The backfilled GEI : a cross-capture modality gait feature for frontal and side-view gait recognition

In this paper, we propose a novel direction for gait recognition research by proposing a new capture-modality independent, appearance-based feature which we call the Back-filled Gait Energy Image (BGEI). It can can be constructed from both frontal depth images, as well as the more commonly used side-view silhouettes, allowing the feature to be applied across these two differing capturing systems using the same enrolled database. To evaluate this new feature, a frontally captured depth-based gait dataset was created containing 37 unique subjects, a subset of which also contained sequences captured from the side. The results demonstrate that the BGEI can effectively be used to identify subjects through their gait across these two differing input devices, achieving rank-1 match rate of 100%, in our experiments. We also compare the BGEI against the GEI and GEV in their respective domains, using the CASIA dataset and our depth dataset, showing that it compares favourably against them. The experiments conducted were performed using a sparse representation based classifier with a locally discriminating input feature space, which show significant improvement in performance over other classifiers used in gait recognition literature, achieving state of the art results with the GEI on the CASIA dataset.

Spatio temporal feature evaluation for action recognition

Spatio-Temporal interest points are the most popular feature representation in the field of action recognition. A variety of methods have been proposed to detect and describe local patches in video with several techniques reporting state of the art performance for action recognition. However, the reported results are obtained under different experimental settings with different datasets, making it difficult to compare the various approaches. As a result of this, we seek to comprehensively evaluate state of the art spatio- temporal features under a common evaluation framework with popular benchmark datasets (KTH, Weizmann) and more challenging datasets such as Hollywood2. The purpose of this work is to provide guidance for researchers, when selecting features for different applications with different environmental conditions. In this work we evaluate four popular descriptors (HOG, HOF, HOG/HOF, HOG3D) using a popular bag of visual features representation, and Support Vector Machines (SVM)for classification. Moreover, we provide an in-depth analysis of local feature descriptors and optimize the codebook sizes for different datasets with different descriptors. In this paper, we demonstrate that motion based features offer better performance than those that rely solely on spatial information, while features that combine both types of data are more consistent across a variety of conditions, but typically require a larger codebook for optimal performance.

Personalized recommender systems integrating tags and item taxonomy

The social tags in Web 2.0 are becoming another important information source to profile users' interests and preferences to make personalized recommendations. To solve the problem of low information sharing caused by the free-style vocabulary of tags and the long tails of the distribution of tags and items, this paper proposes an approach to integrate the social tags given by users and the item taxonomy with standard vocabulary and hierarchical structure provided by experts to make personalized recommendations. The experimental results show that the proposed approach can effectively improve the information sharing and recommendation accuracy.

A taxonomy of model-based testing approaches

Model-based testing (MBT) relies on models of a system under test and/or its environment to derive test cases for the system. This paper discusses the process of MBT and defines a taxonomy that covers the key aspects of MBT approaches. It is intended to help with understanding the characteristics, similarities and differences of those approaches, and with classifying the approach used in a particular MBT tool. To illustrate the taxonomy, a description of how three different examples of MBT tools fit into the taxonomy is provided.

Systematics and biology of the iconic Australian scribbly gum moths Ogmograptis Meyrick (Lepidoptera: Bucculatricidae) and their unique insect-plant interaction

