Recognising facial expressions with noisy data

Techniques to improve the automated analysis of natural and spontaneous facial expressions have been developed. The outcome of the research has applications in several fields including national security (eg: expression invariant face recognition); education (eg: affect aware interfaces); mental and physical health (eg: depression and pain recognition).

Understanding the impact of legislative changes on child car seating and restraint practices in regional Queensland

Road crashes contribute to a significant amount of child mortality and morbidity in Australia. In fact, passenger injuries contribute to the majority of child crash road trauma. A number of factors contribute to child injury and death in motor vehicles, including inappropriate seating position, inappropriate choice of restraint, and incorrect installation and use of child restraints. Prior to March 2010, child restraint legislation in Queensland only required children twelve months and younger to be seated in a properly adjusted and fastened child restraint. This legislation left older infants and young children potentially suboptimally protected. From March 2010, new legislation specified seating position and type of child restraint required, depending on the age of the child. This research was underpinned by the Health Belief Model (HBM), which explores health related behaviour, behaviour change, environmental factors influencing behaviour change (including legislative changes) and is flexible enough to be used in relation to parents' health practices for their children, rather than parent health directly. This thesis investigates the extent to which the changes to child restraint legislation have led parents in regional areas of Queensland to use appropriate restraint practices for their children and determines the extent to which the constructs of the HBM, parental perceptions, barriers and environmental factors contribute to the appropriateness of child seating and restraint use. Study One included three sets of observations taken in two regional cities of Queensland prior to the legislative amendment, during an educative period of six months, and after the enactment of the legislation. Each child's seating position and restraint type were recorded. Results showed that the proportion of children observed occupying the front seat decreased by 15.6 per cent with the announcement the legislation. There was no decrease in front seat use at the enactment of the legislation. The proportion of children observed using dedicated child restraints increased by 8.8 per cent with the announcement of the legislation when there was one child in the vehicle. Further, there was a 10.1 per cent increase in the proportion of children observed using a seat belt that fit with the announcement when there was one child in the vehicle and with the enactment of the legislation regardless of the number of children in the vehicle (21.8 per cent for one child, 39.7 per cent for two children and 40.2 per cent for three or more children). Study Two comprised initial intercept interviews, later followed up by telephone, with parents with children aged eight years and younger at the announcement and telephone interviews at the enactment of the legislation in one regional city in Queensland. Parents reported their child restraint practices, and opinions, knowledge and understanding of the requirements of the new legislation. Parent responses were analysed in terms of the constructs in the HBM. When asked which seating position their child 'usually' used, parents reported child front seat use was nil (0.0 per cent) and did not change with the enactment of the legislative amendment. However, when parents were asked whether they allowed children to use the front seat at some point within the six months prior to the interview, reported child front seat use was 7 (5.4 per cent) children at T2 and 10 (9.6 per cent) at T3. Reported use of age-appropriate child restraints did not increase with the enactment of the legislation (p = 0.77, ns). Parents reported restraint practices were classed as either appropriate or inappropriate. Parents who reported appropriate restraint practices were those whose children were sitting in optimal restraints and seating positions for their age according to the requirements of the legislation. Parents who reported inappropriate restraint practices were those who had one or more children who were suboptimally restrained or seated for their age according to the requirements of the legislation. Neither parents' perceptions about their susceptibility of being in a crash nor the likelihood of severity of child injury if involved in a crash yielded significant differences in the appropriateness of reported parent restraint practices over time with the enactment of the legislation. A trend in the data suggested parents perceived a benefit to using appropriate restraint practices was to avoid fines and demerit points. Over 75 per cent of parents who agreed that child restraints provide better protection for children than an adult seat belt reported appropriately seating and restraining their children (2 (1) = 8.093, p<.05). The self-efficacy measure regarding parents' confidence in installing a child restraint showed a significant association with appropriate parental restraint practices (2 (1) = 7.036, p<.05). Results suggested that some parents may have misinterpreted the announcement of the legislative amendment as the announcement of the enforcement of the legislation instead. Some parents who correctly reported details of the legislation did not report appropriate child restraint practices. This finding shows that parents' knowledge of the legislative amendment does not necessarily have an impact on their behaviour to appropriately seat and restrain children. The results of these studies have important implications for road safety and the prevention of road-related injury and death to children in Queensland. Firstly, parents reported feeling unsure of how to install restraints, which suggests that there may be children travelling in restraints that have not been installed correctly, putting them at risk. Interventions to alert and encourage parents to seek advice when unsure about the correct installation of child restraints could be considered. Secondly, some parents in this study although they were using the most appropriate restraint for their children, reported using a type that was not the most appropriate restraint for the child's age according to the legislation. This suggests that intervention may be effective in helping parents make a more accurate choice of the most appropriate type of restraint to use with children, especially as the child ages and child restraint requirements change. Further research could be conducted to ascertain the most effective methods of informing and motivating parents to use the most appropriate restraints and seating positions for their children, as these results show a concerning disparity between reported restraint practices and those that were observed.

