A Queensland University of Technology (QUT) engineer is developing techniques to automatically identify people in surveillance videos and recognise their movement and behaviour.
The explosion of video surveillance to make public places safer, says Dr Clinton Fookes of the University’s School of Engineering Systems, has created a new challenge for researchers—to make sense of what cameras and computers see. So he is investigating ways to extract and interpret important information from these visual sources.
The data generated by the proliferation of surveillance cameras, as well as the countless images and videos online, he says, are impossible to intelligently use without sophisticated computer vision technology that can automatically extract information from these sources, collate and report on it in real time.
As Clinton’s work is ideally suited to improving security in public places such as airports, one of his roles is technical director of QUT’s Airports of the Future—a major research project aimed at improving the experience of passengers passing through Australia’s airports.
His research in this field could lead to new discoveries in a range of areas including human-computer interaction, security, medical imaging and robotics.
Photo: Clinton Fookes is technical director of QUT’s Airports of the Future.
DNA barcodes could help farmers and conservationists identify wanted and unwanted grasses.
Identifying grasses is difficult especially when they’re not flowering. But identification is important. Australia’s agriculture and ecology are threatened by invading grasses, such as Chilean needle grass (Nassella neesiana) and serrated tussock (N. trichotoma). And efforts to re-introduce native grasses can be hampered if you can’t tell the grasses apart.