The larvae of particular Ogmograptis spp. produce distinctive scribbles on some smooth-barked Eucalyptus spp. which are a common feature on many ornamental and forest trees in Australia. However, although they are conspicuous in the environment the systematics and biology of the genus has been poorly studied. This has been addressed through detailed field and laboratory studies of their biology of three species (O. racemosa Horak sp. nov., O. fraxinoides Horak sp. nov., O. scribula Meyrick), in conjunction with a comprehensive taxonomic revision support by a molecular phylogeny utilising the mitochondrial Cox1 and nuclear 18S genes. In brief, eggs are laid in bark depressions and the first instar larvae bore into the bark to the level where the future cork cambium forms (the phellegen). Early instar larvae bore wide, arcing tracks in this layer before forming a tighter zig-zag shaped pattern. The second last instar turns and bores either closely parallel to the initial mine or doubles its width, along the zig-zag shaped mine. The final instar possesses legs and a spinneret (unlike the earlier instars) and feeds exclusively on callus tissue which forms within the zig-zag shaped mine formed by the previous instar, before emerging from the bark to pupate at the base of the tree. The scars of mines them become visible scribble following the shedding of bark. Sequence data confirm the placement of Ogmograptis within the Bucculatricidae, suggest that the larvae responsible for the ‘ghost scribbles’ (unpigmented, raised scars found on smooth-barked eucalypts) are members of the genus Tritymba, and support the morphology-based species groups proposed for Ogmograptis. The formerly monotypic genus Ogmograptis Meyrick is revised and divided into three species groups. Eleven new species are described: Ogmograptis fraxinoides Horak sp. nov., Ogmograptis racemosa Horak sp. nov. and Ogmograptis pilularis Horak sp. nov. forming the scribula group with Ogmograptis scribula Meyrick; Ogmograptis maxdayi Horak sp. nov., Ogmograptis barloworum Horak sp. nov., Ogmograptis paucidentatus Horak sp. nov., Ogmograptis rodens Horak sp. nov., Ogmograptis bignathifer Horak sp. nov. and Ogmograptis inornatus Horak sp. nov. as the maxdayi group; Ogmograptis bipunctatus Horak sp. nov., Ogmograptis pulcher Horak sp. nov., Ogmograptis triradiata (Turner) comb. nov. and Ogmograptis centrospila (Turner) comb. nov. as the triradiata group. Ogmograptis notosema (Meyrick) cannot be assigned to a species group as the holotype has not been located. Three unique synapomorphies, all derived from immatures, redefine the family Bucculatricidae, uniting Ogmograptis, Tritymba Meyrick (both Australian) and Leucoedemia Scoble & Scholtz (African) with Bucculatrix Zeller, which is the sister group of the southern hemisphere genera. The systematic history of Ogmograptis and the Bucculatricidae is discussed.

Adopting relevance feature to learn personalized ontologies

Relevance feature and ontology are two core components to learn personalized ontologies for concept-based retrievals. However, how to associate user native information with common knowledge is an urgent issue. This paper proposes a sound solution by matching relevance feature mined from local instances with concepts existing in a global knowledge base. The matched concepts and their relations are used to learn personalized ontologies. The proposed method is evaluated elaborately by comparing it against three benchmark models. The evaluation demonstrates the matching is successful by achieving remarkable improvements in information filtering measurements.

Ontology-based feature modeling for construction information extraction from a building information model

A building information model (BIM) provides a rich representation of a building's design. However, there are many challenges in getting construction-specific information from a BIM, limiting the usability of BIM for construction and other downstream processes. This paper describes a novel approach that utilizes ontology-based feature modeling, automatic feature extraction based on ifcXML, and query processing to extract information relevant to construction practitioners from a given BIM. The feature ontology generically represents construction-specific information that is useful for a broad range of construction management functions. The software prototype uses the ontology to transform the designer-focused BIM into a construction-specific feature-based model (FBM). The formal query methods operate on the FBM to further help construction users to quickly extract the necessary information from a BIM. Our tests demonstrate that this approach provides a richer representation of construction-specific information compared to existing BIM tools.

Semantic labelling for document feature patterns using ontological subjects

Finding and labelling semantic features patterns of documents in a large, spatial corpus is a challenging problem. Text documents have characteristics that make semantic labelling difficult; the rapidly increasing volume of online documents makes a bottleneck in finding meaningful textual patterns. Aiming to deal with these issues, we propose an unsupervised documnent labelling approach based on semantic content and feature patterns. A world ontology with extensive topic coverage is exploited to supply controlled, structured subjects for labelling. An algorithm is also introduced to reduce dimensionality based on the study of ontological structure. The proposed approach was promisingly evaluated by compared with typical machine learning methods including SVMs, Rocchio, and kNN.

