The N170 observed ‘in the wild’: robust event-related potentials to faces in cluttered dynamic visual scenes

As a social species in a constantly changing environment, humans rely heavily on the informational richness and communicative capacity of the face. Thus, understanding how the brain processes information about faces in real-time is of paramount importance. The N170 is a high temporal resolution electrophysiological index of the brain's early response to visual stimuli that is reliably elicited in carefully controlled laboratory-based studies. Although the N170 has often been reported to be of greatest amplitude to faces, there has been debate regarding whether this effect might be an artifact of certain aspects of the controlled experimental stimulation schedules and materials. To investigate whether the N170 can be identified in more realistic conditions with highly variable and cluttered visual images and accompanying auditory stimuli we recorded EEG 'in the wild', while participants watched pop videos. Scene-cuts to faces generated a clear N170 response, and this was larger than the N170 to transitions where the videos cut to non-face stimuli. Within participants, wild-type face N170 amplitudes were moderately correlated to those observed in a typical laboratory experiment. Thus, we demonstrate that the face N170 is a robust and ecologically valid phenomenon and not an artifact arising as an unintended consequence of some property of the more typical laboratory paradigm.

An MRF based abnormal event detection approach using motion and appearance features

Abnormal event detection has attracted a lot of attention in the computer vision research community during recent years due to the increased focus on automated surveillance systems to improve security in public places. Due to the scarcity of training data and the definition of an abnormality being dependent on context, abnormal event detection is generally formulated as a data-driven approach where activities are modeled in an unsupervised fashion during the training phase. In this work, we use a Gaussian mixture model (GMM) to cluster the activities during the training phase, and propose a Gaussian mixture model based Markov random field (GMM-MRF) to estimate the likelihood scores of new videos in the testing phase. Further-more, we propose two new features: optical acceleration, and the histogram of optical flow gradients; to detect the presence of any abnormal objects and speed violations in the scene. We show that our proposed method outperforms other state of the art abnormal event detection algorithms on publicly available UCSD dataset.

Evaluation of features for leaf classification in challenging conditions

Fine-grained leaf classification has concentrated on the use of traditional shape and statistical features to classify ideal images. In this paper we evaluate the effectiveness of traditional hand-crafted features and propose the use of deep convolutional neural network (ConvNet) features. We introduce a range of condition variations to explore the robustness of these features, including: translation, scaling, rotation, shading and occlusion. Evaluations on the Flavia dataset demonstrate that in ideal imaging conditions, combining traditional and ConvNet features yields state-of-theart performance with an average accuracy of 97:3%�0:6% compared to traditional features which obtain an average accuracy of 91:2%�1:6%. Further experiments show that this combined classification approach consistently outperforms the best set of traditional features by an average of 5:7% for all of the evaluated condition variations.

Injury narrative text classification : a preliminary study

Description of a patient's injuries is recorded in narrative text form by hospital emergency departments. For statistical reporting, this text data needs to be mapped to pre-defined codes. Existing research in this field uses the Naïve Bayes probabilistic method to build classifiers for mapping. In this paper, we focus on providing guidance on the selection of a classification method. We build a number of classifiers belonging to different classification families such as decision tree, probabilistic, neural networks, and instance-based, ensemble-based and kernel-based linear classifiers. An extensive pre-processing is carried out to ensure the quality of data and, in hence, the quality classification outcome. The records with a null entry in injury description are removed. The misspelling correction process is carried out by finding and replacing the misspelt word with a soundlike word. Meaningful phrases have been identified and kept, instead of removing the part of phrase as a stop word. The abbreviations appearing in many forms of entry are manually identified and only one form of abbreviations is used. Clustering is utilised to discriminate between non-frequent and frequent terms. This process reduced the number of text features dramatically from about 28,000 to 5000. The medical narrative text injury dataset, under consideration, is composed of many short documents. The data can be characterized as high-dimensional and sparse, i.e., few features are irrelevant but features are correlated with one another. Therefore, Matrix factorization techniques such as Singular Value Decomposition (SVD) and Non Negative Matrix Factorization (NNMF) have been used to map the processed feature space to a lower-dimensional feature space. Classifiers with these reduced feature space have been built. In experiments, a set of tests are conducted to reflect which classification method is best for the medical text classification. The Non Negative Matrix Factorization with Support Vector Machine method can achieve 93% precision which is higher than all the tested traditional classifiers. We also found that TF/IDF weighting which works well for long text classification is inferior to binary weighting in short document classification. Another finding is that the Top-n terms should be removed in consultation with medical experts, as it affects the classification performance.

