Type extension and efficient AST manipulation

Oberon-2 is an object-oriented language with a class structure based on type extension. The runtime structure of Oberon-2 is described and the low-level mechanism for dynamic type checking explained. It is shown that the superior type-safety of the language, when used for programming styles based on heterogeneous, pointer-linked data structures, has an entirely negligible cost in runtime performance.

Semantic and context-based retrieval of digital cultural objects [Published in Chinese]

Cultural objects are increasingly generated and stored in digital form, yet effective methods for their indexing and retrieval still remain an important area of research. The main problem arises from the disconnection between the content-based indexing approach used by computer scientists and the description-based approach used by information scientists. There is also a lack of representational schemes that allow the alignment of the semantics and context with keywords and low-level features that can be automatically extracted from the content of these cultural objects. This paper presents an integrated approach to address these problems, taking advantage of both computer science and information science approaches. We firstly discuss the requirements from a number of perspectives: users, content providers, content managers and technical systems. We then present an overview of our system architecture and describe various techniques which underlie the major components of the system. These include: automatic object category detection; user-driven tagging; metadata transform and augmentation, and an expression language for digital cultural objects. In addition, we discuss our experience on testing and evaluating some existing collections, analyse the difficulties encountered and propose ways to address these problems.

Long-term exposure to gaseous air pollutants and cardio-respiratory mortality in Brisbane, Australia

Air pollution is ranked by the World Health Organisation as one of the top ten contributors to the global burden of disease and injury. Exposure to gaseous air pollutants, even at a low level, has been associated with cardiorespiratory diseases (Vedal, Brauer et al. 2003). Most recent epidemiological studies of air pollution have used time-series analyses to explore the relationship between daily mortality or morbidity and daily ambient air pollution concentrations based on the same day or previous days (Hajat, Armstrong et al. 2007). However, most of the previous studies have examined the association between air pollution and health outcomes using air pollution data from a single monitoring site or average values from a few monitoring sites to represent the whole population of the study area. In fact, for a metropolitan city, ambient air pollution levels may differ significantly among the different areas. There is increasing concern that the relationships between air pollution and mortality may vary with geographical area (Chen, Mengersen et al. 2007). Additionally, some studies have indicated that socio-economic status can act as a confounder when investigating the relation between geographical location and health (Scoggins, Kjellstrom et al. 2004). This study examined the spatial variation in the relationship between long-term exposure to gaseous air pollutants (including nitrogen dioxide (NO2), ozone (O3) and sulphur dioxide (SO2)), and cardiorespiratory mortality in Brisbane, Australia, during the period 1996 - 2004.

An ecological assessment of secondary poisoning risk in the Australian sugarcane industry

Rodenticide use in agriculture can lead to the secondary poisoning of avian predators. Currently the Australian sugarcane industry has two rodenticides, Racumin® and Rattoff®, available for in-crop use but, like many agricultural industries, it lacks an ecologically-based method of determining the potential secondary poisoning risk the use of these rodenticides poses to avian predators. The material presented in this thesis addresses this by: a. determining where predator/prey interactions take place in sugar producing districts; b. quantifying the amount of rodenticide available to avian predators and the probability of encounter; and c. developing a stochastic model that allows secondary poisoning risk under various rodenticide application scenarios to be investigated. Results demonstrate that predator/prey interactions are highly constrained by environmental structure. Rodents used crops that provided high levels of canopy cover and therefore predator protection and poorly utilised open canopy areas. In contrast, raptors over-utilised areas with low canopy cover and low rodent densities, but which provided high accessibility to prey. Given this pattern of habitat use, and that industry baiting protocols preclude rodenticide application in open canopy crops, these results indicate that secondary poisoning can only occur if poisoned rodents leave closed canopy crops and become available for predation in open canopy areas. Results further demonstrate that after in-crop rodenticide application, only a small proportion of rodents available in open areas are poisoned and that these rodents carry low levels of toxicant. Coupled with the low level of rodenticide use in the sugar industry, the high toxic threshold raptors have to these toxicants and the low probability of encountering poisoned rodents, results indicate that the risk of secondary poisoning events occurring is minimal. A stochastic model was developed to investigate the effect of manipulating factors that might influence secondary poisoning hazard in a sugarcane agro-ecosystem. These simulations further suggest that in all but extreme scenarios, the risk of secondary poisoning is also minimal. Collectively, these studies demonstrate that secondary poisoning of avian predators associated with the use of the currently available rodenticides in Australian sugar producing districts is minimal. Further, the ecologically-based method of assessing secondary poisoning risk developed in this thesis has broader applications in other agricultural systems where rodenticide use may pose risks to avian predators.

