Large-scale surveillance

Security cameras help keep us safe, but they produce hours of benign footage for operators to sift through. We've discovered new techniques that allow camera operators to focus on the 1% of events that indicate possible security risks. Our algorithms power innovative anomaly detection software called iCetana, which is now used in video surveillance worldwide.

Surveillance's big data question

We are interested in the problems that arise when examining video from hundreds of cameras. Our research examines the following question: how can we select which camera an operator should look at as a priority?

Solving the question

Focusing on large-scale surveillance, we developed new techniques to model 'normal' data from static video cameras. This allows us to detect real-time 'abnormal events' in the footage, so operators can focus on the 1% of events in a video feed that indicate risk.

Our research has resulted in new, intelligent software called iCetana, which is changing how we look at surveillance big data.

How we detect security threats

iCetana uses video analytics to detect potential security threats in large data sets. Algorithms that we developed drive iCetana's innovative anomaly detection software. The software uses compressed sensing concepts to enable simultaneous surveillance of many cameras deployed in diverse settings.

A local city council first used iCetana to detect anti-social behaviour and traffic violations. It's now being used around the world.

iCetana received the Broadband Innovation Award, Tech23 in 2010 and the 2011 WA Innovator of the Year award.

Find out more about iCetana

Contact us

Centre for Pattern Recognition and Data Analytics
+61 3 5227 8797
Email PRaDA

School of Information Technology
Deakin University
Locked Bag 20000
Geelong VIC 3220