Trial control for community anxiety, suicide prevention in metro London, Montreal

Subway systems in the UK and Canada are experimenting with computer vision cameras to prevent crime and suicide, but with no facial recognition, public safety initiatives are going undercover.

Arms, 'anti-social behaviour' among warning triggers for Tube pilot.

In September 2023, the London Underground demonstrated a 'Smart Station' proof of concept at Willesden Green underground station, using AI to provide insights from video analytics and real-time data on customer behaviour. The pilot's final report identifies design principles for further iterations of the system, and from counting customers entering and exiting, whether patrons are evading tolls, leaning into the driveway, vaping, sitting on seats too long or unlocking their e-scooters.

Although the partially amended document indicated that facial recognition would not be implemented through the Smart Station platform, the pilot introduced additional requirements for the fare evasion case, including testing facial image blurring to identify repeat offenders.

According to a Wired report, municipal transit operator Transport for London (TfL) tested 11 algorithms that generated more than 44,000 alerts, 19,000 of which were delivered to staff in real time. It also catalogs a list of expected objections and concerns from privacy and consumer advocates concerned about the accuracy of a system that requires capturing the nuances of behavioral biometrics that are prone to misinterpretation.

Warning signs tell the AI ​​that a Metro patron may be suicidal

A much more altruistic program is testing computer vision and behavioral analysis tools for suicide prevention in the Montreal metro system. CBC reports that researchers from the Société de transport de Montréal (STM) and the Center for Suicide Intervention (CRISE) are working with artificial intelligence to scan CCTV footage for signs of people in distress.

Machine learning capable of recognizing warning signs can send real-time alerts to operators, who can take action to reduce damage. According to Brian Mishara, director of CRISE and professor of psychology at the University of Montreal (UQAM), the algorithm can currently identify one in four people who attempt suicide based on behavioral symptoms.

Mishara says the system is a cheaper alternative to physical barriers or screens, which are still on STM's wish list but cost millions. The organization says it hopes to implement the AI ​​system within two years.

Topics of the article

AI | behavior analysis | computer vision | London | Montreal | Transportation in Kazakhstan | video analysis