Modelling and Removal of Noise and Artefacts in Surveillance and Security Video for Forensic Image Analysis and Enhancement. This project spearheads research in advanced digital image and video processing technology, placing Australia at the forefront of both theoretical and applied research to safeguard Australia. It tackles fundamental issues identified in our earlier research in this area and consulting work for Victoria and NSW Police Departments in forensic investigations since 2000. Althou ....Modelling and Removal of Noise and Artefacts in Surveillance and Security Video for Forensic Image Analysis and Enhancement. This project spearheads research in advanced digital image and video processing technology, placing Australia at the forefront of both theoretical and applied research to safeguard Australia. It tackles fundamental issues identified in our earlier research in this area and consulting work for Victoria and NSW Police Departments in forensic investigations since 2000. Although the main investigation focuses on video surveillance and security systems for public safety, policing, crime prevention and border control, the outcomes of the investigation will have other applications, including digital photography for fine-art, medical imaging, picture archiving and communication systems for telemedicine and rural healthcare systems.Read moreRead less
Kernel and Margin Based Machine Learning Algorithms. Certain machine learning algorithms, such as support vector machines, utilizing
the ideas of large margins and kernels have attracted much attention lately
because of their impressive performance on real world problems such as optical
character recognition. We plan to refine and extend such algorithms to a wide
range of different machine learning problems such as gene sequence analysis,
image processing and text classification. Expected ....Kernel and Margin Based Machine Learning Algorithms. Certain machine learning algorithms, such as support vector machines, utilizing
the ideas of large margins and kernels have attracted much attention lately
because of their impressive performance on real world problems such as optical
character recognition. We plan to refine and extend such algorithms to a wide
range of different machine learning problems such as gene sequence analysis,
image processing and text classification. Expected outcomes include the
development of software that allows the solution of hitherto unsolved machine
learning problems, and the ability to solve problems larger than those solvable
by the current generation of machine learning tools.Read moreRead less
Vision Model Based Perceptual Digital Video Coding. Digital video coding and compression is an enabling technology and has diversified applications in audiovisual communications, multimedia computing, digital television broadcast and electronic entertainment industries. The project aims at spearheading research in theory, techniques and implementation of perceptual video coding in order to achieve constant and guaranteed quality in visual communications and services. It will explore a new appr ....Vision Model Based Perceptual Digital Video Coding. Digital video coding and compression is an enabling technology and has diversified applications in audiovisual communications, multimedia computing, digital television broadcast and electronic entertainment industries. The project aims at spearheading research in theory, techniques and implementation of perceptual video coding in order to achieve constant and guaranteed quality in visual communications and services. It will explore a new approach to digital video coding other than the constant bit rate coding techniques which have dominated digital video research for the past four decades. It will form a part of the theoretical foundation and principles for the next generation video coding and compression techniques, and may lead to new standards and practice.Read moreRead less
Intelligent techniques to exploit the dynamic temporal structure in detection of attacks in credit application fraud. Obtaining credit using fraudulent information costs financial institutions billions of dollars. This project develops fraud detection methods in credit applications, working with credit bureau data. Existing fraud detection models are mostly applicable to transaction fraud, rather than application fraud, and are static. Fraudsters however constantly change their method of attack. ....Intelligent techniques to exploit the dynamic temporal structure in detection of attacks in credit application fraud. Obtaining credit using fraudulent information costs financial institutions billions of dollars. This project develops fraud detection methods in credit applications, working with credit bureau data. Existing fraud detection models are mostly applicable to transaction fraud, rather than application fraud, and are static. Fraudsters however constantly change their method of attack. The temporal characteristics of fraud attacks provide an additional source of information that can be exploited to gain increased predictive power. We propose a hybrid intelligent approach to construct models that are sensitive to the temporal dynamics of fraud attacks, and evolve to acknowledge the changing behaviour of fraudsters.Read moreRead less
Ensembles of Collaborative Neural Networks. Artificial neural networks have been used successfully for data mining and control. A neural network ensemble(NNE) is a collection of networks that exhibits properties of self-organization, plasticity, and adaptive behaviour. The aim of this research is to develop an efficient and theoretically sound algorithm for NNE learning. The outcomes of the project will include insights into self-organization of complex NNE and automatic problem decomposition an ....Ensembles of Collaborative Neural Networks. Artificial neural networks have been used successfully for data mining and control. A neural network ensemble(NNE) is a collection of networks that exhibits properties of self-organization, plasticity, and adaptive behaviour. The aim of this research is to develop an efficient and theoretically sound algorithm for NNE learning. The outcomes of the project will include insights into self-organization of complex NNE and automatic problem decomposition and an efficient algorithm for constructing and training NNE. Practical outcomes will include research training for early career researchers and new modelling tools for data mining, robotics and multi-agent systems. The project contributes to the national priority area of smart information use.Read moreRead less
