TerraSense Analytics Receives Funding for Advanced Airborne Surveillance System

Science fiction movies have chase sequences where cameras pick up the protagonist as he or she is trying to escape from authorities. We know that, given the current limits of technology, this isn’t possible just yet. However, remote sensing company TerraSense believes it can develop technology that might be able to help national defense systems be more aware of threats that may exist. The company received funding to the tune of $977,439 from the Innovation for Defence Excellence and Security (IDEaS) program, an initiative of the Canadian Department of National Defense. The funding was secured to allow TerraSense to accelerate work on its Multimodal Input Surveillance & Tracking (MIST) system.

Airborne Surveillance and Reconnaissance

MIST is as futuristic as current-level technology gets. It’s a system that utilizes the video feeds from surveillance aircraft, combining the input into a composite image. This image also takes into account thermal and color imagery and can be used to track targets in real-time. Additionally, the system will use deep learning to help it identify current and potential future threats. TerraSense expects to spend the rest of this year perfecting its working prototype of MIST. The team states that by the end of the year, they would like to be able to have the software utilize these individual data streams from both Canadian surveillance aircraft as well as international assets to identify and track targets.

Using New Technology

As the roles of surveillance and reconnaissance become more critical to the Canadian Armed Forces, older aircraft are also being outfitted with camera packages that can be used to collect data for MIST. The aim is to shift data processing from a manual annotation system into one that takes advantage of the power of deep learning. Instead of having humans spend hours poring over incoming data streams and annotating threats manually, MIST intends to do so as it ingests and processes streams. Building on its successes, the system will get iteratively better at locating and informing authorities of potential problems before they arise.