DARPA threat detection technology uses a camera to see targets, software and soldier brains to identify them
DARPA aids our military in myriad ways, from designing one shot, one kill weapons to creating robotic pack mules to carry soldiers’ gear. It’s also been building tools for soldiers to better survey their environment and identify threats, and its latest such tool is called the Cognitive Technology Threat Warning System (CT2WS). CT2WS is comprised of a 120-megapixel electro-optical video camera with a 120-degree field of view feeding a laptop running cognitive visual processing algorithms. Those algorithms identify potential targets in the video feed, which are shown to a soldier wearing an EEG cap that monitors brain signals. You see, the human brain is particularly good at perceiving threats, and CT2WS looks for the particular brain wave that occurs when we see one. The human component drastically improves the accuracy with which the system can identify enemies from afar. How accurate? Testing in desert, tropical and open terrain showed that without a solider/EEG filter, the system had 810 false alarms out of 2,304 threat events in an hour. Incorporating the filter resulted in only five false alarms per hour, plus it was able to identify 91 percent of the potential targets successfully. Not good enough, you say? Add commercial radar into the mix and the army becomes omniscient — the system then identified 100 percent of the test targets.