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.

In-vehicle avatars to elicit social response and change driving behaviour

Social and psychological theories have provided a plethora of evidence showing that the physical difficulty to express appropriate social interactions between drivers expresses itself in aggression, selfish driving and anti-social behaviour. Therefore there is a need to improve interactions between drivers and allow clearer collective decision making between them. Personal characteristics and the driving situations play strong roles in driver’s aggression. Our approach is centered around the driving situation as opposed to focusing on personality characteristics. It examines aggression and manipulates contextual variables such as driver’s eye contact exchanges. This paper presents a new unobtrusive in-vehicle system that aims at communicating drivers’ intentions, elicit social responses and increasing mutual awareness. It uses eye gaze as a social cue to affect collective decision making with the view to contribute to safe driving. The authors used a driving simulator to design a case control experiment in which eye gaze movements are conveyed with an avatar. Participants were asked to drive through different types of intersections. An avatar representing the head of the other driver was displayed and driver behaviour was analysed. Significant eye gaze pattern difference where observed when an avatar was displayed. Drivers cautiously refer to the avatar when information is required on the intention of others (e.g. when they do not have the right of way). The majority of participants reported the perception of “being looked at”. The number of glances and time spent gazing at the avatar did not indicate an unsafe distraction by standards of in-vehicle device ergonomic design. Avatars were visually consulted primarily in less demanding driving situations, which underlines their non-distractive nature.

An old language for a new landscape

Because aesthetics can have a profound effect upon the human relationship to the non-human environment the importance of aesthetics to ecologically sustainable designed landscapes has been acknowledged. However, in recognition that the physical forms of designed landscapes are an expression of the social values of the time, some design professionals have called for a new aesthetic ― one that reflects these current ecological concerns. To address this, some authors have suggested various theoretical design frameworks upon which such an aesthetic could be based. Within these frameworks there is an underlying theme that the patterns and processes of natural systems have the potential to form a new aesthetic for landscape design —an aesthetic based on fractal rather than Euclidean geometry. Perry, Reeves and Sim (2008) have shown that it is possible to differentiate between different landscape forms by fractal analysis. However, this research also shows that individual scenes from within very different landscape forms can possess the same fractal properties. Early data, revealed by transforming landscape images from the spatial to the frequency domain, using the fast Fourier transform, suggest that fractal patterning can have a significant effect within the landscape. In fact, it may be argued that any landscape design that includes living processes will include some design element whose ultimate form can only be expressed through the mathematics of fractal geometry. This paper will present ongoing research into the potential role of fractal geometry as a basis for a new form language – a language that may articulate an aesthetic for landscape design that echoes our ecological awakening.

Impact of cognitive load and frustration on drivers’ speech [Abstract]

Secondary tasks such as cell phone calls or interaction with automated speech dialog systems (SDSs) increase the driver’s cognitive load as well as the probability of driving errors. This study analyzes speech production variations due to cognitive load and emotional state of drivers in real driving conditions. Speech samples were acquired from 24 female and 17 male subjects (approximately 8.5 h of data) while talking to a co-driver and communicating with two automated call centers, with emotional states (neutral, negative) and the number of necessary SDS query repetitions also labeled. A consistent shift in a number of speech production parameters (pitch, first format center frequency, spectral center of gravity, spectral energy spread, and duration of voiced segments) was observed when comparing SDS interaction against co-driver interaction; further increases were observed when considering negative emotion segments and the number of requested SDS query repetitions. A mel frequency cepstral coefficient based Gaussian mixture classifier trained on 10 male and 10 female sessions provided 91% accuracy in the open test set task of distinguishing co-driver interactions from SDS interactions, suggesting—together with the acoustic analysis—that it is possible to monitor the level of driver distraction directly from their speech.

