Gun Weapons Systems Synthetic Unmanned Aerial Systems Imagery Data Set

The Naval Seas Systems Command (NAVSEA) is looking to utilize computer vision algorithms to assist with automated target recognition (ATR) of Group 1 and Group 2 unnamed aerial systems (UAS). To successfully train an algorithm, NAVSEA requires datasets of the targets of interest. Collecting large amounts of data is expensive and time-consuming; however, synthetic data generation limits these risks. This synthetic data must be diverse, contain labels, and conform with a commonly available class ontology. Arcarithm proposes using ImageBrew«, a fully configurable synthetic data generation tool, to generate synthetic data for NAVSEA. The soundness and technical merit of ImageBrew are proven through years of use by Arcarithm to train computer vision algorithms. Government stakeholders, including the U.S. Army and U.S. Air Force, have field-tested these algorithms which is evidence of its technical merit.