Dynamic Image-based visual servo control using centroid and optic flow features

This paper considers the question of designing a fully image-based visual servo control for a class of dynamic systems. The work is motivated by the ongoing development of image-based visual servo control of small aerial robotic vehicles. The kinematics and dynamics of a rigid-body dynamical system (such as a vehicle airframe) maneuvering over a flat target plane with observable features are expressed in terms of an unnormalized spherical centroid and an optic flow measurement. The image-plane dynamics with respect to force input are dependent on the height of the camera above the target plane. This dependence is compensated by introducing virtual height dynamics and adaptive estimation in the proposed control. A fully nonlinear adaptive control design is provided that ensures asymptotic stability of the closed-loop system for all feasible initial conditions. The choice of control gains is based on an analysis of the asymptotic dynamics of the system. Results from a realistic simulation are presented that demonstrate the performance of the closed-loop system. To the author's knowledge, this paper documents the first time that an image-based visual servo control has been proposed for a dynamic system using vision measurement for both position and velocity.

Why do poor people perceive poor neighbourhoods? : the role of objective neighbourhood features and psychosocial factors

Compared to people with a high socioeconomic status, those with a lower socioeconomic status are more likely to perceive their neighbourhood as unattractive and unsafe, which is associated with their lower levels of physical activity. Agreement between objective and perceived environmental factors is often found to be moderate or low, so it is questionable to what extent ‘creating supportive neighbourhoods’ would change neighbourhood perceptions. This study among residents (N=814) of fourteen neighbourhoods in the city of Eindhoven (the Netherlands), investigated to what extent socioeconomic differences in perceived neighbourhood safety and perceived neighbourhood attractiveness can be explained by five domains of objective neighbourhood features (i.e. design, traffic safety, social safety, aesthetics, and destinations), and to what extent other factors may play a role. Unfavourable neighbourhood perceptions of low socioeconomic groups partly reflected their actual less aesthetic and less safe neighbourhoods, and partly their perceptions of low social neighbourhood cohesion and adverse psychosocial circumstances.

Robust image hashing using higher order spectral features

Robust image hashing seeks to transform a given input image into a shorter hashed version using a key-dependent non-invertible transform. These image hashes can be used for watermarking, image integrity authentication or image indexing for fast retrieval. This paper introduces a new method of generating image hashes based on extracting Higher Order Spectral features from the Radon projection of an input image. The feature extraction process is non-invertible, non-linear and different hashes can be produced from the same image through the use of random permutations of the input. We show that the transform is robust to typical image transformations such as JPEG compression, noise, scaling, rotation, smoothing and cropping. We evaluate our system using a verification-style framework based on calculating false match, false non-match likelihoods using the publicly available Uncompressed Colour Image database (UCID) of 1320 images. We also compare our results to Swaminathan’s Fourier-Mellin based hashing method with at least 1% EER improvement under noise, scaling and sharpening.

A hidden Markov model and relative entropy rate approach to vision-based dim target detection for UAV sense-and-avoid