The osteogenic properties of CaP/silk composite scaffolds

The rationale for the present study was to develop porous CaP/silk composite scaffolds with a CaP-phase distribution and pore architecture better suited to facilitate osteogenic properties of human bone mesenchymal stromal cells (BMSCs) and in vivo bone formation abilities. This was achieved by first preparing CaP/silk hybrid powders which were then incorporated into silk to obtain uniform CaP/silk composite scaffolds, by means of a freeze-drying method. The composition, microstructure and mechanical properties of the CaP/silk composite scaffolds were ascertained by X-ray diffraction (XRD), Fourier transform infrared spectra (FTIR), scanning electron microscope (SEM) and a universal mechanical testing machine. BMSCs were cultured in these scaffolds and cell proliferation analyzed by confocal microscopy and MTS assay. Alkaline phosphatase (ALP) activity and osteogenic gene expression were assayed to determine if osteogenic differentiation had taken place. A calvarial defect model in SCID mice was used to determine the in vivo bone forming ability of the hybrid CaP/silk scaffolds. Our results showed that incorporating the hybrid CaP/silk powders into silk scaffolds improved both pore structure architecture and distribution of CaP powders in the composite scaffolds. By incorporating the CaP phase into silk scaffolds in vitro osteogenic differentiation of BMSCs was enhanced and there was increased in vivo cancellous bone formation. Here we report a method with which to prepare Ca/P composite scaffolds with a pore structure and Ca/P distribution better suited to facilitate BMSC differentiation and bone formation.

Spontaneous pacing during overground hill running

Purpose: To investigate speed regulation during overground running on undulating terrain. Methods: Following an initial laboratory session to calculate physiological thresholds, eight experienced runners completed a spontaneously paced time trial over 3 laps of an outdoor course involving uphill, downhill and level sections. A portable gas analyser, GPS receiver and activity monitor were used to collect physiological, speed and stride frequency data. Results: Participants ran 23% slower on uphills and 13.8% faster on downhills compared with level sections. Speeds on level sections were significantly different for 78.4 ± 7.0 seconds following an uphill and 23.6 ± 2.2 seconds following a downhill. Speed changes were primarily regulated by stride length which was 20.5% shorter uphill and 16.2% longer downhill, while stride frequency was relatively stable. Oxygen consumption averaged 100.4% of runner’s individual ventilatory thresholds on uphills, 78.9% on downhills and 89.3% on level sections. 89% of group level speed was predicted using a modified gradient factor. Individuals adopted distinct pacing strategies, both across laps and as a function of gradient. Conclusions: Speed was best predicted using a weighted factor to account for prior and current gradients. Oxygen consumption (VO2) limited runner’s speeds only on uphill sections, and was maintained in line with individual ventilatory thresholds. Running speed showed larger individual variation on downhill sections, while speed on the level was systematically influenced by the preceding gradient. Runners who varied their pace more as a function of gradient showed a more consistent level of oxygen consumption. These results suggest that optimising time on the level sections after hills offers the greatest potential to minimise overall time when running over undulating terrain.

Examining the role of human resource management in continuous improvement

The objective of this paper is to take a first step in developing a theoretical framework describing the role of HRM in successful CI, based on the current literature from both fields. To this end, elements from the CI Maturity Model and a framework depicting the role of HRM in innovation serve as a foundation for examining how specific bundles of HRM practices utilised during different phases of the CI implementation process may contribute to sustained organisational and enhanced operational performance. The primary contribution of this paper is theoretical; however, the framework has practical value in that it suggests important relationships between HRM practices and behaviours necessary for successful CI. A preliminary test of the framework in an empirical setting is summarised at the conclusion of this paper, where a number of possible research avenues are also suggested.