A basis for instrusion detection in distributed systems using kernel-level data tainting

This project was a step forward in developing intrusion detection systems in distributed environments such as web services. It investigates a new approach of detection based on so-called "taint-marking" techniques and introduces a theoretical framework along with its implementation in the Linux kernel.

Assessment of police subjective workload and preference for using a voice-based interface during simulated driving

Police in-vehicle systems include a visual output mobile data terminal (MDT) with manual input via touch screen and keyboard. This study investigated the potential for voice-based input and output modalities for reducing subjective workload of police officers while driving. Nineteen experienced drivers of police vehicles (one female) from New South Wales (NSW) Police completed four simulated urban drives. Three drives included a concurrent secondary task: an imitation licence number search using an emulated MDT. Three different interface output-input modalities were examined: Visual-Manual, Visual-Voice, and Audio-Voice. Following each drive, participants rated their subjective workload using the NASA - Raw Task Load Index and completed questions on acceptability. A questionnaire on interface preferences was completed by participants at the end of their session. Engaging in secondary tasks while driving significantly increased subjective workload. The Visual-Manual interface resulted in higher time demand than either of the voice-based interfaces and greater physical demand than the Audio-Voice interface. The Visual-Voice and Audio-Voice interfaces were rated easier to use and more useful than the Visual-Manual interface, although not significantly different from each other. Findings largely echoed those deriving from the analysis of the objective driving performance data. It is acknowledged that under standard procedures, officers should not drive while performing tasks concurrently with certain invehicle policing systems; however, in practice this sometimes occurs. Taking action now to develop voice-based technology for police in-vehicle systems has potential to realise visions for potentially safer and more efficient vehicle-based police work.

Investigating the factors influencing cyclist awareness and behaviour : an on-road study of cyclist situation awareness

Situation awareness, ones understanding of ‘what is going on’, is a critical commodity for road users. Although the concept has received much attention in the driving context, situation awareness in vulnerable road users, such as cyclists, remains unexplored. This paper presents the findings from an exploratory on-road study of cyclist situation awareness, the aim of which was to explore how cyclists develop situation awareness, what their situation awareness comprises, and what the causes of degraded cyclist situation awareness may be. Twenty participants cycled a pre-defined urban on-road study route. A range of data were collected, including verbal protocols, forward scene video and rear video, and a network analysis procedure was used to describe and assess cyclist situation awareness. The analysis produced a number of key findings regarding cyclist situation awareness, including the potential for cyclists’ awareness of other road users to be degraded due to additional situation awareness and decision making requirements that are placed on them in certain road situations. Strategies for improving cyclists’ situation awareness are discussed.