Image synthesis based on a model of human vision

Modern computer graphics systems are able to construct renderings of such high quality that viewers are deceived into regarding the images as coming from a photographic source. Large amounts of computing resources are expended in this rendering process, using complex mathematical models of lighting and shading. However, psychophysical experiments have revealed that viewers only regard certain informative regions within a presented image. Furthermore, it has been shown that these visually important regions contain low-level visual feature differences that attract the attention of the viewer. This thesis will present a new approach to image synthesis that exploits these experimental findings by modulating the spatial quality of image regions by their visual importance. Efficiency gains are therefore reaped, without sacrificing much of the perceived quality of the image. Two tasks must be undertaken to achieve this goal. Firstly, the design of an appropriate region-based model of visual importance, and secondly, the modification of progressive rendering techniques to effect an importance-based rendering approach. A rule-based fuzzy logic model is presented that computes, using spatial feature differences, the relative visual importance of regions in an image. This model improves upon previous work by incorporating threshold effects induced by global feature difference distributions and by using texture concentration measures. A modified approach to progressive ray-tracing is also presented. This new approach uses the visual importance model to guide the progressive refinement of an image. In addition, this concept of visual importance has been incorporated into supersampling, texture mapping and computer animation techniques. Experimental results are presented, illustrating the efficiency gains reaped from using this method of progressive rendering. This visual importance-based rendering approach is expected to have applications in the entertainment industry, where image fidelity may be sacrificed for efficiency purposes, as long as the overall visual impression of the scene is maintained. Different aspects of the approach should find many other applications in image compression, image retrieval, progressive data transmission and active robotic vision.

Idealistic quantum psychopathology : a way forward?

Quantum psychopathology holds the so called “quantum mind” hypothesis, which is controversial. In addition, this hypothesis focuses attention onto quantum processes in the brain, and how this may relate to psychopathological issues. This is very “low level”. As a consequence, it is challenging to form bridges to “higher level” problems related to psychopathology. By adopting the stance used in the quantum interaction community or researchers, this reply puts forward the idea that an idealistic approach may circumvent the controversy and opens the way for addressing challenges at higher levels of psychopathology.

Modelling the process of learning in economics

Student learning research literature has shown that students' learning approaches are influenced by the learning context (Evans, Kirby, & Fabrigar, 2003). Of the many contextual factors, assessment has been found to have the most important influence on the way students go about learning. For example, assessment that is perceived to required a low level of cognitive abilities will more likely elicit a learning approach that concentrate on reproductive learning activities. Moreover, assessment demand will also interact with learning approach to determine academic performance. In this paper an assessment specific model of learning comprising presage, process and product variables (Biggs, 2001) was proposed and tested against data obtained from a sample of introductory economics students (n=434). The model developed was used to empirically investigate the influence of learning inputs and learning approaches on academic performances across assessment types (essay assignment, multiple choice question exam and exam essay). By including learning approaches in the learning model, the mechanism through which learning inputs determine academic performance was examined. Methodological limitations of the study will also be discussed.

A Helicopter named Dolly : behavioural cloning for autonomous helicopter control

This paper considers the pros and cons of using Behavioural cloning for the development of low-level helicopter automation modules. Over the course of this project several Behavioural cloning approaches have been investigated. The results of the most effective Behavioural cloning approach are then compared to PID modules designed for the same aircraft. The comparison takes into consideration development time, reliability, and control performance. It has been found that Behavioural cloning techniques employing local approximators and a wide state-space coverage during training can produce stabilising control modules in less time than tuning PID controllers. However, performance and reliabity deficits have been found to exist with the Behavioural Cloning, attributable largely to the time variant nature of the dynamics due to the operating environment, and the pilot actions being poor for teaching. The final conclusion drawn here is that tuning PID modules remains superior to behavioural cloning for low-level helicopter automation.

Do warnings deter rather than produce more sophisticated malingering?