Taxonomy and palaeoecology of Quaternary faunas from caves in eastern tropical Queensland : a record of broad-scale environmental change

Analysis of fossils from cave deposits at Mount Etna (eastern-central Queensland) has established that a species-rich rainforest palaeoenvironment existed in that area during the middle Pleistocene. This unexpected finding has implications for several fields (e.g., biogeography/phylogeography of rainforest-adapted taxa, and the impact of climate change on rainforest communities), but it was unknown whether the Mount Etna sites represented a small refugial patch of rainforest or was more widespread. In this study numerous bone deposits in caves in north-east Queensland are analysed to reconstruct the environmental history of the area during the late Quaternary. Study sites are in the Chillagoe/Mitchell Palmer and Broken River/Christmas Creek areas. The cave fossil records in these study areas are compared with dated (middle Pleistocene-Holocene) cave sites in the Mount Etna area. Substantial taxonomic work on the Mount Etna faunas (particularly dasyurid marsupials and murine rodents) is also presented as a prerequisite for meaningful comparison with the study sites further north. Middle Pleistocene sites at Mount Etna contain species indicative of a rainforest palaeoenvironment. Small mammal assemblages in the Mount Etna rainforest sites (>500-280 ka) are unexpectedly diverse and composed almost entirely of new species. Included in the rainforest assemblages are lineages with no extant representatives in rainforest (e.g., Leggadina), one genus previously known only from New Guinea (Abeomelomys), and forms that appear to bridge gaps between related but morphologically-divergent extant taxa ('B-rat' and 'Pseudomys C'). Curiously, some taxa (e.g., Melomys spp.) are notable for their absence from the Mount Etna rainforest sites. After 280 ka the rainforest faunas are replaced by species adapted to open, dry habitats. At that time the extinct ‘rainforest’ dasyurids and rodents are replaced by species that are either extant or recently extant. By the late Pleistocene all ‘rainforest’ and several ‘dry’ taxa are locally or completely extinct, and the small mammal fauna resembles that found in the area today. The faunal/environmental changes recorded in the Mount Etna sites were interpreted by previous workers as the result of shifts in climate during the Pleistocene. Many samples from caves in the Chillagoe/Mitchell-Palmer and Broken River/Christmas Creek areas are held in the Queensland Museum’s collection. These, supplemented with additional samples collected in the field as well as samples supplied by other workers, were systematically and palaeoecologically analysed for the first time. Palaeoecological interpretation of the faunal assemblages in the sites suggests that they encompass a similar array of palaeoenvironments as the Mount Etna sites. ‘Rainforest’ sites at the Broken River are here interpreted as being of similar age to those at Mount Etna, suggesting the possibility of extensive rainforest coverage in eastern tropical Queensland during part of the Pleistocene. Likewise, faunas suggesting open, dry palaeoenvironments are found at Chillagoe, the Broken River and Mount Etna, and may be of similar age. The 'dry' faunal assemblage at Mount Etna (Elephant hole Cave) dates to 205-170 ka. Dating of one of the Chillagoe sites (QML1067) produced a maximum age for the deposit of approximately 200 ka, and the site is interpreted as being close to that age, supporting the interpretation of roughly contemporaneous deposition at Mount Etna and Chillagoe. Finally, study sites interpreted as being of late Pleistocene-Holocene age show faunal similarities to sites of that age near Mount Etna. This study has several important implications for the biogeography and phylogeography of murine rodents, and represents a major advance in the study of the Australian murine fossil record. Likewise the survey of the northern study areas is the first systematic analysis of multiple sites in those areas, and is thus a major contribution to knowledge of tropical Australian faunas during the Quaternary. This analysis suggests that climatic changes during the Pleistocene affected a large area of eastern tropical Queensland in similar ways. Further fieldwork and dating is required to properly analyse the geographical extent and timing of faunal change in eastern tropical Queensland.

