An evaluation of different features and learning models for anomalous event detection

The huge amount of CCTV footage available makes it very burdensome to process these videos manually through human operators. This has made automated processing of video footage through computer vision technologies necessary. During the past several years, there has been a large effort to detect abnormal activities through computer vision techniques. Typically, the problem is formulated as a novelty detection task where the system is trained on normal data and is required to detect events which do not fit the learned ‘normal’ model. There is no precise and exact definition for an abnormal activity; it is dependent on the context of the scene. Hence there is a requirement for different feature sets to detect different kinds of abnormal activities. In this work we evaluate the performance of different state of the art features to detect the presence of the abnormal objects in the scene. These include optical flow vectors to detect motion related anomalies, textures of optical flow and image textures to detect the presence of abnormal objects. These extracted features in different combinations are modeled using different state of the art models such as Gaussian mixture model(GMM) and Semi- 2D Hidden Markov model(HMM) to analyse the performances. Further we apply perspective normalization to the extracted features to compensate for perspective distortion due to the distance between the camera and objects of consideration. The proposed approach is evaluated using the publicly available UCSD datasets and we demonstrate improved performance compared to other state of the art methods.

Knowledge unlatched announces the launch of its pilot collection [Blog]

Frances Pinter and I have been visiting fellows at the Big Innovation Centre for more than a year now. Tucked away in a corner, inspired by BIC’s open innovation vision, we have been attempting to solve a problem that continues to perplex many in the era of digital affordance: creating sustainable markets for high quality new content that include free access for end users.

Digital tablets as a tool for blended learning in power engineering education

Digital tablets have been identified as a tool for enabling blended learning and supporting online teaching and learning. A small scale trial was undertaken to assess the effectiveness of this technology when applied to power engineering education. Critical findings and experiences gained from this trial, including potential benefits, presentation techniques and the resulting student feedback are presented in this paper.

A visual profile of Queensland indigenous and non-indigenous school children, and the association between vision and reading

This research investigated the prevalence of vision disorders in Queensland Indigenous primary school children, creating the first comprehensive visual profile of Indigenous children. Findings showed reduced convergence ability and reduced visual information processing skills were more common in Indigenous compared to non-Indigenous children. Reduced visual information processing skills were also associated with reduced reading outcomes in both groups of children. As early detection of visual disorders is important, the research also reviewed the delivery of screening programs across Queensland and proposed a model for improved coordination and service delivery of vision screening to Queensland school children.

Conceptualising the learning organisation : creating a maturity framework to develop a shared understanding of the library's role in literacy and learning

Although they sit outside the formal education sector, libraries are intrinsically centres of learning where people can engage with knowledge and ideas and acquire the literacy skills that are essential for active participation in an increasingly digital society. In Australia, National and State Libraries Australasia (NSLA) has acknowledged the need to not only better understand the general concept of the library as a learning institution, but also to help the individual NSLA libraries specifically identify their capabilities in this arena. The NSLA Literacy and Learning project aimed to improve the members' organisational comprehension and practice as learning institutions and to help them conceptualise their ability to deliver literacy and learning programmes that will benefit their staff and their communities. The NSLA concept of learning institution encompassed two discrete lenses: the internal lens of the library's own organisational understanding and practice, and the external lens of the clients who engage in the literacy and learning programmes delivered by the library. The ultimate goal was to develop a matrix which could enable libraries to assess their perceived levels of maturity as learning institutions along a continuum of emerging to active capabilities. The matrix should also serve as a tool for shared understanding about the NSLA's own strategic directions in the literacy and learning space. This case study documents the evolving process of developing a learning institution maturity framework for libraries that considers individual, team and organisational learning, as well as clients' interactions with the organisation, with the goal of producing a framework that has the potential to measure the value of learning and growth in both the library's staff and the library's communities

Automated diagnosis of Age-related Macular Degeneration using greyscale features from digital fundus images

