Panasonic unmasks deep learning facial recognition technology

Panasonic Business has used IFSEC 2018 to launch its new deep learning facial recognition technology in Europe.

The upgraded platform features a ‘deep learning’ core engine that has ranked as the industry’s most accurate in independent testing by NIST. The Face Server can identify faces that are difficult to recognise using conventional facial recognition technologies, including faces at an angle of up to 45 degrees to the left or right or 30 degrees up or down, and those partially hidden by sunglasses and face masks.

When combined with Panasonic’s iPro Extreme camera range, the Face Server technology makes use of the ‘iA (intelligent Auto) mode’ and ‘best shot feature’. This means the camera automatically adjusts its settings to shoot optimal images. The iA function enables image analysis to be performed on the camera instead of the server. As a result, only the best images are sent, reducing server and network load and total system cost.

Jointly developed with the National University of Singapore, it provides a real-time processing capacity of up to 20 cameras per server and can execute high-speed searches of up to 30,000 registered reference faces.

“Face Server changes the game in terms of detection accuracy,” said Gerard Figols, Category Manager at Panasonic Security. “Not only does it maintain 90%+ accuracy for faces that are partially covered by sunglasses or face masks, it also recognises faces from photographs that are up to 10 years old.”

The platform is already in use at Tokyo’s Haneda airport where it has replaced finger print recognition on the passport entry gates, following three years of verification testing.    

For more information on Panasonic Security Solutions visit Stand No: F608 at IFSEC 2018 (19th – 21st June), or visit http://business.panasonic.co.uk/security-solutions/

For more information on the Tokyo Haneda airport installation, please see:  https://news.panasonic.com/global/press/data/2017/12/en171215-2/en171215-2.html