Information engineering

What needs to be done to improve cyber security for individuals, businesses, government and national infrastructure?

Some of our research in cyber security:

Surveillance of public spaces

Intelligent video surveillance using CCTV systems is a necessary tool for agencies to ensure public safety and protect public assets and critical infrastructure. Commercial ports, railway stations, and other transport hubs around Australia are at constant risk from security incidents that can jeopardise public safety and halt operations.

UQ is developing and deploying advanced technologies for real-time video analysis in order to proactively identify and track people and vehicles, on land and water.

3D video analysis from calibrated cameras will become available for incident detection and recognition of abnormal behaviour. Our aim is to use advanced video analytics as a fully integrated, world-class command, control, and surveillance system for port and rail security that is powerful, cost-effective, and easy to manage.

Records and database security

With computer hacking activity becoming increasingly sophisticated, managing to remain one-step ahead in the software and internet security game is a major challenge for the industry. By investigating ways to close security loop-holes in applications that are built using the most common computer languages, UQ researchers are assisting computer technology corporation Oracle to meet this challenge.

We are tackling bugs in Parfait, an internal Oracle product used by thousands of developers daily, to ensure that applications are reliable and secure from hacking. Our research uses static analysis to find bugs in C, C++ and Java system code such as used in operating systems, databases and virtual machines to greatly improve the quality of thousands of applications that use C and Java programming languages.

Biometrics

Biometrics authentication is used in computing systems as a form of personal identification for authentication and access control in border control, banking, and shopping transactions. Remote biometrics such as face-in-the-crowd recognition may also be used to identify individuals in crowds that are under surveillance.

Current UQ research aims to take biometric systems into the Internet of Things so that household devices can instantly recognise persons they interact with. For example, the refrigerator could recommend what you should eat for dinner based on your personal preferences. Perhaps your car can recognise the driver by their face and instantly change the mirrors and seats to suit your preferences. These applications require fundamental research on reliable and robust biometrics as well as strengthening security via anti-spoofing and cyber security technologies.

How do we interact with technology?

Some of our research in interacting with technology

Human-computer interface

UQ researchers are reimagining ways that people physically and socially interact with the digital world. We are creating new possibilities for how children interact with digital information and technologies as interactive digital systems can support exploratory learning in primary school students to enhance literacy and numeracy learning outcomes.

For example, we’ve developed programs where children can fold their own origami characters (such as a bird or snake) and as the computer recognises the markers, these animals will appear as an animation in the virtual jungle on the screen.

This example is part of our larger ambition in understanding how interactive technologies can be incorporated into people’s everyday lives to enhance how people live, learn, work and play.

How can the global convergence of information and communications technologies work for everyone?

Some of our research in the global convergence of Information and Communication Technology (ICT)

Large-scale databases (Big data)

Big data is being generated by everything around us. The challenge we now face is how can we extract meaningful value from big data sets for useful applications.

One area where UQ is making an impact is through a novel project aiming to make effective use of the massive amounts of data collected by health care providers using problem-specific analysis techniques.

Our research aims to develop techniques for monitoring prescribing patterns of medications that are essential in pain management but have high addiction potential. Outcomes will include new methods for automated detection of anomalous prescribing of controlled drugs, presented in an interactive dashboard, and will lead to improvements in patient outcomes and greater efficiency of the health care system.

Multimedia and pattern recognition

Multimedia exists almost everywhere and has increasingly become the “biggest big data” in a wide range of application domains, including security, transportation, commercials, entertainment, games, arts, health, education, online and social services.

The groups research on multimedia and pattern recognition includes designing innovative methods to collect, manage, index, analyse, and search big multimedia data, with scalable algorithms to recognize and understand multimedia content. The impact of their research has been demonstrated by the practical systems that have been developed for many important applications, including real-time visual search engines to find near-duplicate contents and effective content recognition tools to automatically identify and search objects, scenes, places, and events from millions of images and videos.