Developing optimal synthesis strategies in distributed expert systems. The aim of this project is to investigate synthesis strategies in distributed expert systems (DESs). Such strategies are used to synthesize multiple solutions to the same task from different experts (either human experts or expert systerms) in order to obtain the final solution to the task. These strategies could be used in a wide application of domains such as insurance agencies and medical diagnosis systems. The expected ....Developing optimal synthesis strategies in distributed expert systems. The aim of this project is to investigate synthesis strategies in distributed expert systems (DESs). Such strategies are used to synthesize multiple solutions to the same task from different experts (either human experts or expert systerms) in order to obtain the final solution to the task. These strategies could be used in a wide application of domains such as insurance agencies and medical diagnosis systems. The expected outcomes are to develop computational strategies, neural network strategies, and case-based strategies for solving different synthesis cases.Read moreRead less
Adaptive learning of spatiotemporal patterns: Development of multi-layer spiking neuron networks using Hebbian and competitive learning. The aim of this project is to develop a method for recognising patterns that change in time. The development of a reliable method that is fast and robust to noise will have wide application in many areas, especially computer speech recognition where timing plays a crucial role. Building-blocks similar to those in the brain (spiking neurons) will be used. Aut ....Adaptive learning of spatiotemporal patterns: Development of multi-layer spiking neuron networks using Hebbian and competitive learning. The aim of this project is to develop a method for recognising patterns that change in time. The development of a reliable method that is fast and robust to noise will have wide application in many areas, especially computer speech recognition where timing plays a crucial role. Building-blocks similar to those in the brain (spiking neurons) will be used. Automatic techniques will be used to teach groups of spiking neurons the differences between sequences of events by adjusting connections between them. The significance of this approach is that it captures information about timing that is missed in existing techniques.Read moreRead less
Adaptive learning in networks of spiking neurons for recognising patterns that change with time. The aim of this project is to develop a method for recognising patterns that change with time. Building-blocks similar to those in the brain (spiking neurons) will be used. Automatic techniques will be developed to teach groups of spiking neurons the differences between sequences of events by adjusting connections between neurons. The significance of this approach is that it captures information abou ....Adaptive learning in networks of spiking neurons for recognising patterns that change with time. The aim of this project is to develop a method for recognising patterns that change with time. Building-blocks similar to those in the brain (spiking neurons) will be used. Automatic techniques will be developed to teach groups of spiking neurons the differences between sequences of events by adjusting connections between neurons. The significance of this approach is that it captures information about timing that is missed in existing techniques. The development of a reliable method that is fast and robust to noise will have wide application in many areas, especially computer speech recognition where timing plays a crucial role.Read moreRead less
Temporal Pattern Learning and Recognition in Neural Systems. This project is relevant to the National Research Priority area of Frontier Technologies and addresses fundamental cross-disciplinary issues of how neural systems learn patterns that change with time, which is at the cutting edge of intelligent processing systems. Applications are in rapidly growing fields of automatic speech processing, robotics, machine learning and intelligent systems, all with applications in areas of economic impo ....Temporal Pattern Learning and Recognition in Neural Systems. This project is relevant to the National Research Priority area of Frontier Technologies and addresses fundamental cross-disciplinary issues of how neural systems learn patterns that change with time, which is at the cutting edge of intelligent processing systems. Applications are in rapidly growing fields of automatic speech processing, robotics, machine learning and intelligent systems, all with applications in areas of economic importance. Application to cochlear implant speech processing will provide benefit for the hearing impaired. The project will provide students with training at an international level within Australia, thus helping ensure Australia maintains and extends its science and technology base into the future.Read moreRead less
Special Research Initiatives - Grant ID: SR0567109
Funder
Australian Research Council
Funding Amount
$69,438.00
Summary
Smart astronomy: using computational science to understand distant radio galaxies. Radio galaxies are among the largest galaxies in the universe with their copious radio emission powered by massive black holes. Australian radio telescopes are very effective at tracing these massive galaxies back in time so we can measure how black holes formed and developed. These measurements depend on reliable identification of the radio sources with our optical telescopes to make vital measurements of their d ....Smart astronomy: using computational science to understand distant radio galaxies. Radio galaxies are among the largest galaxies in the universe with their copious radio emission powered by massive black holes. Australian radio telescopes are very effective at tracing these massive galaxies back in time so we can measure how black holes formed and developed. These measurements depend on reliable identification of the radio sources with our optical telescopes to make vital measurements of their distances. Until now this identification process has been straightforward, but the next generation of studies will look so far back in time that the identification will become ambiguous. Our project will develop a software tool that applies techniques from computational science to overcome the ambiguity in this matching problem.Read moreRead less