Korea uses AI to prevent people from committing suicide on the Han River
An artificial intelligence (AI) system has been put into use to help reduce the number of suicides at 28 bridges spanning the Han River.

The 28 bridges across the Han River in Seoul, South Korea, are suicide hotspots. According to The Korea Times, the number of people who attempt suicide at these bridges is up to 500 people a year.

Previously, a series of safety measures were put into operation. For example, the Mapo bridge has been equipped with a pressure sensor system along the railing to help detect people holding their hands more tightly than usual, along with a safety fence to make it difficult to jump into the river.

In an effort to reduce suicide rates on bridges along the Han River, the Seoul government put in place a network of cameras that use machine learning to detect people who are attempting to jump from the bridge through their behavior. .

Accordingly, the system will identify people who linger on the bridge for a long time and conduct further validation through data from other monitoring records in the vicinity. Once identified, notification will be sent immediately to rescue units standing around the bridges.

Upon receiving the notification, the rescue workers will immediately coordinate manpower to deploy the prevention, or rescue when the person has jumped the bridge.

The technology has been developed over the past year and a half by the Seoul Metropolitan Fire and Disaster Headquarters, in collaboration with the Seoul Institute of Technology.

“The most important thing in rescuing people who are trying to commit suicide is to stop them before they jump. Once that has happened, the survival rate drops below 50%,” said a spokesman for the Seoul Metropolitan Fire and Rescue Agency.

“We have to monitor 572 cameras in the control center, so it’s not easy for a limited number of surveillance officers to capture everything. But now the AI ​​system will select the footage of a person with the most suspicious behavior and sound an alarm so that the rescue team can react much faster than before,” he shared.

Dao Vu (According to Thanh Nien, Cafebiz)