Tencent develops image-aging technology for facial recognition of missing children

Police in China are planning to use artificial intelligence technology developed by Tencent to add years of aging to photos of children to use facial recognition in the search for abducted children, China Daily reports.

Chen Shiqu, deputy director of the Ministry of Public Security’s criminal investigation bureau, told the news outlet that the system’s accuracy can be as high as 96 percent.

“Even though a child may have been abducted many years ago and has grown up-and his appearance has greatly changed-faces can be matched through cross-age facial recognition,” Chen says.

The technology was tested in a trial launched by Fujian police in 2017, and helped find 500 missing people in just half a year, according to the report.

The official also said the technology was used last month in Sichuan province to reunite a missing child with his family after ten years. He had been abducted by gang members and sold to a family in Guangdong province. The gang reportedly abducted 13 children before being broken up by police, but authorities were unable to locate the children.

Police used facial recognition, and then took DNA tests to compare children to the parents of abduction victims. Police have located 10 of the children so far, seven using the new AI technology.

Chinese authorities launched an alert system developed by Alibaba in 2016, and say child abductions have fallen dramatically since then.

Researchers in Australia have been developing a similar system, attempting to match photos separated by years to identify child trafficking victims. The International Center for Missing & Exploited Children (ICMEC) has also launched a biometric platform to combat child trafficking.

Related Posts

Article Topics

 |   |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Most Read This Week

Featured Company

Biometrics Research

Biometrics White Papers

Biometrics Events

Explaining Biometrics