Following Youngjohn, Lees-Haley, and Binder's (1999) comment on Johnson and Lesniak-Karpiak's (1997) study that warnings lead to more subtle malingering, researchers have sought to better understand warning effects. However, such studies have been largely atheoretical and may have confounded warning and coaching. This study examined the effect on malingering of a warning that was based on criminological-sociological concepts derived from the rational choice model of deterrence theory. A total of 78 participants were randomly assigned to a control group, an unwarned simulator group, or one of two warned simulator groups. The warning groups comprised low- and high-level conditions depending on warning intensity. Simulator participants received no coaching about how to fake tests. Outcome variables were scores derived from the Test of Memory Malingering and Wechsler Memory Scale-III. When the rate of malingering was compared across the four groups, a high-level warning effect was found such that warned participants were significantly less likely to exaggerate than unwarned simulators. In an exploratory follow-up analysis, the warned groups were divided into those who reported malingering and those who did not report malingering, and the performance of these groups was compared to that of unwarned simulators and controls. Using this approach, results showed that participants who were deterred from malingering by warning performed no worse than controls. However, on a small number of tests, self-reported malingerers in the low-level warning group appeared less impaired than unwarned simulators. This pattern was not observed in the high-level warning condition. Although cautious interpretation of findings is necessitated by the exploratory nature of some analyses, overall results suggest that using a carefully designed warning may be useful for reducing the rate of malingering. The combination of some noteworthy effect sizes, despite low power and the small size of some groups, suggests that further investigation of the effects of warnings needs to continue to determine their effect more fully.

Multi-robot control in highly dynamic, competitive environments

The control and coordination of multiple mobile robots is a challenging task; particularly in environments with multiple, rapidly moving obstacles and agents. This paper describes a robust approach to multi-robot control, where robustness is gained from competency at every layer of robot control. The layers are: (i) a central coordination system (MAPS), (ii) an action system (AES), (iii) a navigation module, and (iv) a low level dynamic motion control system. The multi-robot coordination system assigns each robot a role and a sub-goal. Each robots action execution system then assumes the assigned role and attempts to achieve the specified sub-goal. The robots navigation system directs the robot to specific goal locations while ensuring that the robot avoids any obstacles. The motion system maps the heading and speed information from the navigation system to force-constrained motion. This multi-robot system has been extensively tested and applied in the robot soccer domain using both centralized and distributed coordination.

Pandemic (H1N1) 2009 Influenza Outbreak in Australia: Impact on Emergency Departments.

