Closed-circuit television (CCTV), also known as video surveillance,[1][2] is the use of video cameras to transmit a signal to a specific place, on a limited set of monitors. It differs from broadcast television in that the signal is not openly transmitted, though it may employ point-to-point (P2P), point-to-multipoint (P2MP), or mesh wired or wireless links. Even though almost all video cameras fit this definition, the term is most often applied to those used for surveillance in areas that require additional security or ongoing monitoring (Videotelephony is seldom called "CCTV"[3][4]).
Surveillance of the public using CCTV is common in many areas around the world. Video surveillance has generated significant debate about balancing its use with individuals' right to privacy even when in public.[5][6][7]
In industrial plants, CCTV equipment may be used to observe parts of a process from a central control room, especially if the environments observed are dangerous or inaccessible to humans. CCTV systems may operate continuously or only as required to monitor a particular event. A more advanced form of CCTV, using digital video recorders (DVRs), provides recording for possibly many years, with a variety of quality and performance options and extra features (such as motion detection and email alerts). More recently, decentralized IP cameras, perhaps equipped with megapixel sensors, support recording directly to network-attached storage devices, or internal flash for completely stand-alone operation.
1. SmartCatch Systems
Putting Intelligence into Surveillance
19th Annual NDIA Security Technology Symposium
June 16-19, 2003
Brooks McChesney
Entrepreneur-in-Residence
NEC USA Inc
brooks@ccrl.sj.nec.com (408-863-6017)
2. Smart Surveillance: Features & Benefits
• Video intelligence software that detects, identifies
and tracks objects within existing CCTV systems
– Monitors events in real time to insure activities are limited
to specific tasks & within authorized areas of authority
– Automatically detects suspicious behaviors and other
violations of established security policies and procedures
– Distinguishes between objects: person(s), bag(s), vehicles,
etc.
– Sounds alerts & alarms according to user escalation
procedures
– Supports incident analysis and forensic search (in real time
or via preserved archives)
– Maintains confidentiality through restricted user and role
based access
3. The New Face of Physical Security
Then
• Reliance upon semi-
skilled guards with 90-
100% turnover and whose
costs exceeds that of
products & systems
• A highly fragmented,
unstable industry
• Products typically passive,
standalone sensors
• Thinking in terms of
borders, fences & doors
New Focus
• Reduce people cost with smart
technology
• Integrate systems with advanced
surveillance technology
• Prevent actions of those who
pose a threat while facilitating
actions of the honest majority
• Elevate operational effectiveness
& efficiency
• Deploy human-centric framework
that can associate actions with
identity, behavior and credentials
4. Human Costs of Security
In-House
Guard
$11B
Products &
Systems
$19B
Total Security Market
Contract
Guard
$11B
Position Staff
Annual
Salary
Annual
Cost
Security
Guard 5 X $26,628 = $133,140
Security
Director 1 X $64,133 = $64,133
Your
Current
Costs $197,273
Guard Cost
(staff one post, 24/7 at one facility)
Source : JP Freeman, GE, Security magazine, OnGuard.Net
6. Airport Examples
• Unattended baggage
– PanAm Lockerbie disaster: Heathrow Airport
• Unidentified person walks around the security gate
– Denver Airport, August 2002
• Unidentified person in unauthorized AOA
– Detroit Metro Airport , May 2002
• Piggybacking/Tailgating at secure access points
• Failure to recognize unattended vehicles
– Louisville Intl. Airport, Sept 2001
• Unidentified person gains access to the runway and stows
away
– Logan International Airport, Sept 1999
• Unidentified person disappears in the parking lot
• Suspicious movement against normal traffic patterns
8. System Overview
Card Reader Biometrics
Door Alarm
Live Camera
Feeds
Card Reader
Database
Door Alarm
Database
Biometrics
Database
SmartCatch
Vision
Central Matrix Switch
ALERTS
RULES
Card Reader
signal
9. SFO requirements
• Detect tailgating: more than one person
enters the door with one card swipe.
• Detect piggybacking: one person swipes
the card and another person enters the
door.
• Detect abnormal behaviors:
– Person swiping the card many times.
– Person loitering in front of the door.
13. SmartCatch Differentiators
Limitations of the current systems
• Difficulties of human tracking: non-rigid motion, changing
image size, occlusion and intersections
• Unable to handle changing shape in relation to camera angle
• Generates very high false alerts
• Cannot handle cluttered environment with multiple objects
SmartCatch Approach
• Neural Network that handles both humans and non-humans in
cluttered envionrments
• Tight Integration of Object detection & tracking resulting in very
high system accuracy
• Robust object classification engine
• Background modeling that automatically adjusts to changing
lighting conditions
• Real-time processing speed
14. SmartCatch: Benefits
• Provides a quantifiable increase in security
• Increases public/employee confidence
• Provides 24x7 protection without delays or
inconvenience to customers or employees
• Reduces risks of liability
• Increases operational efficiency
