US Military developing thermal face recognition to identify people in the dark

Using thermal imagery, the US military is figuring out how to apply it to facial recognition

Ramy Caspi
Read +
Follow Us

A recently released white paper by Army Research Laboratory (ARL) scientists Benjamin S. Riggan, Nathaniel J. Short, and Shuowen Hu have detailed the military’s efforts in developing thermal imaging devices applied to facial recognition technology.

In today's modern militaries, and in the case of the US Military, multiple platforms have thermal imaging cameras installed to detect heat signatures from humans during low visibility and at night. From jeeps, to armour carriers, and Apache helicopters, thermal imaging cameras are now relatively cheap and readily available.

In the paper, the researchers say:

“When using thermal cameras to capture facial imagery, the main challenge is that the captured thermal image must be matched against a watch list or gallery that only contains conventional visible imagery from known persons of interest.”

By deploying Artificial Intelligence (AI) technology, and applying deep learning against the captured footage, the researchers believe that the tech does not need to be installed directly onto the camera, rather can be applied against the videos in the archive.

Similar to the recent ‘buzz’ around Project Maven, in which the military wants to review drone video footage, or in the applicational use described above, AI is now being developed in order to identify a person against his facial image.

While the technology is still in its development stages, the program has already shown that using a small data-set, the AI has the ability to pair thermal vs actual facial imagery and with further development, the system could be expanded to conduct real-time facial recognition.

In an article by the Army’s public affairs office, the division has indicated that the technology is being developed to aid in battlefield detection and assisting soldiers with identifying enemy VIPs or individuals on government watch lists.

Should the Military, or perhaps a country’s police department have access to such technology, and if, for example those organisations have a ‘face book’ data-set, then the identification of persons day, or night may prove to be highly valuable when tracking down wanted individuals.

Facial recognition has taken on a new paradigm. With the recent launch of Apple's iPhone X, the technology has now been socially accepted throughout the world and we have various use cases and the deployment from the industrial security complex in China using the tech.