Executive summary Objective: The aims of this study were to identify the impact of Pandemic (H1N1) 2009 Influenza on Australian Emergency Departments (EDs) and their staff, and to inform planning, preparedness, and response management arrangements for future pandemics, as well as managing infectious patients presenting to EDs in everyday practice. Methods This study involved three elements: 1. The first element of the study was an examination of published material including published statistics. Standard literature research methods were used to identify relevant published articles. In addition, data about ED demand was obtained from Australian Government Department of Health and Ageing (DoHA) publications, with several state health departments providing more detailed data. 2. The second element of the study was a survey of Directors of Emergency Medicine identified with the assistance of the Australasian College for Emergency Medicine (ACEM). This survey retrieved data about demand for ED services and elicited qualitative comments on the impact of the pandemic on ED management. 3. The third element of the study was a survey of ED staff. A questionnaire was emailed to members of three professional colleges—the ACEM; the Australian College of Emergency Nursing (ACEN); and the College of Emergency Nursing Australasia (CENA). The overall response rate for the survey was 18.4%, with 618 usable responses from 3355 distributed questionnaires. Topics covered by the survey included ED conditions during the (H1N1) 2009 influenza pandemic; information received about Pandemic (H1N1) 2009 Influenza; pandemic plans; the impact of the pandemic on ED staff with respect to stress; illness prevention measures; support received from others in work role; staff and others’ illness during the pandemic; other factors causing ED staff to miss work during the pandemic; and vaccination against Pandemic (H1N1) 2009 Influenza. Both qualitative and quantitative data were collected and analysed. Results: The results obtained from Directors of Emergency Medicine quantifying the impact of the pandemic were too limited for interpretation. Data sourced from health departments and published sources demonstrated an increase in influenza-like illness (ILI) presentations of between one and a half and three times the normal level of presentations of ILIs. Directors of Emergency Medicine reported a reasonable level of preparation for the pandemic, with most reporting the use of pandemic plans that translated into relatively effective operational infection control responses. Directors reported a highly significant impact on EDs and their staff from the pandemic. Growth in demand and related ED congestion were highly significant factors causing distress within the departments. Most (64%) respondents established a ‘flu clinic’ either as part of Pandemic (H1N1) 2009 Influenza Outbreak in Australia: Impact on Emergency Departments. the ED operations or external to it. They did not note a significantly higher rate of sick leave than usual. Responses relating to the impact on staff were proportional to the size of the colleges. Most respondents felt strongly that Pandemic (H1N1) 2009 Influenza had a significant impact on demand in their ED, with most patients having low levels of clinical urgency. Most respondents felt that the pandemic had a negative impact on the care of other patients, and 94% revealed some increase in stress due to lack of space for patients, increased demand, and filling staff deficits. Levels of concern about themselves or their family members contracting the illness were less significant than expected. Nurses displayed significantly higher levels of stress overall, particularly in relation to skill-mix requirements, lack of supplies and equipment, and patient and patients’ family aggression. More than one-third of respondents became ill with an ILI. Whilst respondents themselves reported taking low levels of sick leave, respondents cited difficulties with replacing absent staff. Ranked from highest to lowest, respondents gained useful support from ED colleagues, ED administration, their hospital occupational health department, hospital administration, professional colleges, state health department, and their unions. Respondents were generally positive about the information they received overall; however, the volume of information was considered excessive and sometimes inconsistent. The media was criticised as scaremongering and sensationalist and as being the cause of many unnecessary presentations to EDs. Of concern to the investigators was that a large proportion (43%) of respondents did not know whether a pandemic plan existed for their department or hospital. A small number of staff reported being redeployed from their usual workplace for personal risk factors or operational reasons. As at the time of survey (29 October –18 December 2009), 26% of ED staff reported being vaccinated against Pandemic (H1N1) 2009 Influenza. Of those not vaccinated, half indicated they would ‘definitely’ or ‘probably’ not get vaccinated, with the main reasons being the vaccine was ‘rushed into production’, ‘not properly tested’, ‘came out too late’, or not needed due to prior infection or exposure, or due to the mildness of the disease. Conclusion: Pandemic (H1N1) 2009 Influenza had a significant impact on Australian Emergency Departments. The pandemic exposed problems in existing plans, particularly a lack of guidelines, general information overload, and confusion due to the lack of a single authoritative information source. Of concern was the high proportion of respondents who did not know if their hospital or department had a pandemic plan. Nationally, the pandemic communication strategy needs a detailed review, with more engagement with media networks to encourage responsible and consistent reporting. Also of concern was the low level of immunisation, and the low level of intention to accept vaccination. This is a problem seen in many previous studies relating to seasonal influenza and health care workers. The design of EDs needs to be addressed to better manage infectious patients. Significant workforce issues were confronted in this pandemic, including maintaining appropriate staffing levels; staff exposure to illness; access to, and appropriate use of, personal protective equipment (PPE); and the difficulties associated with working in PPE for prolonged periods. An administrative issue of note was the reporting requirement, which created considerable additional stress for staff within EDs. Peer and local support strategies helped ensure staff felt their needs were provided for, creating resilience, dependability, and stability in the ED workforce. Policies regarding the establishment of flu clinics need to be reviewed. The ability to create surge capacity within EDs by considering staffing, equipment, physical space, and stores is of primary importance for future pandemics.

Secure tracking for critical applications : communications, GPS and future Galileo services

Tracking/remote monitoring systems using GNSS are a proven method to enhance the safety and security of personnel and vehicles carrying precious or hazardous cargo. While GNSS tracking appears to mitigate some of these threats, if not adequately secured, it can be a double-edged sword allowing adversaries to obtain sensitive shipment and vehicle position data to better coordinate their attacks, and to provide a false sense of security to monitoring centers. Tracking systems must be designed with the ability to perform route-compliance and thwart attacks ranging from low-level attacks such as the cutting of antenna cables to medium and high-level attacks involving radio jamming and signal / data-level simulation, especially where the goods transported have a potentially high value to terrorists. This paper discusses the use of GNSS in critical tracking applications, addressing the mitigation of GNSS security issues, augmentation systems and communication systems in order to provide highly robust and survivable tracking systems.