Age-related Macular Degeneration (AMD) is one of the major causes of vision loss and blindness in ageing population. Currently, there is no cure for AMD, however early detection and subsequent treatment may prevent the severe vision loss or slow the progression of the disease. AMD can be classified into two types: dry and wet AMDs. The people with macular degeneration are mostly affected by dry AMD. Early symptoms of AMD are formation of drusen and yellow pigmentation. These lesions are identified by manual inspection of fundus images by the ophthalmologists. It is a time consuming, tiresome process, and hence an automated diagnosis of AMD screening tool can aid clinicians in their diagnosis significantly. This study proposes an automated dry AMD detection system using various entropies (Shannon, Kapur, Renyi and Yager), Higher Order Spectra (HOS) bispectra features, Fractional Dimension (FD), and Gabor wavelet features extracted from greyscale fundus images. The features are ranked using t-test, Kullback–Lieber Divergence (KLD), Chernoff Bound and Bhattacharyya Distance (CBBD), Receiver Operating Characteristics (ROC) curve-based and Wilcoxon ranking methods in order to select optimum features and classified into normal and AMD classes using Naive Bayes (NB), k-Nearest Neighbour (k-NN), Probabilistic Neural Network (PNN), Decision Tree (DT) and Support Vector Machine (SVM) classifiers. The performance of the proposed system is evaluated using private (Kasturba Medical Hospital, Manipal, India), Automated Retinal Image Analysis (ARIA) and STructured Analysis of the Retina (STARE) datasets. The proposed system yielded the highest average classification accuracies of 90.19%, 95.07% and 95% with 42, 54 and 38 optimal ranked features using SVM classifier for private, ARIA and STARE datasets respectively. This automated AMD detection system can be used for mass fundus image screening and aid clinicians by making better use of their expertise on selected images that require further examination.

Street reads - Creating mobile walking events

By 2012 mobile devices had become the main interface for people to access information about anything from their current GPS position to the latest book reviews. What was less accessible were tools and techniques for writers to leverage this new technology to construct and distribute located stories. This project began with a series of master classes for local Brisbane writers to demonstrate processes and techniques for imagining, constructing and distributing stories. Most significantly, this project equipped writers with how to identify and adopt various mobile services and applications to research, produce and deliver packaged multi-modal content for readers to access and experience stories in the very locations from which they were inspired. Four stories by four writers were selected to be developed and published as location-based events in four different neighbourhoods across Brisbane. These writers were mentored throughout the writing process and a model was developed for them to simply upload several multi-modal chapters for access on location by readers using QR codes. These activities culminated in a major 25 day event presented by Brisbane City Council and supported by Brisbane Writers Festival and Queensland Writers Centre. The 'Street Reads' event presented the four stories on location in Cannon Hill, Darra, Toowong and West End. The significance of the Street Reads project went beyond extending the capacity for writers to access mobile technologies as a new platform for distributing stories. This event also motivated readers to travel to neighbourhoods to experience them in ways that had not previously imagined possible. These located stories were fictionalisations of actual events and characters that have current and historic importance to these places. These histories are hidden from view and yet can provide locals and visitors with a new found appreciation for the past and set an example for how neighbourhoods can become active stages for the sharing of stories inspiring a deeper connection with each other and an agency for participating in the development of the identity of the local places they inhabit together. Due to the success of the project and by employing more advanced tools now available, Street reads has been further developed by Brisbane City Council and is now available as a the Story City App available for download at itunes.

Enterprise Business Technology Governance: Three Competencies to Build Board Digital Leadership Capability.

The competent leadership of digital transformation needs to involve the board of directors. The reported lack of such capability in boards is becoming a pressing issue. A part of leadership in such transformation is the board of director’s competence to lead Enterprise Business Technology Governance (EBTG). In this paper we take the position that EBTG competencies are essential in boards, because competent EBTG has been shown to contribute to increased revenue, profit, and returns. We update and expand on the results of a multi-method approach to the development of a set of three board of director competencies needed for effective EBTG.

100 futures now : designing sustainable futures

This presentation tells the story of an initiative in middle schooling at Kelvin Grove State College that begins in the Art studios, but reaches out to other disciplines and approaches, and to community and industry partners. It is inspired by the potential of 'future thinking' to become a compelling focus in contemporary art and design. Ethically it espouses a simple premise": every student in our classrooms now has a stake in creating livable, democratic and creative futures. Every student has the potential to be an active force in making that future. "100 Futures Now" is a project that envisages creative and imaginative students working in collaboration with artists and designers to visualize amazing futures and communicate their vision through art and design. "100 Futures Now" is one in a series of innovative curriculum initiatives at Kelvin Grove State College designed to build sustainable practice in arts education with the support of partners in industry and universities and with resident artists and designers. The model blends elements of art and design methodology to focus on the critical and creative thinking skills prioritised in ACARA and 21st century curriculum. The organisers are developing a sustainable model for working with resident artists that goes beyond a single arts intervention or extension/enrichment experience. In this model artists and designers are collaborators in the design of learning experiences that support future programs. This model also looks to transfer the benefits of residencies to the wider school community (in this case to middle schooling curriculum) and to teachers in other curriculum areas, and not exclusively to the immediate target group. In "100 Futures Now", story-making is the engine that powers the creative process. For this reason the program uses a series of imaginative scenarios, including those of speculative fiction and science, as departure points for inquiry, and applies the methodologies of arts and design practice to explore and express student story telling and story making. The story-making responses of student teams will naturally be expressed multimodally through visual art, design artifacts, installation, performance and digital works. The project’s focus on narratives and its modes of communication (performance/installation) are inspired by the work of experimental contemporary design practices and the speculative scenarios of U.K. based designers Anthony Dunne and Fiona Raby. Thanks to the support of an Arts Queensland Artist-in -Residence grant in 2014, resident artists and designers who work with a diversity of ideas and approaches ranging over science, bio-ethics, biodiversity, behavior and ethics, ambient sound, urbanism, food, and wearable design, will work with middle school students as catalysts for deeper thinking and creative action. All these rich fields for future speculation will become triggers for team inquiry into the deeper connections between the past, the present, and future challenges such as climate, waste, energy, sustainability and resilience. These imagined futures will form the platform for a critical, sustainability/design futures approach that will involve questioning assumptions and empowering students as agents rather than consumers of change.

