Deep learning algorithm to aid in Army intelligence gatheringNews
July 31, 2019
ADELPHI, Md. Scientists at the U.S. Army's corporate research laboratory are developing a new algorithm that could improve image and audio identification for intelligence gathering on the battlefield.
The U.S. Army Combat Capabilities Development Command Army Research Laboratory is developing a deep-learning algorithm called a shortcut autoencoder that can restore single audio clips and images corrupted by various types of random noise.
According to researchers, these advancements differ from previous studies because of improved applicability to 1-D signals like human speech, and are testing against stronger noise sources than usually considered.
The lab also claims that target Army applications are numerous, including eavesdropping, demodulating communications in the presence of strong jammers and perception of objects in image/video that are obscured intentionally, by darkness (low-light) or by weather events such as fog and rain.