 report by Yole Developpement
Embedded Image and Vision Processing 2017 report by Yole Developpement
 
IoT Meets Exhibition Areas: a Modular Architecture to Improve Proximity Inter...
IoT Meets Exhibition Areas: a Modular Architecture to Improve Proximity Inter...IoT Meets Exhibition Areas: a Modular Architecture to Improve Proximity Inter...
IoT Meets Exhibition Areas: a Modular Architecture to Improve Proximity Inter...
 
My documentation
My documentationMy documentation
My documentation
 
My documentation
My documentationMy documentation
My documentation
 

Recently uploaded

development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virus
NazaninKarimi6
 
Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptx
MohamedFarag457087
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
Cherry
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Sérgio Sacani
 
Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.
Cherry
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
Cherry
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptx
seri bangash
 

Recently uploaded (20)

development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virus
 
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
 
Genetics and epigenetics of ADHD and comorbid conditions
Genetics and epigenetics of ADHD and comorbid conditionsGenetics and epigenetics of ADHD and comorbid conditions
Genetics and epigenetics of ADHD and comorbid conditions
 
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLGwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
 
GBSN - Microbiology (Unit 3)Defense Mechanism of the body
GBSN - Microbiology (Unit 3)Defense Mechanism of the body GBSN - Microbiology (Unit 3)Defense Mechanism of the body
GBSN - Microbiology (Unit 3)Defense Mechanism of the body
 
Site specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfSite specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdf
 
Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptx
 
Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Plasmid: types, structure and functions.
Plasmid: types, structure and functions.
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
 
Role of AI in seed science Predictive modelling and Beyond.pptx
Role of AI in seed science  Predictive modelling and  Beyond.pptxRole of AI in seed science  Predictive modelling and  Beyond.pptx
Role of AI in seed science Predictive modelling and Beyond.pptx
 
PATNA CALL GIRLS 8617370543 LOW PRICE ESCORT SERVICE
PATNA CALL GIRLS 8617370543 LOW PRICE ESCORT SERVICEPATNA CALL GIRLS 8617370543 LOW PRICE ESCORT SERVICE
PATNA CALL GIRLS 8617370543 LOW PRICE ESCORT SERVICE
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.
 
Cyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptxCyanide resistant respiration pathway.pptx
Cyanide resistant respiration pathway.pptx
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
 
Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.Reboulia: features, anatomy, morphology etc.
Reboulia: features, anatomy, morphology etc.
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
 
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS ESCORT SERVICE In Bhiwan...
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS  ESCORT SERVICE In Bhiwan...Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS  ESCORT SERVICE In Bhiwan...
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS ESCORT SERVICE In Bhiwan...
 
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRingsTransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
 
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate ProfessorThyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptx
 