Hybrid robot control and SLAM for persistent navigation and mapping

For a mobile robot to operate autonomously in real-world environments, it must have an effective control system and a navigation system capable of providing robust localization, path planning and path execution. In this paper we describe the work investigating synergies between mapping and control systems. We have integrated development of a control system for navigating mobile robots and a robot SLAM system. The control system is hybrid in nature and tightly coupled with the SLAM system; it uses a combination of high and low level deliberative and reactive control processes to perform obstacle avoidance, exploration, global navigation and recharging, and draws upon the map learning and localization capabilities of the SLAM system. The effectiveness of this hybrid, multi-level approach was evaluated in the context of a delivery robot scenario. Over a period of two weeks the robot performed 1143 delivery tasks to 11 different locations with only one delivery failure (from which it recovered), travelled a total distance of more than 40km, and recharged autonomously a total of 23 times. In this paper we describe the combined control and SLAM system and discuss insights gained from its successful application in a real-world context.

Security metrics for object-oriented designs

Several studies have developed metrics for software quality attributes of object-oriented designs such as reusability and functionality. However, metrics which measure the quality attribute of information security have received little attention. Moreover, existing security metrics measure either the system from a high level (i.e. the whole system’s level) or from a low level (i.e. the program code’s level). These approaches make it hard and expensive to discover and fix vulnerabilities caused by software design errors. In this work, we focus on the design of an object-oriented application and define a number of information security metrics derivable from a program’s design artifacts. These metrics allow software designers to discover and fix security vulnerabilities at an early stage, and help compare the potential security of various alternative designs. In particular, we present security metrics based on composition, coupling, extensibility, inheritance, and the design size of a given object-oriented, multi-class program from the point of view of potential information flow.

Ontology-based image annotation

With regard to the long-standing problem of the semantic gap between low-level image features and high-level human knowledge, the image retrieval community has recently shifted its emphasis from low-level features analysis to high-level image semantics extrac- tion. User studies reveal that users tend to seek information using high-level semantics. Therefore, image semantics extraction is of great importance to content-based image retrieval because it allows the users to freely express what images they want. Semantic content annotation is the basis for semantic content retrieval. The aim of image anno- tation is to automatically obtain keywords that can be used to represent the content of images. The major research challenges in image semantic annotation are: what is the basic unit of semantic representation? how can the semantic unit be linked to high-level image knowledge? how can the contextual information be stored and utilized for image annotation? In this thesis, the Semantic Web technology (i.e. ontology) is introduced to the image semantic annotation problem. Semantic Web, the next generation web, aims at mak- ing the content of whatever type of media not only understandable to humans but also to machines. Due to the large amounts of multimedia data prevalent on the Web, re- searchers and industries are beginning to pay more attention to the Multimedia Semantic Web. The Semantic Web technology provides a new opportunity for multimedia-based applications, but the research in this area is still in its infancy. Whether ontology can be used to improve image annotation and how to best use ontology in semantic repre- sentation and extraction is still a worth-while investigation. This thesis deals with the problem of image semantic annotation using ontology and machine learning techniques in four phases as below. 1) Salient object extraction. A salient object servers as the basic unit in image semantic extraction as it captures the common visual property of the objects. Image segmen- tation is often used as the �rst step for detecting salient objects, but most segmenta- tion algorithms often fail to generate meaningful regions due to over-segmentation and under-segmentation. We develop a new salient object detection algorithm by combining multiple homogeneity criteria in a region merging framework. 2) Ontology construction. Since real-world objects tend to exist in a context within their environment, contextual information has been increasingly used for improving object recognition. In the ontology construction phase, visual-contextual ontologies are built from a large set of fully segmented and annotated images. The ontologies are composed of several types of concepts (i.e. mid-level and high-level concepts), and domain contextual knowledge. The visual-contextual ontologies stand as a user-friendly interface between low-level features and high-level concepts. 3) Image objects annotation. In this phase, each object is labelled with a mid-level concept in ontologies. First, a set of candidate labels are obtained by training Support Vectors Machines with features extracted from salient objects. After that, contextual knowledge contained in ontologies is used to obtain the �nal labels by removing the ambiguity concepts. 4) Scene semantic annotation. The scene semantic extraction phase is to get the scene type by using both mid-level concepts and domain contextual knowledge in ontologies. Domain contextual knowledge is used to create scene con�guration that describes which objects co-exist with which scene type more frequently. The scene con�guration is represented in a probabilistic graph model, and probabilistic inference is employed to calculate the scene type given an annotated image. To evaluate the proposed methods, a series of experiments have been conducted in a large set of fully annotated outdoor scene images. These include a subset of the Corel database, a subset of the LabelMe dataset, the evaluation dataset of localized semantics in images, the spatial context evaluation dataset, and the segmented and annotated IAPR TC-12 benchmark.