Patriot One’s Xtract AI Division has been Awarded a Department of National Defence Contract

AI team based in Vancouver to work with the Canadian Armed Forces

VANCOUVER, British Columbia, May 27, 2020 (GLOBE NEWSWIRE) -- Patriot One Technologies Inc.’s (TSX: PAT) (OTCQX: PTOTF) (FRANKFURT: 0PL) wholly-owned subsidiary Xtract Technologies (“Xtract AI”)  is pleased to announce it has secured a $199,000CAD contract with Canada’s Department of National Defence through the Public Works and Government Services Canada Division, as part of the Innovation for Defence Excellence and Security (IDEaS) Program.

Operating covertly is key for Army operations and the “Now you see me, now you don’t” project is looking for solutions to actively modify the visual and infrared signatures (the pattern emitted by an object) of Army platforms in order to decrease detectability. Xtract AI proposes to leverage recent advances in deep learning and computer vision, along with novel deep learning technology to develop a system capable of assessing the efficacy of any existing or developed adaptive and multi-spectral concealment and camouflage technologies or materials. This will guide the development of technologies that optimally reduce visual and infrared detection of soldiers and vehicles, and enhance the performance of materials designed to eliminate detection in dynamic environments in real time.

The proposed system will combine two deep learning components: a soldier and vehicle detector; and a soldier and vehicle concealer. The detector will be a computer vision model that is able to identify people and vehicles from visual and infrared video streams. Meanwhile the concealer will be a computer vision model that uses real-time information from the context (e.g., background visual and heat patterns) to determine the optimal visual and infrared signature to facilitate concealment.

Using these two components, several outstanding problems with soldier and vehicle concealment can be addressed. Concealment systems will be able to be assessed, for example the efficacy of new uniforms, textiles and vehicle add-ons. Pitting the two components against each other will also facilitate the development of optimal concealment methods. Additionally, the concealer will be trained to generate modified visual and infrared signatures that are able to prevent detection by the deep learning detection system.