Skip to main content

Advertisement

Advertisement

Facial recognition, detection technology helps locate lost elders in matter of hours

SINGAPORE — While China’s recent use of surveillance footage to nab a criminal among a sea of concertgoers has raised ethical concerns, such facial-recognition technology has also helped locate, in a few hours, older pedestrians who were lost.

Facial recognition, detection technology helps locate lost elders in matter of hours

China-based firm SenseTime’s systems allow the detection and recognition of vehicle and pedestrian attributes, including a person’s gender, age and clothing.

SINGAPORE — While China’s recent use of surveillance footage to nab a criminal among a sea of concertgoers has raised ethical concerns, such facial-recognition technology has also helped locate, in a few hours, older pedestrians who were lost.

This is done through algorithms that let computers pinpoint pedestrians from video footage, based on attributes such as gender, clothing and even age.

The technology, powered by artificial intelligence, was among others on display at the inaugural Singapore Defence Technology Summit at the Shangri-La Hotel.

Using facial-recognition technology, China-based artificial intelligence company SenseTime works with the authorities in places such as Yunnan, Chongqing and Shenzhen to help nab criminals and also reunite seniors with their families.

In January, for instance, the Shenzhen police, aided by the technology, located one such lost person in just three hours, SenseTime’s marketing officer Gao Yuchen told TODAY on Thursday (June 28). Realising that the person had gone missing, family members alerted the police and provided attributes such as the type of clothing worn, age and a photo of the person.

The firm’s system then matched the person’s facial features, attributes and age with people appearing on surveillance footage, before the lost family member was found.

Apart from locating missing persons, the company’s technology has also helped police arrest scores of criminals, Mr Gao said.

In the Chinese city of Guangzhou, a public-security bureau uses a surveillance system developed by the company to compare and run crime-scene facial images with photos in its database to establish the identities of suspects. Since last year, the system has helped identify more than 2,000 suspects, captured more than 800 people and solved nearly 100 cases.

In April, Chinese police arrested a 31-year-old crime suspect wanted for “economic crimes”. He was among a 60,000-strong crowd at a concert in Nanchang, China.

Several cameras at the venue’s ticket entrance had been equipped with facial-recognition technology, and the system was linked to a national database. The suspect was identified and the authorities were notified.

While it raises the efficiency of law enforcement agencies, such technology has thrown up major ethical concerns, with human-rights groups speaking out against the mass surveillance of citizens.

SenseTime’s systems also allow the detection of vehicle attributes, right down to the make and model of cars and the licence-plate numbers, from video footage. It also recognises various features of pedestrians, including their gender, age, the colour of their clothing, and whether they are wearing long- or short-sleeved clothes.

Among the other exhibits at the three-day summit, which runs until Friday, is a new fleet management system being developed by the Defence Science and Technology Agency and set to be introduced in stages across Singapore’s army, navy and air force by year’s end.

In the air force, for instance, the system will help uncover failure trends and patterns so that preventive measures can be put in place quickly to raise the combat-readiness of aircraft.

For instance, if certain aircraft components are found to be the leading causes of instrument failure, these can be strengthened.

Read more of the latest in

Advertisement

Popular

Advertisement

Stay in the know. Anytime. Anywhere.

Subscribe to get daily news updates, insights and must reads delivered straight to your inbox.

By clicking subscribe, I agree for my personal data to be used to send me TODAY newsletters, promotional offers and for research and analysis.

Aa