Every year people and objects on the tracks cause serious accidents, driver trauma and unnecessary stops for metro and rail traffic around the globe. In 2019 alone, there were 1 516 significant railway accidents in the EU, with a total of 802 fatalities and 612 persons seriously injured, according to Eurostat.
To increase personal safety and security for all travelers at their stations, Stockholm Metro (SL) has decided to invest in the new AI-based surveillance system IRIS™ Rail that will send real-time alerts whenever someone or something enters the track area. This will in time enable the Swedish rail operator to take quick actions and might save many lives. During a first stage SL is now testing the system on recorded video material with a full implementation scheduled to take place this fall.
“We are excited to be on this journey with SL to improve safety in Stockholm Metro stations. We believe in delivering highly reliable and precise analytics to the rail industry in order to help prevent serious accidents and save lives”, says Victor Hagelbäck, CPO at Irisity.
IRIS™ Rail by Irisity is an AI-based system, deep-learning trained to analyze and filter streaming video with the highest possible precision and reliability. It detects human movement or objects on tracks and will immediately send alarms to a safety central whenever an incident is detected within specified zones. An operator at the safety central can then make sure to take the right action such as calling a guard or potentially even stopping the train. The system knows the difference between a human, animals, objects and trains and will not send alarms on normal activity such as trains entering or exiting the station or passengers boarding the train within the specified zone. It is a camera-agnostic analytics compatible with any existing or new infrastructure and can operate fully offline to meet any data security demands if needed.
“There are many benefits of using AI-technology in video surveillance. It is like we implement an AI in front of every camera, that never gets tired and that always has 100% attention”, comments Hagelbäck.
“When it comes to personal integrity protection AI is often also preferable, as it always follows a firm set of rules. As such AI never discriminates, it does not have any preconceived notions and does not make assumptions based on gender, ethnicity or class”, Hagelbäck continues.
At the vanguard of rail infrastructure safety, subway and rail operators globally are eager to implement AI-based analytics. In addition to Stockholm Metro (SL), subway and rail operators in several countries internationally including France, Italy, Spain and Chile are also looking into implementing IRIS™ Rail.