Quantum Leap in Video Analytics for Large Crowd Surveillance

Quantum Leap in Video Analytics for Large Crowd Surveillance

Photo illus. crowd by US National Archives
A big crowd of spectators at the National Victory Celebration parade honoring the coalition forces of Desert Storm.

This post is also available in: heעברית (Hebrew)

Video analytics surveillance have accomplished a quantum leap thanks to artificial intelligence (AI) applications, which have turned surveillance into a smart technology and have substantially decreased the human and time resources required. A new video analytics solution acquired by the Canadian government will provide advanced object recognition capabilities that can provide a higher level of safety to people on the streets, as well as the surveillance of large crowds of people through integration with drones.

Canada’s Department of National Defense (DND) has awarded Patriot One Technologies a contract to further develop an artificial intelligence-enabled full-motion video analytics application. This technology will be used to detect, classify and track objects and events of interest, initially for DND and the Canadian Armed Forces applications, to improve situational awareness. In the future it will be applied across multiple other domains. 

The contract is worth nearly $1 million Canadian dollars. The project, named “RECCE,” is being developed by Xtract AI, the AI innovation and development arm of Patriot One, and the solution will be delivered to DND by March 2022.

The technology provides high threat detection and categorization in environments where people, cameras and other elements are in motion, according to the company’s announcement. 

The company integrates advanced AI analytics into its products developing technologies suited for enhanced crowd or threat detection management for large entertainment venues.

The RECCE project is being funded as part of the DND’s Innovation for Defense Excellence and Security (IDEaS) program and is sponsored by the Canadian Joint Operations Command (CJOC).