Wireless channel modelling for indoor populated environments : statistical characterization of narrowband channels at 2.45 Ghz and 5.2 Ghz

The recent development of indoor wireless local area network (WLAN) standards at 2.45 GHz and 5 GHz has led to increased interest in propagation studies at these frequency bands. Within the indoor environment, human body effects can strongly reduce the quality of wireless communication systems. Human body effects can cause temporal variations and shadowing due to pedestrian movement and antenna- body interaction with portable terminals. This book presents a statistical characterisation, based on measurements, of human body effects on indoor narrowband channels at 2.45 GHz and at 5.2 GHz. A novel cumulative distribution function (CDF) that models the 5 GHz narrowband channel in populated indoor environments is proposed. This novel CDF describes the received envelope in terms of pedestrian traffic. In addition, a novel channel model for the populated indoor environment is proposed for the Multiple-Input Multiple-Output (MIMO) narrowband channel in presence of pedestrians at 2.45 GHz. Results suggest that practical MIMO systems must be sufficiently adaptive if they are to benefit from the capacity enhancement caused by pedestrian movement.

Contrasting scenarios : embracing speech recognition

The purpose of this chapter is to describe the use of caricatured contrasting scenarios (Bødker, 2000) and how they can be used to consider potential designs for disruptive technologies. The disruptive technology in this case is Automatic Speech Recognition (ASR) software in workplace settings. The particular workplace is the Magistrates Court of the Australian Capital Territory.----- Caricatured contrasting scenarios are ideally suited to exploring how ASR might be implemented in a particular setting because they allow potential implementations to be “sketched” quickly and with little effort. This sketching of potential interactions and the emphasis of both positive and negative outcomes allows the benefits and pitfalls of design decisions to become apparent.----- A brief description of the Court is given, describing the reasons for choosing the Court for this case study. The work of the Court is framed as taking place in two modes: Front of house, where the courtroom itself is, and backstage, where documents are processed and the business of the court is recorded and encoded into various systems.----- Caricatured contrasting scenarios describing the introduction of ASR to the front of house are presented and then analysed. These scenarios show that the introduction of ASR to the court would be highly problematic.----- The final section describes how ASR could be re-imagined in order to make it useful for the court. A final scenario is presented that describes how this re-imagined ASR could be integrated into both the front of house and backstage of the court in a way that could strengthen both processes.

What are students' understandings of how digital tools contribute to learning in design disciplines?

Building Information Modelling (BIM) is evolving in the Construction Industry as a successor to CAD. CAD is mostly a technical tool that conforms to existing industry practices, however BIM has the capacity to revolutionise industry practice. Rather than producing representations of design intent, BIM produces an exact Virtual Prototype of any building that in an ideal situation is centrally stored and freely exchanged between the project team, facilitating collaboration and allowing experimentation in design. Exposing design students to this technology through their formal studies allows them to engage with cutting edge industry practices and to help shape the industry upon their graduation. Since this technology is relatively new to the construction industry, there are no accepted models for how to “teach” BIM effectively at university level. Developing learning models to enable students to make the most out of their learning with BIM presents significant challenges to those teaching in the field of design. To date there are also no studies of students experiences of using this technology. This research reports on the introduction of Building Information Modeling (BIM) software into a second year Bachelor of Design course. This software has the potential to change industry standards through its ability to revolutionise the work practices of those involved in large scale design projects. Students’ understandings and experiences of using the software in order to complete design projects as part of their assessment are reported here. In depth semi-structured interviews with 6 students revealed that students had views that ranged from novice to sophisticate about the software. They had variations in understanding of how the software could be used to complete course requirements, to assist with the design process and in the workplace. They had engaged in limited exploration of the collaborative potential of the software as a design tool. Their understanding of the significance of BIM for the workplace was also variable. The results indicate that students are beginning to develop an appreciation for how BIM could aid or constrain the work of designers, but that this appreciation is highly varied and likely to be dependent on the students’ previous experiences of working in a design studio environment. Their range of understandings of the significance of the technology is a reflection of their level of development as designers (they are “novice” designers). The results also indicate that there is a need for subjects in later years of the course that allow students to specialise in the area of digital design and to develop more sophisticated views of the role of technology in the design process. There is also a need to capitalise on the collaborative potential inherent in the software in order to realise its capability to streamline some aspects of the design process. As students become more sophisticated designers we should explore their understanding of the role of technology as a design tool in more depth in order to make recommendations for improvements to teaching and learning practice related to BIM and other digital design tools.