On-road driving studies to understand why drivers behave as they do at regional rail level crossings

Improving safety at rail level crossings is an important part of both road and rail safety strategies. While low in number, crashes between vehicles and trains at level crossings are catastrophic events typically involving multiple fatalities and serious injuries. Advances in driving assessment methods, such as the provision of on-road instrumented test vehicles with eye and head tracking, provide researchers with the opportunity to further understand driver behaviour at such crossings in ways not previously possible. This paper describes a study conducted to further understand the factors that shape driver behaviour at rail level crossings using instrumented vehicles. Twenty-two participants drove an On-Road Test Vehicle (ORTeV) on a predefined route in regional Victoria with a mix of both active (flashing lights with/without boom barriers) and passively controlled (stop, give way) crossings. Data collected included driving performance data, head checks, and interview data to capture driver strategies. The data from an integrated suite of methods demonstrated clearly how behaviour differs at active and passive level crossings, particularly for inexperienced drivers. For example, the head check data clearly show the reliance and expectancies of inexperienced drivers for active warnings even when approaching passively controlled crossings. These studies provide very novel and unique insights into how level crossing design and warnings shape driver behaviour.

Correcting radiation survey data to account for increased leakage during intensity modulated radiotherapy treatments

Purpose Intensity modulated radiotherapy (IMRT) treatments require more beam-on time and produce more linac head leakage to deliver similar doses to conventional, unmodulated, radiotherapy treatments. It is necessary to take this increased leakage into account when evaluating the results of radiation surveys around bunkers that are, or will be, used for IMRT. The recommended procedure of 15 applying a monitor-unit based workload correction factor to secondary barrier survey measurements, to account for this increased leakage when evaluating radiation survey measurements around IMRT bunkers, can lead to potentially-costly over estimation of the required barrier thickness. This study aims to provide initial guidance on the validity of reducing the value of the correction factor when applied to different radiation barriers (primary barriers, doors, maze walls and other walls) by 20 evaluating three different bunker designs. Methods Radiation survey measurements of primary, scattered and leakage radiation were obtained at each of five survey points around each of three different radiotherapy bunkers and the contribution of leakage to the total measured radiation dose at each point was evaluated. Measurements at each survey point were made with the linac gantry set to 12 equidistant positions from 0 to 330o, to 25 assess the effects of radiation beam direction on the results. Results For all three bunker designs, less than 0.5% of dose measured at and alongside the primary barriers, less than 25% of the dose measured outside the bunker doors and up to 100% of the dose measured outside other secondary barriers was found to be caused by linac head leakage. Conclusions Results of this study suggest that IMRT workload corrections are unnecessary, for 30 survey measurements made at and alongside primary barriers. Use of reduced IMRT workload correction factors is recommended when evaluating survey measurements around a bunker door, provided that a subset of the measurements used in this study are repeated for the bunker in question. Reduction of the correction factor for other secondary barrier survey measurements is not recommended unless the contribution from leakage is separetely evaluated.

