Wir lernen Ihre Sprache noch!

Wir arbeiten fieberhaft daran, alle Seiten auf milestonesys.com in so vielen Sprachen wie möglich verfügbar zu machen. Aber dieser Prozess braucht Zeit. Während alle unsere Funktionen mehrsprachig sind. Einige Seiten, wie diese hier, sind noch nicht in Ihrer Landessprache verfügbar.
Vielen Dank für Ihr Verständnis!

OK, ICH HAB‘S VERSTANDEN!

Academic study: Intelligent video technology can save lives

April 25, 2024

Video technology can save lives – the proof is in this award-winning research project.

Researcher Neelu Madan has received the Spar Nord Foundation’s esteemed Research Prize for her Aalborg University PhD project exploring the use of artificial intelligence (AI) and video surveillance, in close collaboration with Milestone Systems.

With funding from Milestone, a leading provider of video technology, Neelu and the research teams developed a model that can train itself, i.e. self-supervise, to monitor and assess abnormal behavior and prevent accidents, for example. near water.

In Denmark alone, city harbors have unfortunately witnessed numerous drowning incidents over the years. From 2001 to 2015, a staggering 1,647 lives were lost in Danish waters, with a quarter of these tragedies occurring in harbors.

Neelu Madan’s award-winning research demonstrates how fatal accidents can be detected using AI-trained models for thermal cameras and Milestone’s Video Management Software (VMS).

“By knowing the normal behavior on the waterfront, the model can detect abnormal events such as a fall off the dock and immediately trigger an alarm. That way, rescue personnel can be called out quickly – even in cases where the accident occurs without witnesses,” explained Neelu Madan.

The primary challenge faced by Neelu was training the system to identify instances where a person was too close to the waterfront or fell into the harbor. This was achieved using live models, dummies, and synthetic data – information that is artificially generated using video game software to simulate real-world events.

According to Neelu Madan, the same model could be used in other situations. For example, it could be used by manufacturing companies to improve production processes or enhance diagnostics in the healthcare industry.

“In a production company, for instance, it is absolutely crucial that automated processes proceed as they should. If not, the final product may end up with undetected errors and that can be expensive in many ways, not least financially,” she said.

The partnership between Milestone Systems and Aalborg University (AAU) has demonstrated the possibilities of combining video surveillance and AI to the benefit of society, said Rahul Yadav, Chief Technology Officer at Milestone.

“We are incredibly proud of Neelu and excited about the potential application of her work. The development of the AAU/Milestone thermal dataset at Aalborg Harbour represents a groundbreaking step forward in video technology, providing the community with the largest annotated thermal dataset.” he said.

Bo Uggerhøj, fund director at Spar Nord Foundation, said that the research has already been noted around the world.

“In the Spar Nord Foundation, we want to highlight researchers who, with passion and commitment, contribute to finding new solutions to important problems,” he said.

About Milestone Systems

Milestone Systems is a leading provider of data-driven video technology software in and beyond security that helps the world see how to ensure safety, protect assets, and increase business efficiency. Milestone enables an open platform community that drives collaboration and innovation in the development and use of network video technology, with reliable and scalable solutions that are proven in more than 500,000 customer sites worldwide. Founded in 1998, Milestone is a stand-alone company in the Canon Group. For more information visit: www.milestonesys.com. For news and other press releases, visit our Newsroom.

Tags
Lars Lemche
Global External Communication Manager

Global 

lle@milestone.dk

You will be logged out in
5 minutes and 0 seconds
For your security, sessions automatically end after 15 minutes of inactivity unless you choose to stay logged in.