The Queensland University of Technology (QUT) will assist the Gold Coast Waterways Authority (GCWA) in its smart camera trial that aims to build a detailed picture of the activity on the city's waterway network.
The GCWA has turned to QUT for help in using the technology to collect data on the use of waterways in an effort to help improve planning, network management and safety.
QUT said his research team will use advanced machine learning and statistical methods to estimate the number and types of ships using the waterways.
Combined with information about incidents at sea and weather conditions, QUT said it will identify usage trends and provide insight into future usage patterns and pressure points on waterways.
A second part of the project is to investigate the feasibility of using the same camera technology to develop a more accurate way to measure the speed of ships on the water, the university added.
“We will use the cameras to get a more complete picture of who uses the waterways, where they are going, in which type of vessel or watercraft they sail and how they interact with other users and the environment,” said Hal. Morris, CEO of GCWA.
“This information has not been collected up to this level of detail on the Gold Coast before. It is important because, in order to plan successfully for the future, we need to understand the consequences of population growth and the increasing ownership of boats, so that we can plan for these changes, protect the environment and ensure that locals and visitors continue to have safe access to our beautiful water town “.
The cameras have started rolling out at 20 locations around the Coomera River and southern Broadwater, with GCWA saying that there are several “ready to start breaking” on the long weekend of Australia Day that according to Morris traditionally one of the busiest on waterways.
The cameras continuously take photos at their locations at any time of the week and in all weather conditions.
QUT then uses image analysis to automatically process the photos and advanced machine learning methods to understand which characteristics of the images can be used to identify the ship type and determine the number of users.
“From this we will develop a statistical model that contains additional information about, for example, weather and incidents at sea, to give an indication of the future pattern of use of the waterways,” said QUT project manager, associate professor James McGree.
The project will also investigate whether computers can be trained to recognize ship registration numbers to help identify ships that are sailing too fast, QUT said.Tags: #ArtificialIntelligence, #latestNewsAI, #researchAi, QUT