Uninhabited aerial vehicles (UAVs) are a cutting-edge technology that is at the forefront of aviation/aerospace research and development worldwide. Many consider their current military and defence applications as just a token of their enormous potential. Unlocking and fully exploiting this potential will see UAVs in a multitude of civilian applications and routinely operating alongside piloted aircraft. The key to realising the full potential of UAVs lies in addressing a host of regulatory, public relation, and technological challenges never encountered be- fore. Aircraft collision avoidance is considered to be one of the most important issues to be addressed, given its safety critical nature. The collision avoidance problem can be roughly organised into three areas: 1) Sense; 2) Detect; and 3) Avoid. Sensing is concerned with obtaining accurate and reliable information about other aircraft in the air; detection involves identifying potential collision threats based on available information; avoidance deals with the formulation and execution of appropriate manoeuvres to maintain safe separation. This thesis tackles the detection aspect of collision avoidance, via the development of a target detection algorithm that is capable of real-time operation onboard a UAV platform. One of the key challenges of the detection problem is the need to provide early warning. This translates to detecting potential threats whilst they are still far away, when their presence is likely to be obscured and hidden by noise. Another important consideration is the choice of sensors to capture target information, which has implications for the design and practical implementation of the detection algorithm. The main contributions of the thesis are: 1) the proposal of a dim target detection algorithm combining image morphology and hidden Markov model (HMM) filtering approaches; 2) the novel use of relative entropy rate (RER) concepts for HMM filter design; 3) the characterisation of algorithm detection performance based on simulated data as well as real in-flight target image data; and 4) the demonstration of the proposed algorithm's capacity for real-time target detection. We also consider the extension of HMM filtering techniques and the application of RER concepts for target heading angle estimation. In this thesis we propose a computer-vision based detection solution, due to the commercial-off-the-shelf (COTS) availability of camera hardware and the hardware's relatively low cost, power, and size requirements. The proposed target detection algorithm adopts a two-stage processing paradigm that begins with an image enhancement pre-processing stage followed by a track-before-detect (TBD) temporal processing stage that has been shown to be effective in dim target detection. We compare the performance of two candidate morphological filters for the image pre-processing stage, and propose a multiple hidden Markov model (MHMM) filter for the TBD temporal processing stage. The role of the morphological pre-processing stage is to exploit the spatial features of potential collision threats, while the MHMM filter serves to exploit the temporal characteristics or dynamics. The problem of optimising our proposed MHMM filter has been examined in detail. Our investigation has produced a novel design process for the MHMM filter that exploits information theory and entropy related concepts. The filter design process is posed as a mini-max optimisation problem based on a joint RER cost criterion. We provide proof that this joint RER cost criterion provides a bound on the conditional mean estimate (CME) performance of our MHMM filter, and this in turn establishes a strong theoretical basis connecting our filter design process to filter performance. Through this connection we can intelligently compare and optimise candidate filter models at the design stage, rather than having to resort to time consuming Monte Carlo simulations to gauge the relative performance of candidate designs. Moreover, the underlying entropy concepts are not constrained to any particular model type. This suggests that the RER concepts established here may be generalised to provide a useful design criterion for multiple model filtering approaches outside the class of HMM filters. In this thesis we also evaluate the performance of our proposed target detection algorithm under realistic operation conditions, and give consideration to the practical deployment of the detection algorithm onboard a UAV platform. Two fixed-wing UAVs were engaged to recreate various collision-course scenarios to capture highly realistic vision (from an onboard camera perspective) of the moments leading up to a collision. Based on this collected data, our proposed detection approach was able to detect targets out to distances ranging from about 400m to 900m. These distances, (with some assumptions about closing speeds and aircraft trajectories) translate to an advanced warning ahead of impact that approaches the 12.5 second response time recommended for human pilots. Furthermore, readily available graphic processing unit (GPU) based hardware is exploited for its parallel computing capabilities to demonstrate the practical feasibility of the proposed target detection algorithm. A prototype hardware-in- the-loop system has been found to be capable of achieving data processing rates sufficient for real-time operation. There is also scope for further improvement in performance through code optimisations. Overall, our proposed image-based target detection algorithm offers UAVs a cost-effective real-time target detection capability that is a step forward in ad- dressing the collision avoidance issue that is currently one of the most significant obstacles preventing widespread civilian applications of uninhabited aircraft. We also highlight that the algorithm development process has led to the discovery of a powerful multiple HMM filtering approach and a novel RER-based multiple filter design process. The utility of our multiple HMM filtering approach and RER concepts, however, extend beyond the target detection problem. This is demonstrated by our application of HMM filters and RER concepts to a heading angle estimation problem.

Crowd counting using group tracking and local features

In public venues, crowd size is a key indicator of crowd safety and stability. In this paper we propose a crowd counting algorithm that uses tracking and local features to count the number of people in each group as represented by a foreground blob segment, so that the total crowd estimate is the sum of the group sizes. Tracking is employed to improve the robustness of the estimate, by analysing the history of each group, including splitting and merging events. A simplified ground truth annotation strategy results in an approach with minimal setup requirements that is highly accurate.

Immunogenicity and protective efficacy of an anti-Streptococcus pyogenes vaccine candidate in multiple animal species