Building an augmented map for road risk assessment

Recent road safety statistics show that the decades-long fatalities decreasing trend is stopping and stagnating. Statistics further show that crashes are mostly driven by human error, compared to other factors such as environmental conditions and mechanical defects. Within human error, the dominant error source is perceptive errors, which represent about 50% of the total. The next two sources are interpretation and evaluation, which accounts together with perception for more than 75% of human error related crashes. Those statistics show that allowing drivers to perceive and understand their environment better, or supplement them when they are clearly at fault, is a solution to a good assessment of road risk, and, as a consequence, further decreasing fatalities. To answer this problem, currently deployed driving assistance systems combine more and more information from diverse sources (sensors) to enhance the driver's perception of their environment. However, because of inherent limitations in range and field of view, these systems' perception of their environment remains largely limited to a small interest zone around a single vehicle. Such limitations can be overcomed by increasing the interest zone through a cooperative process. Cooperative Systems (CS), a specific subset of Intelligent Transportation Systems (ITS), aim at compensating for local systems' limitations by associating embedded information technology and intervehicular communication technology (IVC). With CS, information sources are not limited to a single vehicle anymore. From this distribution arises the concept of extended or augmented perception. Augmented perception allows extending an actor's perceptive horizon beyond its "natural" limits not only by fusing information from multiple in-vehicle sensors but also information obtained from remote sensors. The end result of an augmented perception and data fusion chain is known as an augmented map. It is a repository where any relevant information about objects in the environment, and the environment itself, can be stored in a layered architecture. This thesis aims at demonstrating that augmented perception has better performance than noncooperative approaches, and that it can be used to successfully identify road risk. We found it was necessary to evaluate the performance of augmented perception, in order to obtain a better knowledge on their limitations. Indeed, while many promising results have already been obtained, the feasibility of building an augmented map from exchanged local perception information and, then, using this information beneficially for road users, has not been thoroughly assessed yet. The limitations of augmented perception, and underlying technologies, have not be thoroughly assessed yet. Most notably, many questions remain unanswered as to the IVC performance and their ability to deliver appropriate quality of service to support life-saving critical systems. This is especially true as the road environment is a complex, highly variable setting where many sources of imperfections and errors exist, not only limited to IVC. We provide at first a discussion on these limitations and a performance model built to incorporate them, created from empirical data collected on test tracks. Our results are more pessimistic than existing literature, suggesting IVC limitations have been underestimated. Then, we develop a new CS-applications simulation architecture. This architecture is used to obtain new results on the safety benefits of a cooperative safety application (EEBL), and then to support further study on augmented perception. At first, we confirm earlier results in terms of crashes numbers decrease, but raise doubts on benefits in terms of crashes' severity. In the next step, we implement an augmented perception architecture tasked with creating an augmented map. Our approach is aimed at providing a generalist architecture that can use many different types of sensors to create the map, and which is not limited to any specific application. The data association problem is tackled with an MHT approach based on the Belief Theory. Then, augmented and single-vehicle perceptions are compared in a reference driving scenario for risk assessment,taking into account the IVC limitations obtained earlier; we show their impact on the augmented map's performance. Our results show that augmented perception performs better than non-cooperative approaches, allowing to almost tripling the advance warning time before a crash. IVC limitations appear to have no significant effect on the previous performance, although this might be valid only for our specific scenario. Eventually, we propose a new approach using augmented perception to identify road risk through a surrogate: near-miss events. A CS-based approach is designed and validated to detect near-miss events, and then compared to a non-cooperative approach based on vehicles equiped with local sensors only. The cooperative approach shows a significant improvement in the number of events that can be detected, especially at the higher rates of system's deployment.

The consistency of crash involvement recall across time

Introduction Road safety researchers rely heavily on self-report data to explore the aetiology of crash risk. However, researchers consistently acknowledge a range of limitations associated with this methodological approach (e.g., self-report bias), which has been hypothesised to reduce the predictive efficacy of scales. Although well researched in other areas, one important factor often neglected in road safety studies is the fallibility of human memory. Given accurate recall is a key assumption in many studies, the validity and consistency of self-report data warrants investigation. The aim of the current study was to examine the consistency of self-report data of crash history and details of the most recent reported crash on two separate occasions. Materials & Method A repeated measures design was utilised to examine the self-reported crash involvement history of 214 general motorists over a two month period. Results A number of interesting discrepancies were noted in relation to number of lifetime crashes reported by the participants and the descriptions of their most recent crash across the two occasions. Of the 214 participants who reported having been involved in a crash, 35 (22.3%) reported a lower number of lifetime crashes as Time 2, than at Time 1. Of the 88 drivers who reported no change in number of lifetime crashes, 10 (11.4%) described a different most recent crash. Additionally, of the 34 reporting an increase in the number of lifetime crashes, 29 (85.3%) of these described the same crash on both occasions. Assessed as a whole, at least 47.1% of participants made a confirmed mistake at Time 1 or Time 2. Conclusions These results raise some doubt in regard to the accuracy of memory recall across time. Given that self-reported crash involvement is the predominant dependent variable used in the majority of road safety research, this issue warrants further investigation. Replication of the study with a larger sample size that includes multiple recall periods would enhance understanding into the significance of this issue for road safety methodology.

