 Analysis Capability
Additional analysis is made equally straight-forward by offering rapid answers to questions such as:
- Could a given passenger have been at a station or on a train at a particular time?
- Is it possible for a given passenger to have visited a specified station during a nominated time period?
- How long could a given passenger have been present at a particular station during their journey?
- Is it possible for a given passenger to have been in contact with a specified train during their journey?
- Which routes could a passenger have taken during their journey?
- Is it possible for two or more nominated passengers to have been in contact with each other on the network?
Route analysis capabilities are incorporated to allow ambiguous routing options to be quickly resolved. Ambiguities can arise in very highly interconnected networks offering frequent train services. When ambiguities are detected, the technology calculates and prioritises the minimum amount of CCTV footage that will most efficiently allow each passenger’s actual route to be determined.
Scenarios
Consider a 7 July 2005 London Bombing style attack on a rail network. By entering the location and time of each bombing into the Innovation Science Passenger Tracking software, the system determines a short-list of candidate passengers that could have been caught up in each incident. The search only takes a couple of seconds.
A picture of every candidate passenger entering the network is either retrieved from recorded CCTV footage (if on-line access is available) or logged for manual retrieval. The pictures are paired with information that is already associated with the candidate passenger’s ticket (which will vary depending upon the ticketing policies and level of anonymity offered by the given transit network).
Pictures can be used by emergency services to identify casualties. Even though a name and address may not immediately be available for a given casualty, information such as the passenger’s station of origin will be known and could help to speed the process of formal identification.
Video Prioritisation
17,000 surveillance video tapes were retrieved from the London Underground as part of the 7 July 2005 London Bombing investigations [1]. With over 8,500 cameras on the London Underground network (est. 12,000 by 2012), the potential for an unmanageable volume of CCTV footage is substantial.
The Rapid Passenger Tracking software identifies which video footage could contain one or more passengers of interest for any given incident. Video of interest is then prioritised for analyst review. This allows scarce human resources to be devoted to processing video that is most likely to result in the greatest benefit. Prioritisation is based on a combination of the calculated likelihood that the passenger of interest will be present within the video segment, and the effect that a positive identification within the video segment will have on isolating ambiguities associated with the incident or passenger's routing.
When analysing a journey taken by a suspect, relevant video footage from different CCTV cameras can be automatically spliced together to present a thorough record of the individual passenger’s journey.
Video Surveillance and Facial Recognition
The Innovation Science Rapid Passenger Tracking capability complements and enhances the value of any video based security solution. Significant effort is being spent by numerous organisations around the world in developing facial recognition and other biometric identification technologies. While impressive advances have been made in facial recognition over recent years, it remains impractical to reliably utilise facial recognition within crowded environments such as railway platforms to track the journeys of millions of passengers.
The Innovation Science Rapid Passenger Tracking capability massively simplifies the facial recognition problem. The solution isolates small segments of video from potentially thousands of cameras distributed around the rail network, and associates each video segment with one or more passengers that are of interest to an investigation. Facial recognition software then only has to consider a handful of potential faces for any video segment instead of a million or more that are somewhere on the network at the time.
Where To From Here?
To discuss how the Innovation Science Rapid Passenger Tracking could be applied to your city’s mass-transit rail network, or to enquire about licensing the technology to enhance your organisation’s video surveillance and facial recognition capability, contact us by phone or email.
Point of Contact:
Michael Haddy
Innovation Science Pty Ltd
109-110 Balcony Level, Gay’s Arcade
Twin Street
Adelaide SA 5000
Australia
ph. +61 (0) 8 7127 3516
[1] “Statement of Mike Brown, Chief Operating Officer, London Underground”, Senate Committee on Homeland Security & Governmental Affairs, 21 September 2005, USA, p5., (http://hsgac.senate.gov/public/_files/092105Brown.pdf, accessed 3 June 2011).
| 