Streptococcus pyogenes, also known as Group A Streptococcus (GAS) has been associated with a range of diseases from the mild pharyngitis and pyoderma to more severe invasive infections such as streptococcal toxic shock. GAS also causes a number of non-suppurative post-infectious diseases such as rheumatic fever, rheumatic heart disease and glomerulonephritis. The large extent of GAS disease burden necessitates the need for a prophylactic vaccine that could target the diverse GAS emm types circulating globally. Anti-GAS vaccine strategies have focused primarily on the GAS M-protein, an extracellular virulence factor anchored to GAS cell wall. As opposed to the hypervariable N-terminal region, the C-terminal portion of the protein is highly conserved among different GAS emm types and is the focus of a leading GAS vaccine candidate, J8-DT/alum. The vaccine candidate J8-DT/alum was shown to be immunogenic in mice, rabbits and the non-human primates, hamadryas baboons. Similar responses to J8-DT/alum were observed after subcutaneous and intramuscular immunization with J8-DT/alum, in mice and in rabbits. Further assessment of parameters that may influence the immunogenicity of J8-DT demonstrated that the immune responses were identical in male and female mice and the use of alum as an adjuvant in the vaccine formulation significantly increased its immunogenicity, resulting in a long-lived serum IgG response. Contrary to the previous findings, the data in this thesis indicates that a primary immunization with J8-DT/alum (50ƒÊg) followed by a single boost is sufficient to generate a robust immune response in mice. As expected, the IgG response to J8- DT/alum was a Th2 type response consisting predominantly of the isotype IgG1 accompanied by lower levels of IgG2a. Intramuscular vaccination of rabbits with J8-DT/alum demonstrated that an increase in the dose of J8-DT/alum up to 500ƒÊg does not have an impact on the serum IgG titers achieved. Similar to the immune response in mice, immunization with J8-DT/alum in baboons also established that a 60ƒÊg dose compared to either 30ƒÊg or 120ƒÊg was sufficient to generate a robust immune response. Interestingly, mucosal infection of naive baboons with a M1 GAS strain did not induce a J8-specific serum IgG response. As J8-DT/alum mediated protection has been previously reported to be due to the J8- specific antibody formed, the efficacy of J8-DT antibodies was determined in vitro and in vivo. In vitro opsonization and in vivo passive transfer confirmed the protective potential of J8-DT antibodies. A reduction in the bacterial burden after challenge with a bioluminescent M49 GAS strain in mice that were passively administered J8-DT IgG established that protection due to J8-DT was mediated by antibodies. The GAS burden in infected mice was monitored using bioluminescent imaging in addition to traditional CFU assays. Bioluminescent GAS strains including the ‘rheumatogenic’ M1 GAS could not be generated due to limitations with transformation of GAS, however, a M49 GAS strain was utilized during BLI. The M49 serotype is traditionally a ‘nephritogenic’ serotype associated with post-streptococcal glomerulonephritis. Anti- J8-DT antibodies now have been shown to be protective against multiple GAS strains such as M49 and M1. This study evaluated the immunogenicity of J8-DT/alum in different species of experimental animals in preparation for phase I human clinical trials and provided the ground work for the development of a rapid non-invasive assay for evaluation of vaccine candidates.

The candidate : a novella and examination of Australian Gothic crime fiction

Australia is a land without haunted castles or subterranean corridors, without ancient graveyards or decaying monasteries, a land whose climate is rarely gloomy. Yet, the literary landscape is splattered with shades of the Gothic genre. This Gothic heritage is especially evident within elements of nineteenth century Australian sensation fiction. Australian crime fiction in the twentieth century, in keeping with this lineage, repeatedly employs elements of the Gothic, adapting and appropriating these conventions for literary effect. I believe that a ‘mélange’ of historical Gothic crime traditions could produce an exciting new mode of Gothic crime writing in the Australian context. As such, I have written a contemporary literary experiment in a Gothic crime ‘hybrid’ style: this novella forms my creative practice. The accompanying exegesis is a critical study of a selection of Australian literary works that exhibit the characteristics of both Gothic and crime genres. Through an analysis of these creative works, this study argues that the interlacing of Gothic traditions with crime writing conventions has been a noteworthy practice in Australian fiction during both the nineteenth and twentieth centuries and these literary tropes are interwoven in the writing of ‘The Candidate’, a Gothic crime novella.

Candidate metastasis suppressor genes uncovered by array comparative genomic hybridization in a mouse allograft model of prostate cancer.

Background The purpose of this study was to identify candidate metastasis suppressor genes from a mouse allograft model of prostate cancer (NE-10). This allograft model originally developed metastases by twelve weeks after implantation in male athymic nude mice, but lost the ability to metastasize after a number of in vivo passages. We performed high resolution array comparative genomic hybridization on the metastasizing and non-metastasizing allografts to identify chromosome imbalances that differed between the two groups of tumors. Results This analysis uncovered a deletion on chromosome 2 that differed between the metastasizing and non-metastasizing tumors. Bioinformatics filters were employed to mine this region of the genome for candidate metastasis suppressor genes. Of the 146 known genes that reside within the region of interest on mouse chromosome 2, four candidate metastasis suppressor genes (Slc27a2, Mall, Snrpb, and Rassf2) were identified. Quantitative expression analysis confirmed decreased expression of these genes in the metastasizing compared to non-metastasizing tumors. Conclusion This study presents combined genomics and bioinformatics approaches for identifying potential metastasis suppressor genes. The genes identified here are candidates for further studies to determine their functional role in inhibiting metastases in the NE-10 allograft model and human prostate cancer.