Smart Video Surveillance and Privacy - CRISP Final Conference

  • 1. © Fraunhofer Smart Video Surveillance and Privacy  Dr.-Ing. Erik Krempel,  Fraunhofer IOSB, Karlsruhe, Germany
  • 2. © Fraunhofer IOSB 2 What can we do with video analytics… SMART VIDEO SURVEILLANCE
  • 3. © Fraunhofer IOSB 3 From Visual to Symbolic Video Surveillance • The classic approach of video surveillance is visual video monitoring by security staff. • Today, video analytics are more and more in use as assistance systems. Mainly to attract and sustain attention of the staff to possible points of interst. • Current research is going towards useage of advanced video analytics for symbolic situation awareness and „privacy-aware management by exception“ Classic Approach Intelligent Video Analytics Privacy-aware Situation Awareness Tools
  • 4. © Fraunhofer IOSB 4 Video Analytics for Scene Understanding Object Tracking Activity Recognition Semantic Scene Understanding Scene Context Recognition Object DetectionImage Enhancement
  • 5. © Fraunhofer IOSB 5 Current Research: Advanced Objekt Detection, Tracking and Re-Identification • Multi-camera multi-object detection and tracking of persons and vehicles • indoor / outdoor applications with complex illumination conditions • self-calibrating cameras (automated geometric and color calibration) • New approaches on dynamic learning of objects‘ unique features
  • 6. © Fraunhofer IOSB 6 Towards Activity Recognition for (Global) Scene Understanding • Classification of activities in videos.
  • 7. © Fraunhofer IOSB 7 Person Retrieval • Component that is looking for similar persons in archive in a specific time interval • Features of the person of interest (color features and texture features) are compared to already computed features of other tracks • System generates a list of tracks, sorted based on their similarity to the person of interest • Operator can verify the results visually
  • 8. © Fraunhofer IOSB 8 Why do we need to certify smart video surveillance systems CERTIFICATION
  • 9. © Fraunhofer IOSB 9 Security  What is the error rate of video processing algorithms?  False positive rates  False negative rates  Is the system robust against environment conditions?  Rain on the camera  Light conditions  Is the system safe against attacks?  The systems keeps secrets for itself (Confidentiality)  The systems stays operational in the presence of an attacker (Availability)  The systems detects the same results in the presence of an attacker (Integrity)
  • 10. © Fraunhofer IOSB 10 Trust / Transparency
  • 11. © Fraunhofer IOSB 11 Trust  How can you achieve real transparency in such complex systems?  Is there a human preventing the system from making wrong decisions?  Does the systems prevent discrimination? © Axis Communications
  • 12. © Fraunhofer IOSB 12 Efficiency dimension  Is the system able to reduce crime / increase security?  Is the system easy to use?  Can you extend the system / use components of another manufacturer?
  • 13. © Fraunhofer IOSB 13 Freedom Infringement  Does the system prevent discrimination?  Are video processing algorithms restricted to legal applications?  Does the system prevent misuse by the operator?  Security of the data processing…  Secure against data theft…  Purpose limitation…  Errors by the systems…
  • 14. © Fraunhofer IOSB 14 RESEARCH WORK Privacy in smart video surveillance
  • 15. © Fraunhofer IOSB 15 Situation-dependent video surveillance  Alexander Roßnagel, Monika Desoi, and Gerrit Hornung (2011)
  • 16. © Fraunhofer IOSB 16 Idea: Privacy-aware Surveillance Workflows Default Mode: Optimized for privacy Assessment Mode: Event-specific, privacy preserving Event detected Event confirmed Event resolved Investigation Mode: Event specific functions unlocked
  • 17. © Fraunhofer IOSB 17 Default Mode: Optimized for Privacy  Computer vision algorithms in background  Abstract representation of observed environment visualized  No access to video data  No access to archived data  Exposes as little information about observed persons as possible
  • 18. © Fraunhofer IOSB 18 Assessment Mode: Event-specific, Privacy Preserving Event detected  Assessment view according to event type  Anonymized live video data released  Possibly anonymized access to a limited buffer of recorded (video) data  Only selected cameras
  • 19. © Fraunhofer IOSB 19 Investigation Mode: Event-specific Functions Unlocked Event confirmed Event resolved  Allows additional privacy intrusions for investigation purposes  I.e., retrieving the person who dropped a piece of luggage  Restrict additional analyses to persons related to the event under investigation
  • 20. © Fraunhofer IOSB 20 IMPLEMENTATION How to build systems with this concept…
  • 22. © Fraunhofer IOSB 22 Safety Use-Case: Prototype NurseEye
  • 23. © Fraunhofer IOSB 23 Operator interaction  Default mode:  No Access  Assessment mode:  Access to anonymized video  Investigation mode:  Full access to video to help in emergency handling
  • 24. © Fraunhofer IOSB 24 Transparency  All cameras come with displays showing the current mode and data usage  Displays become monitor for chat in the investigation mode Default mode Assessment mode Investigation mode
  • 27. © Fraunhofer IOSB 27 User Study on Anonymization Techniques: Introduction  Which obfuscation technique should be used?  Regarding privacy (identity leakage), utility and perceptual video quality ROI Anonymized person Blurring Silhouette Edge detection PixelizationBlurring (gray scale)
  • 28. © Fraunhofer IOSB 28 Results: Subjective assessments 1 = silhouette; 2 = pixelization; 3,5 = edge detection; 4 gray scale blurring; 6-8 color blurring Error bars represent one standard deviation of the data sample Perceived privacy protection and perceived image quality
  • 29. © Fraunhofer IOSB 29 Video anonymisation Edge detectionPixelization Gaussian blurring Silhouette
  • 30. © Fraunhofer IOSB 30 Adaptive „Privacy-Masking“ for Pan/Tilt/Zoom Camera • Advanced Privacy Masking for semi-stationary PTZ-cameras • Given position of camera over ground (or an elevation model of the site) optical distance to objects in the scene is estimated • Given a parameter for „privacy-preserving resolution“ video is pixelized or blurred inhomoge- nously depending on distance to object. • Highly senstive areas (e.g. buildings / windows) can be blacked out of the stream by dynamic adaptable polygons (depending on pan/tilt/zoom settings)
  • 31. © Fraunhofer IOSB 31 Information Flow Tracking in NEST Video & Observation archive PEP HMI GeoViewer Observations PEP Requests HMI StreamViewer PEP Image exploitation algorithm Image exploitation algorithm Image exploitation algorithm PEP IMG Policy Decision Point Policy Information Point Policy Observations Control PEP Access Control OOWM Association PEP HLS-Alarms PEP Classification PEP Fusion
  • 32. © Fraunhofer IOSB 32 Discussion “Data is not an asset, it’s a liability” -Marko Karppinen- Contact Information: Erik Krempel Fraunhofer IOSB Fraunhoferstr. 1 76131 Karlsruhe, Germany erik.krempel@iosb.fraunhofer.de +49-721-6091-292