8/12/2019 08 NATO Talk
1/23
Automated video surveillance:challenges and solutions.
ACE Surveillance (Annotated Critical Evidence)
case study.
Dmitry Gorodnichy and Tony Mungham
Laboratory & Scientific Services DirectorateCanada Border Services Agency
www.videorecognition.com/ACE
http://www.videorecognition.com/ACEhttp://www.videorecognition.com/ACE8/12/2019 08 NATO Talk
2/23
2
Outline
Problems with status-quo Video Surveillance
Real-time and archival problems
Operational considerations
Next generation solution - Video Analytics based Motion detection myth and problem
Object detection as example of real intelligence
ACE Surveillancefirst fully-functional object-detection-based prototype
Year long tests with different levels of complexity
What that means for future of Video Surveillance
Conclusions
8/12/2019 08 NATO Talk
3/23
3
Role of Video Technology (VT)
In the context of enhancing security, VideoTechnology (VT) is one of the most demandedtechnologies of the 21st century
It is publicly acceptable It provides rich in content data
Multi-million funding in Canada and worldwide:
CBSA Port Runner project invested 10s of Millions inCCTV upgrades
Transport Canada opens $35M of funding towardsprocurement of CCTV
8/12/2019 08 NATO Talk
4/23
4
VT at CBSA
CBSA is a major user of CCTV systems at POEs
Most major CCTV installations start to leverage VT
Current task: to lead applied R&D to push VT to help
CBSA apply S&T innovative approaches to bordermanagement:
Event detection and notification to provide effectiveresponse to events
Traffic trends analysis to assist with border management
Video storage management to manage the cost of storageand meet obligations under the privacy act
Data integration/fusion of contextualised video information
8/12/2019 08 NATO Talk
5/23
5
Problem with status quo use of CCTVsurveillance
Modes of operation:
1. Active - personnel watch video at all times
2. Passive - in conjunction with other duties
3. Archival - for post-event analysis
Current systems and protocols are not efficient
in either mode!
Problem in real-time modes: an event may easilypass unnoticed .
due to false or simultaneous alarms,
lack of time needed to rewind and analyse all video
streams.
8/12/2019 08 NATO Talk
6/23
6
Problems in Archival mode:
Due to temporal nature of data:1. Storage space consumption problem
Typical assignment:2-16 cameras, 7 or 30 days of recording, 2-10 Mb / min.1.5 GB per day per camera / 20 - 700 GB total !
2. Data management and retrieval problem
London bombing video backtracking experience:
Manual browsing of millions of hours of digitized video fromthousands of cameras proved impossible within time-sensed period
[by the Scotland Yard trying to back-track the suspects]
8/12/2019 08 NATO Talk
7/237
Operational considerations
Lots of CCTV infrastructure: Many local initiatives, notcoordinated
Most video technology decisions are influenced byvendors - short-term solutions
Over 30 different video systems within the same dept. (atRCMP)
A national program with proper benchmark-based planningand evaluation of VT is required
Leveraging advances recently made in S&T
Technical standards for capturing /saving video data.
Policies in when, where and how VT should be used.
8/12/2019 08 NATO Talk
8/238
Video Technology today
Video Analytics (Video Recognition)
Analog
20th
century
21st century
First video recording
Digital
Wireless, Network Connected (IP)
8/12/2019 08 NATO Talk
9/239
Next generation Video Technology
Is Video Analytics based
also identified as:
Video Recognition,
Intelligent Video, Smart Video / Smart Camera
Video Analysis & Content Extraction
Perceptual Vision is not much about capturing better data (better
lenses, grabbers, coders, transmitters)
but about understanding captured data (better
theory)
8/12/2019 08 NATO Talk
10/2310
Status-quo video intelligence
Transport Canada CCTV Reference Manual forSecurity Application .
Australian Government National code of practice for
CCTV applications in urban transportUSA Government :recommended security Guidelines
for Airport Planning, Design and Construction.
. refer toMotion-based capture as IntelligentSurveillance Technology, and make theirrecommendations based on thereon.
8/12/2019 08 NATO Talk
11/2311
Motion-detection is not intelligent!
Term Motion-based is coined to make people believe thatvideo recognition is happening, which is not!
Its actually illumination-change-based, as it uses simplepoint brightness comparison:
Which often happens not because of motion! Changing light / weather (esp. in 24/7 monitoring)
Against sun/light, out of focus, blurred, thru glass
Reflections, diffraction, optical interferences
Image transmission, compression losses
8/12/2019 08 NATO Talk
12/2312
Object-detection is intelligent
but few can do it, since necessary advances in videorecognition theory became possible only recently (>2002).
