TweakTown
Tech content trusted by users in North America and around the world
6,012 Reviews & Articles | 38,778 News Posts
TRENDING NOW: Windows 9 logo teased, with Microsoft's next OS 'coming soon'

FBI to spend $1 billion on next-generation facial recognition system

FBI to build next-gen facial recognition system, would identify someone under varying conditions at

| Hacking & Security News | Posted: Sep 10, 2012 6:29 am

If all the existing cameras on our streets, front-facing cameras on our smart devices, and even the ones that are now being baked into our TVs aren't enough, the Federal Bureau of Investigation (FBI) are spending $1 billion on a next-generation facial recognition system.

 

TweakTown image news/2/5/25729_07_fbi_to_spend_1_billion_on_next_generation_facial_recognition_system.jpg

 

This new next-gen system would be capable of identifying someone under various conditions with as much as 92% accuracy. The Next Generation Identification (NGI) pilot program was launched two years ago, and if the results are anything to go by, the best algorithms were capable of narrowing down someone's identity 92% of the time, from 1.6 million mugshots, impressive.

 

The person doesn't even need to be looking directly at the camera, as the technology is able to match the person to the available mugshot in the database using various biometric analyses. Said algorithms can analyze features on front and side views of mugshots, create a 3D model of the face, and even rotate the model as much as 70 degrees in order to match the angle of the face in the photo.

 

The program is meant to go into action by 2014.

NEWS SOURCES:Vr-zone.com

Related Tags

Further Reading: Read and find more Hacking & Security news at our Hacking & Security news index page.

Do you get our news RSS feed? Get It!

Got an opinion on this news? Post a comment below!

Latest Tech News Posts

View More News Posts
Check out TweakTown Polls on LockerDome on LockerDome

Forum Activity

View More Forum Posts

Press Releases

View More Press Releases