Exploration and exploitation : the interplay between knowledge and continuous innovation

In rapidly changing environments, organisations require dynamic capabilities to integrate, build and reconfigure resources and competencies to achieve continuous innovation. Although tangible resources are important to promoting the firm’s ability to act, capabilities fundamentally rest in the knowledge created and accumulated by the firm through human capital, organisational routines, processes, practices and norms. The exploration for new ideas, technologies and knowledge – to one side – and – on the other one – the exploitation of existing and new knowledge is essential for continuous innovation. Firms need to decide how best to allocate their scarce resources for both activities and at the same time build dynamic capabilities to keep up with changing market conditions. This in turn, is influenced by the absorptive capacity of the firm to assimilate knowledge. This paper presents a case study that investigates the sources of knowledge in an engineering firm in Australia, and how it is organised and processed. As information pervades the firm from both internal and external sources, individuals integrate knowledge using both exploration and exploitation approaches. The findings illustrate that absorptive capacity can encourage greater leverage for exploration potential leading to radical innovation; and reconfiguring exploitable knowledge for incremental improvements. This study provides an insight for managers in quest of improving knowledge strategies and continuous innovation. It also makes significant theoretical contributions to the literature through extending the concepts of

Solar powered UAV for fire prevention and planning

This project aims to develop a methodology for designing and conducting a systems engineering analysis to build and fly continuously, day and night, propelled uniquely by solar energy for one week with a 0.25Kg payload consuming 0.5 watt without fuel or pollution. An airplane able to fly autonomously for many days could find many applications. Including coastal or border surveillance, atmospherical and weather research and prediction, environmental, forestry, agricultural, and oceanic monitoring, imaging for the media and real-estate industries, etc. Additional advantages of solar airplanes are their low cost and the simplicity with which they can be launched. For example, in the case of potential forest fire risks during a warm and dry period, swarms of solar airplanes, easily launched with the hand, could efficiently monitor a large surface, reporting rapidly any fire starts. This would allow a fast intervention and thus reduce the cost of such disaster, in terms of human and material losses. At higher dimension, solar HALE platforms are expected to play a major role as communication relays and could replace advantageously satellites in a near future.

Robust speech recognition using speech enhancement

Automatic Speech Recognition (ASR) has matured into a technology which is becoming more common in our everyday lives, and is emerging as a necessity to minimise driver distraction when operating in-car systems such as navigation and infotainment. In “noise-free” environments, word recognition performance of these systems has been shown to approach 100%, however this performance degrades rapidly as the level of background noise is increased. Speech enhancement is a popular method for making ASR systems more ro- bust. Single-channel spectral subtraction was originally designed to improve hu- man speech intelligibility and many attempts have been made to optimise this algorithm in terms of signal-based metrics such as maximised Signal-to-Noise Ratio (SNR) or minimised speech distortion. Such metrics are used to assess en- hancement performance for intelligibility not speech recognition, therefore mak- ing them sub-optimal ASR applications. This research investigates two methods for closely coupling subtractive-type enhancement algorithms with ASR: (a) a computationally-efficient Mel-filterbank noise subtraction technique based on likelihood-maximisation (LIMA), and (b) in- troducing phase spectrum information to enable spectral subtraction in the com- plex frequency domain. Likelihood-maximisation uses gradient-descent to optimise parameters of the enhancement algorithm to best fit the acoustic speech model given a word se- quence known a priori. Whilst this technique is shown to improve the ASR word accuracy performance, it is also identified to be particularly sensitive to non-noise mismatches between the training and testing data. Phase information has long been ignored in spectral subtraction as it is deemed to have little effect on human intelligibility. In this work it is shown that phase information is important in obtaining highly accurate estimates of clean speech magnitudes which are typically used in ASR feature extraction. Phase Estimation via Delay Projection is proposed based on the stationarity of sinusoidal signals, and demonstrates the potential to produce improvements in ASR word accuracy in a wide range of SNR. Throughout the dissertation, consideration is given to practical implemen- tation in vehicular environments which resulted in two novel contributions – a LIMA framework which takes advantage of the grounding procedure common to speech dialogue systems, and a resource-saving formulation of frequency-domain spectral subtraction for realisation in field-programmable gate array hardware. The techniques proposed in this dissertation were evaluated using the Aus- tralian English In-Car Speech Corpus which was collected as part of this work. This database is the first of its kind within Australia and captures real in-car speech of 50 native Australian speakers in seven driving conditions common to Australian environments.

