Guarding the Tracks in Montreal

Oct. 7, 2013
Successful deployment of a video-based intrusion detection system on such a large scale is difficult in a metro rail environment

The Montreal Metro is Canada's second busiest subway system, and North America's fourth busiest in total daily passenger usage, delivering an average of 1,241,000 daily unlinked passenger trips per weekday. The Metro is operated by the Société de transport de Montréal (STM), Montreal’s metro transit authority.  STM also operates one of North America's largest urban bus and rail rapid transit schemes, serving the third-largest number of passengers overall behind New York and Mexico City, and attracting the second-highest ridership per capita behind New York

In 2008, STM sought to improve public safety by automating its detection of people intruding on the tracks and into the tunnels of the Metro system.  STM already had a Loronix Windows-based software management system running its video surveillance system, with more than 300 of the several thousand cameras viewing the Metro tracks and tunnels.  The next logical step for STM was to migrate to an intrusion detection solution using video analytics software that also ran on Windows servers that would fit well into its existing infrastructure and information technology skills sets.

Successful deployment of a video-based intrusion detection system on such a large scale is difficult in a metro rail environment, where camera views are often filled with the intense motion and illumination variations of a train.  Such conditions usually render motion-based video analytics temporarily useless. 

Vidient Systems of California, however, had successfully deployed its SmartCatch video analytics software running on Windows servers for rail intrusion detection on a small scale.  SmartCatch already had advanced capability to reliably recognize a walking or running human being – even with only 60-70 pixels on the target.  Since Vidient had committed to solving the problem of dominant train motion, STM contracted them to help solve the issue at the Metro.

STM was well aware that the performance of video analytics for intrusion detection is sometimes a trade-off between the probability of correctly detecting an intrusion and the rate of false alarms.  STM realized that too many false alarms would cause operators to lose confidence in the SmartCatch solution, and even cause some operators to turn the system off. 

Many video-based intrusion detection installations can tolerate up to one false alarm per camera per day. But with almost 300 cameras running video analytics in the Montreal Metro, a much lower false alarm rate was essential.  Even if these cameras were generating one false alarm per camera per day, there would be a flood of false alarms -- one every 4-5 minutes on the average – perhaps worst at busy train times because train motion is the primary trigger of false alarms.  This would be an intolerable false alarm rate, considering STM needed to maintain at least a 90 percent-plus correct detection rate.

Normal operations in Montreal generate about 200 authorized intrusions by work crews every day, giving a very useful ongoing check of sensitivity.  The Montreal Metro authority required that there be no more than 0.5 false alarms per camera per day – about 150 for the whole system.  Thus correct detections would at least be noticeably greater than false alarms.

For almost a year, Vidient and STM cooperated in the development of improved SmartCatch software. The goal was to successfully detect a human intruder on a catwalk next to a moving train that filled most of the camera’s view, while generating a minimal number of false alarms. The final hurdle came during a week-long acceptance test in August, 2009, and again in the summer of 2011. The result demonstrated a false alarm rate of 0.15 to 0.20 false alarms per camera per day.  The ratio of correct detections to false alarms was at least 5 to 1, providing the systems operators the confidence they needed as the probability rate averaged approximately 95 percent.

STM insists it has made notable improvement in its public safety environment with the SmartCatch deployment.  According to Claude Ouellet, professional engineer and manager of telecom operations engineering at STM, “The principal reason to use SmartCatch is to be notified right away when a person gets onto the tracks from the platform, or is getting into the tunnels.  When this happens, our control center can immediately turn off the power on the rails, tell the people to evacuate with speakers in the station, and get our agents on site immediately. 

“This may prevent people from getting electrocuted or hit by an oncoming train.  It also reduces the time for our police officers to get into the tunnel to arrest the person.  Every time a person gets into the tunnels also causes a service interruption of many precious minutes.  SmartCatch is helping us to reduce this time, and better protect our public.”

SmartCatch has also been quite useful for detecting graffiti vandals who enter train yards to tag parked trains.  The system has nearly negated this threat, providing STM prompt detection and almost immediate arrests.

SmartCatch combines advanced computer vision with the detailed knowledge of human operators.  While SmartCatch can filter out trains, handle fast-changing illumination, and distinguish a person from other motion, skilled staff must precisely define various regions of interest in each field of view.

 “When setting up a new SmartCatch system, it is very important to precisely define the borders of the zones that need to be monitored for intrusions in order to prevent false alarms when people merely get close to the controlled areas -- for example, the top versus the bottom of stairs that lead from the platforms down to the catwalk in the tunnel,” says Ouellet.

Successful use of SmartCatch also requires thoughtful integration with the operations of multiple departments.  Crews who maintain the video surveillance cameras must constantly inform SmartCatch administrators of activities that might change any camera’s resolution or field of view. This is crucial since the borders of the controlled zones must be carefully aligned to the changes. 

 “Since SmartCatch detects any human intrusion into the tunnels or onto the rails, good processes and employee collaboration are required. The control center must be notified every time normal operations require an authorized track or tunnel intrusion.  If employees forget to advise the control center in advance, it generates an unwanted distraction and loss of precious time for control center operators,” Ouellet explains.

STM’s confidence in its SmartCatch solution led to a recent upgraded, as several thousand video surveillance cameras were migrated to the Verint Nextiva video management software.  Since 2012, SmartCatch has been developed, delivered, and supported by AgilityVideo LLC ( under an exclusive license to all of the SmartCatch technology.

 STM contracted AgilityVideo to integrate SmartCatch with the new Verint Nextiva video management system and to provide ongoing technical support.   This conversion was completed in spring of 2013. STM and AgilityVideo continue to cooperate to maintain and improve the performance of SmartCatch in Montreal.  Several future feature improvements are being considered:

  1. Delivery of SmartCatch intrusion alarms to STM’s overall command-and-control system built by Alstom
  2. Automatic audio warnings from loudspeakers triggered by SmartCatch intrusion alarms
  3. Using electronic access control information to filter out SmartCatch detection of authorized intrusions

About the Author

J. Michael Rozmus ( is the founder and CEO of AgilityVideo LLC (, provider of SmartCatch automated video surveillance and other services related to large-scale video surveillance – especially for the protection of critical infrastructure.  He can be reached at [email protected].