Marking out the pitch : a historiography and taxonomy of football fiction

Football, or soccer as it is more commonly referred to in Australia and the US, is arguably the world’s most popular sport. It generates a proportionate volume of related writing. Within this landscape, works of novel-length fiction are seemingly rare. This paper establishes and maps a substantial body of football fiction works, explores elements and qualities exhibited individually and collectively. In bringing together current, limited surveys of the field, it presents the first rigorous definition of football fiction and captures the first historiography of the corpus. Drawing on distant reading methods developed in conjunction with closer textual analyses, the historiography and subsequent taxonomy represent the first articulation of relationships across the body of work, identify growth areas and establish a number of movements and trends. In advancing the understanding of football fiction as a collective body, the paper lays foundations for further research and consideration of the works in generic terms.

Feature-domain super-resolution for iris recognition

Uncooperative iris identification systems at a distance suffer from poor resolution of the acquired iris images, which significantly degrades iris recognition performance. Super-resolution techniques have been employed to enhance the resolution of iris images and improve the recognition performance. However, most existing super-resolution approaches proposed for the iris biometric super-resolve pixel intensity values, rather than the actual features used for recognition. This paper thoroughly investigates transferring super-resolution of iris images from the intensity domain to the feature domain. By directly super-resolving only the features essential for recognition, and by incorporating domain specific information from iris models, improved recognition performance compared to pixel domain super-resolution can be achieved. A framework for applying super-resolution to nonlinear features in the feature-domain is proposed. Based on this framework, a novel feature-domain super-resolution approach for the iris biometric employing 2D Gabor phase-quadrant features is proposed. The approach is shown to outperform its pixel domain counterpart, as well as other feature domain super-resolution approaches and fusion techniques.

Visual guidance for fixed-wing unmanned aerial vehicles using feature tracking : application to power line inspection

This thesis presents novel vision based control solutions that enable fixed-wing Unmanned Aerial Vehicles to perform tasks of inspection over infrastructure including power lines, pipe lines and roads. This is achieved through the development of techniques that combine visual servoing with alternate manoeuvres that assist the UAV in both following and observing the feature from a downward facing camera. Control designs are developed through techniques of Image Based Visual Servoing to utilise sideslip through Skid-to-Turn and Forward-Slip manoeuvres. This allows the UAV to simultaneously track and collect data over the length of infrastructure, including straight segments and the transition where these meet.

Multi-gene phylogenetic analysis of south-east Asian pest members of the Bactrocera dorsalis species complex (Diptera: Tephritidae) does not support current taxonomy

Bactrocera dorsalis sensu stricto, B. papayae, B. philippinensis and B. carambolae are serious pest fruit fly species of the B. dorsalis complex that predominantly occur in south-east Asia and the Pacific. Identifying molecular diagnostics has proven problematic for these four taxa, a situation that cofounds biosecurity and quarantine efforts and which may be the result of at least some of these taxa representing the same biological species. We therefore conducted a phylogenetic study of these four species (and closely related outgroup taxa) based on the individuals collected from a wide geographic range; sequencing six loci (cox1, nad4-3′, CAD, period, ITS1, ITS2) for approximately 20 individuals from each of 16 sample sites. Data were analysed within maximum likelihood and Bayesian phylogenetic frameworks for individual loci and concatenated data sets for which we applied multiple monophyly and species delimitation tests. Species monophyly was measured by clade support, posterior probability or bootstrap resampling for Bayesian and likelihood analyses respectively, Rosenberg's reciprocal monophyly measure, P(AB), Rodrigo's (P(RD)) and the genealogical sorting index, gsi. We specifically tested whether there was phylogenetic support for the four 'ingroup' pest species using a data set of multiple individuals sampled from a number of populations. Based on our combined data set, Bactrocera carambolae emerges as a distinct monophyletic clade, whereas B. dorsalis s.s., B. papayae and B. philippinensis are unresolved. These data add to the growing body of evidence that B. dorsalis s.s., B. papayae and B. philippinensis are the same biological species, which poses consequences for quarantine, trade and pest management.

