Army Researchers Develop Facial Recognition that Works at Night

19 April 2018
The system works with thermal imaging in both low light and at night. Source: Army Research Lab

Researchers at the U.S. Army Research Laboratory (ARL) have developed an artificial intelligence and machine vision system that produces a visible face image from a thermal image of a person’s face captured in either low light or nighttime conditions.

The system could be used to enhance current biometric systems or post-mission forensic analysis for covert nighttime operations, ARL says.

Thermal cameras are deployed on aerial and ground vehicles, in watchtowers and at checkpoints for surveillance purposes. Thermal cameras are also increasingly being used as body-worn cameras. The ability to perform automatic face recognition at night using thermal cameras would benefit a soldier by helping identify a person of interest such as someone on a watch list.

"This technology enables matching between thermal face images and existing biometric face databases/watch lists that only contain visible face imagery," said Dr. Benjamin Riggan, a research scientist at ARL. "The technology provides a way for humans to visually compare visible and thermal facial imagery through thermal-to-visible face synthesis."

Typically at night or in low-light conditions, there is insufficient light for conventional cameras to capture facial imagery for recognition without a flash or spotlight, which may give away the position of such surveillance cameras. Thermal cameras don’t require additional light but have difficulty discerning facial features against a gallery database.

The ARL approach uses deep neural networks that provide a non-linear regression model that maps a given thermal image into corresponding visible latent representation and also optimization problem that projects the latent projection back into the image space.

The ARL demonstrated the system in real-time using a FLIR Boson 320 thermal camera and a laptop running the algorithm. The demo showed that a captured thermal image of a person can be used to produce a synthesized visible image in situ.

The next steps are to extend the research under the sponsorship of the Defense Forensics and Biometrics Agency to develop a robust nighttime face recognition capability for soldiers.

To contact the author of this article, email peter.brown@ieeeglobalspec.com

Powered by CR4, the Engineering Community

Discussion – 0 comments

By posting a comment you confirm that you have read and accept our Posting Rules and Terms of Use.
Engineering Newsletter Signup
Get the Engineering360
Stay up to date on:
Our flagship newsletter covers all the technologies engineers need for new product development across disciplines and industries.

Upcoming Events