A convolutional neural network for automatic analysis of aerial imagery

This paper introduces a new method to automate the detection of marine species in aerial imagery using a Machine Learning approach. Our proposed system has at its core, a convolutional neural network. We compare this trainable classifier to a handcrafted classifier based on color features, entropy and shape analysis. Experiments demonstrate that the convolutional neural network outperforms the handcrafted solution. We also introduce a negative training example-selection method for situations where the original training set consists of a collection of labeled images in which the objects of interest (positive examples) have been marked by a bounding box. We show that picking random rectangles from the background is not necessarily the best way to generate useful negative examples with respect to learning.

Using digital projection to evoke aesthetic ideas in performance

This practice-led research project investigated the use of digital projection as a compositional tool in live performance. The project was carried out through the creation of a new Australian theatre work called Genesis that poetically integrated digital projection and live performance. The investigation produced a framework for creating powerful theatrical sequences where the themes and ideas of the show were embedded inside particular performance gestures prompting an expanded aesthetic perception of the content.

Large scale multiuser digital mind mapping tool : a study in usability

Project work can involve multiple people from varying disciplines coming together to solve problems as a group. Large scale interactive displays are presenting new opportunities to support such interactions with interactive and semantically enabled cooperative work tools such as intelligent mind maps. In this paper, we present a novel digital, touch-enabled mind-mapping tool as a first step towards achieving such a vision. This first prototype allows an evaluation of the benefits of a digital environment for a task that would otherwise be performed on paper or flat interactive surfaces. Observations and surveys of 12 participants in 3 groups allowed the formulation of several recommendations for further research into: new methods for capturing text input on touch screens; inclusion of complex structures; multi-user environments and how users make the shift from single- user applications; and how best to navigate large screen real estate in a touch-enabled, co-present multi-user setting.

From ideation to sculpture : digital sculpting as the primary 'ideation' tool for creature design

This research project examined the potential for circumventing drawing in the ideation process by adopting digital sculpture as the primary conceptual development and design tool for the digital sculpting of creature designs. Through a series of experimental research cycles, multiple frameworks were explored with the aim of identifying a methodology for creating '3D sculpted sketches' for the initial phases of the ideation process. This research project acknowledges that drawing still remains the predominant method of visualising design ideas for characters and creatures for many artists. However, alongside other ideation techniques digital sculpting can function as a rapid and responsive tool to visualize and explore forms in a digital sculpting environment for the conceptualisation of multiple creature design variations. The results of this study are significant for emerging digital sculptors who may not necessarily have a well-defined creative brief or initial concept.

Creating a legal framework for copyright management of open access within the Australian academic and research sector

This chapter analyses the copyright law framework needed to ensure open access to outputs of the Australian academic and research sector such as journal articles and theses. It overviews the new knowledge landscape, the principles of copyright law, the concept of open access to knowledge, the recently developed open content models of copyright licensing and the challenges faced in providing greater access to knowledge and research outputs.

Settings, subjects and stories: Creating Australian cinema

The story of Australian cinema is often told as one of brave and often futile struggle by passionate and talented filmmakers to tell Australian stories against the backdrop of an industry dominated locally as well as globally by Hollywood and its agents. In theses narratives international interests are often cast as the villains in the valiant struggle for national filmic self-expression. But such a focus on the national aspects of Australian cinema elides the depth of the international aspect of Australian cinema. A legend has grown around the last decade of the nineteenth century as a time of intense artistic and political activity when a national sensibility welled in writing, poetry and painting. Film too played a part in creating and sharing a vision of a nation, but from the earliest days film also linked Australia to the world.