Patient error : a preliminary taxonomy

PURPOSE Current research on errors in health care focuses almost exclusively on system and clinician error. It tends to exclude how patients may create errors that influence their health. We aimed to identify the types of errors that patients can contribute and help manage, especially in primary care. METHODS Eleven nominal group interviews of patients and primary health care professionals were held in Auckland, New Zealand, during late 2007. Group members reported and helped to classify types of potential error by patients. We synthesized the ideas that emerged from the nominal groups into a taxonomy of patient error. RESULTS Our taxonomy is a 3-level system encompassing 70 potential types of patient error. The first level classifies 8 categories of error into 2 main groups: action errors and mental errors. The action errors, which result in part or whole from patient behavior, are attendance errors, assertion errors, and adherence errors. The mental errors, which are errors in patient thought processes, comprise memory errors, mindfulness errors, misjudgments, and—more distally—knowledge deficits and attitudes not conducive to health. CONCLUSION The taxonomy is an early attempt to understand and recognize how patients may err and what clinicians should aim to influence so they can help patients act safely. This approach begins to balance perspectives on error but requires further research. There is a need to move beyond seeing patient, clinician, and system errors as separate categories of error. An important next step may be research that attempts to understand how patients, clinicians, and systems interact to cocreate and reduce errors.

Personalized ontology learning for enhancing text mining effectiveness

Over the last decade, the majority of existing search techniques is either keyword- based or category-based, resulting in unsatisfactory effectiveness. Meanwhile, studies have illustrated that more than 80% of users preferred personalized search results. As a result, many studies paid a great deal of efforts (referred to as col- laborative filtering) investigating on personalized notions for enhancing retrieval performance. One of the fundamental yet most challenging steps is to capture precise user information needs. Most Web users are inexperienced or lack the capability to express their needs properly, whereas the existent retrieval systems are highly sensitive to vocabulary. Researchers have increasingly proposed the utilization of ontology-based tech- niques to improve current mining approaches. The related techniques are not only able to refine search intentions among specific generic domains, but also to access new knowledge by tracking semantic relations. In recent years, some researchers have attempted to build ontological user profiles according to discovered user background knowledge. The knowledge is considered to be both global and lo- cal analyses, which aim to produce tailored ontologies by a group of concepts. However, a key problem here that has not been addressed is: how to accurately match diverse local information to universal global knowledge. This research conducts a theoretical study on the use of personalized ontolo- gies to enhance text mining performance. The objective is to understand user information needs by a \bag-of-concepts" rather than \words". The concepts are gathered from a general world knowledge base named the Library of Congress Subject Headings. To return desirable search results, a novel ontology-based mining approach is introduced to discover accurate search intentions and learn personalized ontologies as user profiles. The approach can not only pinpoint users' individual intentions in a rough hierarchical structure, but can also in- terpret their needs by a set of acknowledged concepts. Along with global and local analyses, another solid concept matching approach is carried out to address about the mismatch between local information and world knowledge. Relevance features produced by the Relevance Feature Discovery model, are determined as representatives of local information. These features have been proven as the best alternative for user queries to avoid ambiguity and consistently outperform the features extracted by other filtering models. The two attempt-to-proposed ap- proaches are both evaluated by a scientific evaluation with the standard Reuters Corpus Volume 1 testing set. A comprehensive comparison is made with a num- ber of the state-of-the art baseline models, including TF-IDF, Rocchio, Okapi BM25, the deploying Pattern Taxonomy Model, and an ontology-based model. The gathered results indicate that the top precision can be improved remarkably with the proposed ontology mining approach, where the matching approach is successful and achieves significant improvements in most information filtering measurements. This research contributes to the fields of ontological filtering, user profiling, and knowledge representation. The related outputs are critical when systems are expected to return proper mining results and provide personalized services. The scientific findings have the potential to facilitate the design of advanced preference mining models, where impact on people's daily lives.