In 2002 National Research Council of Canada (NRC)starts
developing Video Recognition Systems to leverage itsscientific Video Recognition expertise for the industry.
In 2005, it develops ACE Surveillance:
an object-detection-based Automated surveillanCEprototype capable of automatically extractingAnnotated Critical Evidence from live video.
NRC becomes also the organizer of the first Canadian academic workshopsdedicated to Video Processing for Security (since 2004)
8/12/2019 08 NATO Talk
13/2313
What is ACE Surveillance?
A Windows software that performs real-time video analytics byintegrating best object detection and tracking algorithms.
Replaces video clips with annotated still images:
Compresses 1 Gb of video into 2 Mb of easy to browse and
analyze still images ACE Surveillance output:
A 7-hour activity from day tonight (17:00 - 24:00) is
summarized into 2 minutes(600Kb) of Annotated CriticalEvidence snapshots.
Note illumination changes! - Watch treeshadows and sun light.
8/12/2019 08 NATO Talk
14/23
8/12/2019 08 NATO Talk
15/2315
Adds on top of existing infrastructure
8/12/2019 08 NATO Talk
16/2316
Status quo Motion-based capture(Courtesy: NRC-IIT Video Recognition Systems project)
1. Many captured snapshots are
useless: either noise or
redundant
2. Without visual annotation,
motion information is lost.
3. Hourly distribution ofsnapshots is not useful
8/12/2019 08 NATO Talk
17/2317
ACE Surveillance Object-based capture(Courtesy: NRC Video Recognition Systems project)
1. Each captured shot is useful.
2. Object location and velocity
shown augmentent.
3. Hourly distribution of shots is
indicative of what happened in
each hour, provides good
summarization of activities overlon eriod of time.
8/12/2019 08 NATO Talk
18/2318
ACE Surveillance testing benchmarks
Tested in different levels of complexity:
lighting conditions,
object motion patterns,
camera location environmental constraints.
most difficult - outdoors in unconstrained environments withlittle or no object motion consistency (as around a private
house in a regular neighbourhood).most easy - in controlled indoor environment where minimaldirect sunlight is present and where all objects are ofapproximately the same size and exhibit similar motionpattern (as at access gate inside the business building).
Outdoor wireless eye-level Outdoor, webcam, overview Indoor with sunlight, CCTV Indoor w/o sunlight, CCTV
8/12/2019 08 NATO Talk
19/23
VT within CBSA 19
Outdoor, wireless, eye-level Outdoor, webcam, overview Indoor with sunlight, CCTV Indoor w/o sunlight, CCTV
8/12/2019 08 NATO Talk
20/2320
Enables efficient detection of abnormalactivities Delivery EntryBack Door Entry
On
week-day
Onweek-e
nd
More than usual
8/12/2019 08 NATO Talk
21/2321
ACE Surveillance results
In real-time mode: alarm sounds & last captured evidence(time-stamped) is shown.
In archival mode: Zoom on the evidence browsing of captured
evidences zoom on a day, on hour, then on event - point
and click (for high res as needed)Made Commissioners much more aware of activities.
8/12/2019 08 NATO Talk
22/2322
Conclusions
Affordable automated (intelligent) video surveillance(AVS) is possible! To replace traditional DVR
OR to supplement them: DVR for 1 month + AVS for 1 year
However: Requires extra training from security officers.
Requires new protocols to handle automatically extractedevidence. - From forensic prospective, data that are not original and have
been processed by a computer can not be considered as evidence.
Requires new privacy policies. - Surveillance data are normally not kept for a long period of time
(
8/12/2019 08 NATO Talk
23/23
ACE surveillance case study outcome
ACE Surveillance(which is developed by a research lab)provides a reference standard against which can bemeasured solutions coming from industry.
It deals with common misconceptions related to deploying
intelligent video surveillance systems (IVS): motion detection myth vs real object detection and tracking.
The one-fit-all myth. - Extra video analytics expertise is required toset and operate IVS.
better video data (better resolution or compression) do not implybetter video intelligence. - ACE Surveillance is shown to work withregular TV quality data (320 by 240 pixels).
Howeverbetter quality of video image is needed for forensicpurposes as evidence
Due to closing of the project by NRC CBSA takes lead on it