The efficiency of corpus-based distributional models for literature-based discovery on large data sets

This paper evaluates the efficiency of a number of popular corpus-based distributional models in performing discovery on very large document sets, including online collections. Literature-based discovery is the process of identifying previously unknown connections from text, often published literature, that could lead to the development of new techniques or technologies. Literature-based discovery has attracted growing research interest ever since Swanson's serendipitous discovery of the therapeutic effects of fish oil on Raynaud's disease in 1986. The successful application of distributional models in automating the identification of indirect associations underpinning literature-based discovery has been heavily demonstrated in the medical domain. However, we wish to investigate the computational complexity of distributional models for literature-based discovery on much larger document collections, as they may provide computationally tractable solutions to tasks including, predicting future disruptive innovations. In this paper we perform a computational complexity analysis on four successful corpus-based distributional models to evaluate their fit for such tasks. Our results indicate that corpus-based distributional models that store their representations in fixed dimensions provide superior efficiency on literature-based discovery tasks.

The wrong tool for the job? The predictive powers of the DBQ in a sample of Queensland motorists

Introduction This study reports on the application of the Manchester Driver Behaviour Questionnaire (DBQ) to examine the self-reported driving behaviours (e.g., speeding, errors & aggressive manoeuvres) and predict crash involvement among a sample of general Queensland motorists. Material and Methods Surveys were completed by 249 general motorists on-line or via a pen-and-paper format. Results A factor analysis revealed a three factor solution for the DBQ which was consistent with previous Australian-based research. It accounted for 40.5% of the total variance, although some cross-loadings were observed on nine of the twenty items. The internal reliability of the DBQ was satisfactory. However, multivariate analysis using the DBQ revealed little predictive ability of the tool to predict crash involvement or demerit point loss e.g. violation notices. Rather, exposure to the road was found to be predictive of crashes, although speeding did make a small contribution to those who recently received a violation notice. Conclusions Taken together, the findings contribute to a growing body of research that raises questions about the predictive ability of the most widely used driving assessment tool globally. Ongoing research (which also includes official crash and offence outcomes) is required to better understand the actual contribution that the DBQ can make to understanding and improving road safety. Future research should also aim to confirm whether this lack of predictive efficacy originates from broader issues inherent within self-report data (e.g., memory recall problems) or issues underpinning the conceptualisation of the scale.

Characteristics of police-reported road traffic crashes in Ethiopia over a six year period

Ethiopia has one of Africa’s fastest growing non-oil producing economies and an increasing level of motorisation (AfDB, OECD, UNDP, & UNECA, 2012). This rapidly increasing mobility has created some unique road safety concerns; however there is scant published information and related commentary (United Nations Economic Commission for Africa, 2009). The objective of this paper is to quantify police-reported traffic crashes in Ethiopia and characterise the existing state of road safety. Six years (July 2005 - June 2011) of police-reported crash data were analysed, consisting of 12,140 fatal and 29,454 injury crashes on the country’s road network. The 12,140 fatal crashes involved 1,070 drivers, 5,702 passengers, and 7,770 pedestrians, totalling 14,542 fatalities, an average of 1.2 road user fatalities per crash. An important and glaring trend that emerges is that more than half of the fatalities in Ethiopia involve pedestrians. The majority of the crashes occur during daytime hours, involve males, and involve persons in the 18-50 age group—Ethiopia’s active workforce. Crashes frequently occur in mid blocks or roadways. The predominant collision between motor vehicles and pedestrians was a rollover on a road tangent section. Failing to observe the priority of pedestrians and speeding were the major causes of crashes attributed by police. Trucks and minibus taxis were involved in the majority of crashes, while automobiles (small vehicles) were less involved in crashes relative to other vehicle types, partially because small vehicles tend to be driven fewer kilometres per annum. These data illustrate and justify a high priority to identify and implement effective programs, policies, and countermeasures focused on reducing pedestrian crashes.

