New computer vision technologies detect enemy drones faster

Standard unmanned aerial vehicles (UAVs) carrying a payload of explosives or biological material can enter a crowded building or military base. The ability of terrorists or enemy armed forces to send such “packages” is called the “future of war.” University of Technology Sydney (UTS) and DroneShield have developed next-generation drone technology to better detect threats from aggressive UAVs.

Through the partnership, UTS and DroneShield, an Australian developer of counter-drone solutions, have created an optical system to detect, identify, and track rapidly spreading threats such as enemy UAVs. It consists of a camera and a super-precise neural network (CNN).

UTS and DroneShield began working together in October 2019 – just a month after the aggressive use of drones against oil facilities in Abqayk Hurais in Saudi Arabia.

The new technology was recently demonstrated at the Sydney Science Park.

UTS Vice Chancellor Attila Brungs said the project is an excellent example of what the university is striving for: creating and applying technologies that can be used “here and now”.

“We are using CNN and deep learning to provide DroneShield with a solution to identify potential threats to drones,” explains project manager Nabin Sharma. – The algorithm allows the computer vision system to see what is happening, compare the data, and process it for ultra-fast object recognition and image analysis. This allows for near-instant and effective threat assessment and decision making in response to it. The system is capable of detecting different types of drones and checking for the presence of a payload. ”

The new development could save many lives in the event of a terrorist attack, the scientists conclude.

or as guest:
Comments: 0