Korea Set to Use AI to Prevent Suicide Attempts on Bridges

The South Korean Seoul Metropolitan Government has announced that a new project is underway to address a large number of suicides on bridges that are taking place on the Han River – the fourth longest river on the Korean peninsula.

Their researchers and emergency services are working on an AI (artificial intelligence) to help detect and reduce suicide attempts on bridges. It stems from how difficult it can be for human surveillance teams to tell whether someone is admiring the view off the bridge or if they have something much more painful on their minds.

Monitoring CCTV footage of bridges along the Han River in Seoul, South Korea, June 30, 2021. REUTERS/Minwoo Park

South Korean researchers from the Seoul Institute of Technology are developing an AI system paired with a new CCTV camera control hub. It has been using deep learning patterns of behavior to identify the “patterns” of people in crisis by analyzing data from cameras, sensors, and the dispatch records of rescue services since April last year.

Based on information from hours of CCTV footage and assessing details such as the hesitation of the person, the AI can then forecast a hazardous situation and immediately alert rescue teams, said Kim Jun-chul, who is the principal researcher on this project.

With this, the question of its accuracy may arise. Prashanth Guruswamy, who is the co-Founder of InstaSafe Technologies – a cybersecurity solutions company, explained how CCTV surveillance camera works by saying, “Behavioral analysis is done after using data collected from an array of sources, including CCTV footage, bridge sensors, previous suicide attempts, information from people who had previously attempted suicide, phone calls, and text messages. This data is then collated to determine a possible hazard”.

According to CTV, the Seoul Metropolitan Government said it has been working on a CCTV surveillance and response system since 2012 that operates on the Mapo and Seogang Bridges spanning the Han River. As the highest number of suicide attempts have been recorded at these bridges, they have been monitoring the bridges 24 hours a day and “proactively responding to suicide attempts.”

Kim Hyeong-gil, who is in charge of the Yeouido Water Rescue Brigade, said to Reuters, “We believe the new CCTV will enable our crews to detect the cases a bit faster and help us head to a call more promptly.”

He monitored real-time footage from bridges on Seoul’s Han River and his team has been working with the researchers to develop the technology that his crew and the Seoul Fire and Disaster Headquarters will be piloting from October.

A report by the Organization for Economic Co-operation and Development (OECD) revealed that with a population of 52 million people in 2019, South Korea had the highest suicide rate, with more than 13,700 people ending their lives. Moreover, a press release by the city stated that almost 500 suicide attempts are reported on 27 bridges over the nearly 500-kilometer-long (300 miles) Han River every year.

“This system allows rapid responses to suicide attempts—both before or after an incident—and minimizes surveillance loopholes. Not only that, but it can also dramatically reduce warning errors thanks to AI’s ability to reflect environmental factors, including illumination levels and weather, as well as characteristics of Han River bridges such as wobbling caused by winds and traffic. As data accumulates, the accuracy of the system will increase further,” the release stated.

Due to the coronavirus pandemic, economic hardships and battles for jobs heightened to a great extent in 2020 as many were laid off and businesses came to a standstill. Consequently, the number of rescue dispatches surged about thirty percent in 2020 compared to 2019. Many of the suicide attempts made in 2020 were by people in their 20s and 30s said the rescue brigade’s Kim.

“The system learns the footage itself, which can bring about improved results by greatly reducing false alarms,” said Kim Jun-Chul.

Spread the word