Automated systems for analysing road conditions and road safety are being developed.
Researchers aim to adapt ‘image pattern recognition’ software for that purpose.
The present manual systems used for road safety, not only in Australia but around the world, are inefficient and prone to many errors, Central Queensland University professor Brijesh Verma says.
“The major challenges are to accurately detect, segment and classify all road objects and also calculate the distance between objects,” he says.
“A recent software breakthrough involving ‘deep learning’ has the ability to address such major challenges.”
Verma says digital video road data is collected over every state road in Queensland through vehicle-mounted video and mobile laser scanning. This has the potential to provide a range of value-added products for safer roads, he says.
New identification techniques and software can then be incorporated with road data collection systems.
Automatic analysis of digital video road data for road safety and conditions will reduce accidents, improve road asset management and make Australian roads much safer, with fewer fatalities, Verma says.
Successful outcomes could find automated solutions for road safety rating required by national and international safety and risk assessment models, such as the Australian Road Assessment Program (AusRAP).
The research project will be supported by an Australian Research Council Linkage grant worth around $390,000.
Of that funding, $240,519 comes from the Australian Research Council and $150,000 from the Queensland Department of Transport and Main Roads.
Join the conversation