Dynamic visual features for audio-visual speaker verification

The cascading appearance-based (CAB) feature extraction technique has established itself as the state-of-the-art in extracting dynamic visual speech features for speech recognition. In this paper, we will focus on investigating the effectiveness of this technique for the related speaker verification application. By investigating the speaker verification ability of each stage of the cascade we will demonstrate that the same steps taken to reduce static speaker and environmental information for the visual speech recognition application also provide similar improvements for visual speaker recognition. A further study is conducted comparing synchronous HMM (SHMM) based fusion of CAB visual features and traditional perceptual linear predictive (PLP) acoustic features to show that higher complexity inherit in the SHMM approach does not appear to provide any improvement in the final audio-visual speaker verification system over simpler utterance level score fusion.

Evaluation of spectral and texture features for object-based vegetation species classification using Support Vector Machines

The use of appropriate features to characterize an output class or object is critical for all classification problems. This paper evaluates the capability of several spectral and texture features for object-based vegetation classification at the species level using airborne high resolution multispectral imagery. Image-objects as the basic classification unit were generated through image segmentation. Statistical moments extracted from original spectral bands and vegetation index image are used as feature descriptors for image objects (i.e. tree crowns). Several state-of-art texture descriptors such as Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Patterns (LBP) and its extensions are also extracted for comparison purpose. Support Vector Machine (SVM) is employed for classification in the object-feature space. The experimental results showed that incorporating spectral vegetation indices can improve the classification accuracy and obtained better results than in original spectral bands, and using moments of Ratio Vegetation Index obtained the highest average classification accuracy in our experiment. The experiments also indicate that the spectral moment features also outperform or can at least compare with the state-of-art texture descriptors in terms of classification accuracy.

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.

Predicting single-vehicle fatal crashes for two-lane rural highways in Southeastern United States

The rural two-lane highway in the southeastern United States is frequently associated with a disproportionate number of serious and fatal crashes and as such remains a focus of considerable safety research. The Georgia Department of Transportation spearheaded a regional fatal crash analysis to identify various safety performances of two-lane rural highways and to offer guidance for identifying suitable countermeasures with which to mitigate fatal crashes. The fatal crash data used in this study were compiled from Alabama, Georgia, Mississippi, and South Carolina. The database, developed for an earlier study, included 557 randomly selected fatal crashes from 1997 or 1998 or both (this varied by state). Each participating state identified the candidate crashes and performed physical or video site visits to construct crash databases with enhance site-specific information. Motivated by the hypothesis that single- and multiple-vehicle crashes arise from fundamentally different circumstances, the research team applied binary logit models to predict the probability that a fatal crash is a single-vehicle run-off-road fatal crash given roadway design characteristics, roadside environment features, and traffic conditions proximal to the crash site. A wide variety of factors appears to influence or be associated with single-vehicle fatal crashes. In a model transferability assessment, the authors determined that lane width, horizontal curvature, and ambient lighting are the only three significant variables that are consistent for single-vehicle run-off-road crashes for all study locations.

Noise robust voice activity detection using features extracted from the time-domain autocorrelation function

This paper presents a method of voice activity detection (VAD) for high noise scenarios, using a noise robust voiced speech detection feature. The developed method is based on the fusion of two systems. The first system utilises the maximum peak of the normalised time-domain autocorrelation function (MaxPeak). The second zone system uses a novel combination of cross-correlation and zero-crossing rate of the normalised autocorrelation to approximate a measure of signal pitch and periodicity (CrossCorr) that is hypothesised to be noise robust. The score outputs by the two systems are then merged using weighted sum fusion to create the proposed autocorrelation zero-crossing rate (AZR) VAD. Accuracy of AZR was compared to state of the art and standardised VAD methods and was shown to outperform the best performing system with an average relative improvement of 24.8% in half-total error rate (HTER) on the QUT-NOISE-TIMIT database created using real recordings from high-noise environments.

Interactional competence in a paired speaking test : features salient to raters

Paired speaking tests are now commonly used in both high-stakes testing and classroom assessment contexts. The co-construction of discourse by candidates is regarded as a strength of paired speaking tests, as candidates have the opportunity to display a wider range of interactional competencies, including turn taking, initiating topics and engaging in extended discourse with a partner, rather than an examiner. However, the impact of the interlocutor in such jointly negotiated discourse and the implications for assessing interactional competence are areas of concern. This article reports on the features of interactional competence that were salient to four trained raters of 12 paired speaking tests through the analysis of rater notes, stimulated verbal recalls and rater discussions. Findings enabled the identification of features of the performance noted by raters when awarding scores for interactional competence, and the particular features associated with higher and lower scores. A number of these features were seen by the raters as mutual achievements, which raises the issue of the extent to which it is possible to assess individual contributions to the co-constructed performance. The findings have implications for defining the construct of interactional competence in paired speaking tests and operationalising this in rating scales.