Use of advanced technology videotapes in the delivery of structural engineering courses

A teaching and learning development project is currently under way at Queensland University of Technology to develop advanced technology videotapes for use with the delivery of structural engineering courses. These tapes consist of integrated computer and laboratory simulations of important concepts, and behaviour of structures and their components for a number of structural engineering subjects. They will be used as part of the regular lectures and thus will not only improve the quality of lectures and learning environment, but also will be able to replace the ever-dwindling laboratory teaching in these subjects. The use of these videotapes, developed using advanced computer graphics, data visualization and video technologies, will enrich the learning process of the current diverse engineering student body. This paper presents the details of this new method, the methodology used, the results and evaluation in relation to one of the structural engineering subjects, steel structures.

Improving automatic speaker verification using SVM techniques

Automatic recognition of people is an active field of research with important forensic and security applications. In these applications, it is not always possible for the subject to be in close proximity to the system. Voice represents a human behavioural trait which can be used to recognise people in such situations. Automatic Speaker Verification (ASV) is the process of verifying a persons identity through the analysis of their speech and enables recognition of a subject at a distance over a telephone channel { wired or wireless. A significant amount of research has focussed on the application of Gaussian mixture model (GMM) techniques to speaker verification systems providing state-of-the-art performance. GMM's are a type of generative classifier trained to model the probability distribution of the features used to represent a speaker. Recently introduced to the field of ASV research is the support vector machine (SVM). An SVM is a discriminative classifier requiring examples from both positive and negative classes to train a speaker model. The SVM is based on margin maximisation whereby a hyperplane attempts to separate classes in a high dimensional space. SVMs applied to the task of speaker verification have shown high potential, particularly when used to complement current GMM-based techniques in hybrid systems. This work aims to improve the performance of ASV systems using novel and innovative SVM-based techniques. Research was divided into three main themes: session variability compensation for SVMs; unsupervised model adaptation; and impostor dataset selection. The first theme investigated the differences between the GMM and SVM domains for the modelling of session variability | an aspect crucial for robust speaker verification. Techniques developed to improve the robustness of GMMbased classification were shown to bring about similar benefits to discriminative SVM classification through their integration in the hybrid GMM mean supervector SVM classifier. Further, the domains for the modelling of session variation were contrasted to find a number of common factors, however, the SVM-domain consistently provided marginally better session variation compensation. Minimal complementary information was found between the techniques due to the similarities in how they achieved their objectives. The second theme saw the proposal of a novel model for the purpose of session variation compensation in ASV systems. Continuous progressive model adaptation attempts to improve speaker models by retraining them after exploiting all encountered test utterances during normal use of the system. The introduction of the weight-based factor analysis model provided significant performance improvements of over 60% in an unsupervised scenario. SVM-based classification was then integrated into the progressive system providing further benefits in performance over the GMM counterpart. Analysis demonstrated that SVMs also hold several beneficial characteristics to the task of unsupervised model adaptation prompting further research in the area. In pursuing the final theme, an innovative background dataset selection technique was developed. This technique selects the most appropriate subset of examples from a large and diverse set of candidate impostor observations for use as the SVM background by exploiting the SVM training process. This selection was performed on a per-observation basis so as to overcome the shortcoming of the traditional heuristic-based approach to dataset selection. Results demonstrate the approach to provide performance improvements over both the use of the complete candidate dataset and the best heuristically-selected dataset whilst being only a fraction of the size. The refined dataset was also shown to generalise well to unseen corpora and be highly applicable to the selection of impostor cohorts required in alternate techniques for speaker verification.