Injury severity of pedestrian crashes in Singapore

Singapore is a highly urbanized city-state country where walking is an important mode of travel. Pedestrians form about 25% of road fatalities every year, making them one of the most vulnerable road user groups in Singapore. Engineering measures like provision of overhead pedestrian crossings and raised zebra crossings tend to address pedestrian safety in general, but there may be occasions where pedestrians are particularly vulnerable so that targeted interventions are more appropriate. The objective of this study is to identify factors and situations that affect the injury severity of pedestrians involved in traffic crashes. Six years of crash data from 2003 to 2008 containing around four thousands pedestrian crashes at roadway segments were analyzed. Injury severity of pedestrians—recorded as slight injury, major injury and fatal—were modeled as a function of roadway characteristics, traffic features, environmental factors and pedestrian demographics by an ordered probit model. Results suggest that the injury severity of pedestrians involved in crashes during night time is higher indicating that pedestrian visibility during night is a key issue in pedestrian safety. The likelihood of fatal or serious injuries is higher for crashes on roads with high speed limit, center and median lane of multi-lane roads, school zones, roads with two-way divided traffic type, and when pedestrians cross the roads. Elderly pedestrians appear to be involved in fatal and serious injury crashes more when they attempt to cross the road without using nearby crossing facilities. Specific countermeasures are recommended based on the findings of this study.

Traffic density estimation of signalised arterials with stop line detector and probe data

This research aims to develop a reliable density estimation method for signalised arterials based on cumulative counts from upstream and downstream detectors. In order to overcome counting errors associated with urban arterials with mid-link sinks and sources, CUmulative plots and Probe Integration for Travel timE estimation (CUPRITE) is employed for density estimation. The method, by utilizing probe vehicles’ samples, reduces or cancels the counting inconsistencies when vehicles’ conservation is not satisfied within a section. The method is tested in a controlled environment, and the authors demonstrate the effectiveness of CUPRITE for density estimation in a signalised section, and discuss issues associated with the method.

Enabling open access to and re-use of publicly funded research data in Malaysian public universities : a legal and policy analysis

Numerous statements and declarations have been made over recent decades in support of open access to research data. The growing recognition of the importance of open access to research data has been accompanied by calls on public research funding agencies and universities to facilitate better access to publicly funded research data so that it can be re-used and redistributed as public goods. International and inter-governmental bodies such as the ICSU/CODATA, the OECD and the European Union are strong supporters of open access to and re-use of publicly funded research data. This thesis focuses on the research data created by university researchers in Malaysian public universities whose research activities are funded by the Federal Government of Malaysia. Malaysia, like many countries, has not yet formulated a policy on open access to and re-use of publicly funded research data. Therefore, the aim of this thesis is to develop a policy to support the objective of enabling open access to and re-use of publicly funded research data in Malaysian public universities. Policy development is very important if the objective of enabling open access to and re-use of publicly funded research data is to be successfully achieved. In developing the policy, this thesis identifies a myriad of legal impediments arising from intellectual property rights, confidentiality, privacy and national security laws, novelty requirements in patent law and lack of a legal duty to ensure data quality. Legal impediments such as these have the effect of restricting, obstructing, hindering or slowing down the objective of enabling open access to and re-use of publicly funded research data. A key focus in the formulation of the policy was the need to resolve the various legal impediments that have been identified. This thesis analyses the existing policies and guidelines of Malaysian public universities to ascertain to what extent the legal impediments have been resolved. An international perspective is adopted by making a comparative analysis of the policies of public research funding agencies and universities in the United Kingdom, the United States and Australia to understand how they have dealt with the identified legal impediments. These countries have led the way in introducing policies which support open access to and re-use of publicly funded research data. As well as proposing a policy supporting open access to and re-use of publicly funded research data in Malaysian public universities, this thesis provides procedures for the implementation of the policy and guidelines for addressing the legal impediments to open access and re-use.