Higher order spectra based support vector machine for arrhythmia classification

Heart rate variability (HRV) refers to the regulation of the sinoatrial node, the natural pacemaker of the heart by the sympathetic and parasympathetic branches of the autonomic nervous system. HRV analysis is an important tool to observe the heart’s ability to respond to normal regulatory impulses that affect its rhythm. Like many bio-signals, HRV signals are non-linear in nature. Higher order spectral analysis (HOS) is known to be a good tool for the analysis of non-linear systems and provides good noise immunity. A computer-based arrhythmia detection system of cardiac states is very useful in diagnostics and disease management. In this work, we studied the identification of the HRV signals using features derived from HOS. These features were fed to the support vector machine (SVM) for classification. Our proposed system can classify the normal and other four classes of arrhythmia with an average accuracy of more than 85%.

Ranking based clustering for social event detection

The problem of clustering a large document collection is not only challenged by the number of documents and the number of dimensions, but it is also affected by the number and sizes of the clusters. Traditional clustering methods fail to scale when they need to generate a large number of clusters. Furthermore, when the clusters size in the solution is heterogeneous, i.e. some of the clusters are large in size, the similarity measures tend to degrade. A ranking based clustering method is proposed to deal with these issues in the context of the Social Event Detection task. Ranking scores are used to select a small number of most relevant clusters in order to compare and place a document. Additionally,instead of conventional cluster centroids, cluster patches are proposed to represent clusters, that are hubs-like set of documents. Text, temporal, spatial and visual content information collected from the social event images is utilized in calculating similarity. Results show that these strategies allow us to have a balance between performance and accuracy of the clustering solution gained by the clustering method.

Lighting invariant urban street classification

In this paper we propose the hybrid use of illuminant invariant and RGB images to perform image classification of urban scenes despite challenging variation in lighting conditions. Coping with lighting change (and the shadows thereby invoked) is a non-negotiable requirement for long term autonomy using vision. One aspect of this is the ability to reliably classify scene components in the presence of marked and often sudden changes in lighting. This is the focus of this paper. Posed with the task of classifying all parts in a scene from a full colour image, we propose that lighting invariant transforms can reduce the variability of the scene, resulting in a more reliable classification. We leverage the ideas of “data transfer” for classification, beginning with full colour images for obtaining candidate scene-level matches using global image descriptors. This is commonly followed by superpixellevel matching with local features. However, we show that if the RGB images are subjected to an illuminant invariant transform before computing the superpixel-level features, classification is significantly more robust to scene illumination effects. The approach is evaluated using three datasets. The first being our own dataset and the second being the KITTI dataset using manually generated ground truth for quantitative analysis. We qualitatively evaluate the method on a third custom dataset over a 750m trajectory.

Turning event logs into process movies: animating what has really happened

Today’s information systems log vast amounts of data. These collections of data (implicitly) describe events (e.g. placing an order or taking a blood test) and, hence, provide information on the actual execution of business processes. The analysis of such data provides an excellent starting point for business process improvement. This is the realm of process mining, an area which has provided a repertoire of many analysis techniques. Despite the impressive capabilities of existing process mining algorithms, dealing with the abundance of data recorded by contemporary systems and devices remains a challenge. Of particular importance is the capability to guide the meaningful interpretation of “oceans of data” by process analysts. To this end, insights from the field of visual analytics can be leveraged. This article proposes an approach where process states are reconstructed from event logs and visualised in succession, leading to an animated history of a process. This approach is customisable in how a process state, partially defined through a collection of activity instances, is visualised: one can select a map and specify a projection of events on this map based on the properties of the events. This paper describes a comprehensive implementation of the proposal. It was realised using the open-source process mining framework ProM. Moreover, this paper also reports on an evaluation of the approach conducted with Suncorp, one of Australia’s largest insurance companies.

SAIVT-ADMRG @ MediaEval 2014 social event detection

This paper outlines the approach taken by the Speech, Audio, Image and Video Technologies laboratory, and the Applied Data Mining Research Group (SAIVT-ADMRG) in the 2014 MediaEval Social Event Detection (SED) task. We participated in the event based clustering subtask (subtask 1), and focused on investigating the incorporation of image features as another source of data to aid clustering. In particular, we developed a descriptor based around the use of super-pixel segmentation, that allows a low dimensional feature that incorporates both colour and texture information to be extracted and used within the popular bag-of-visual-words (BoVW) approach.

Classification of echolocation calls from 14 species of bat by Support Vector Machines and Ensembles of Neural Networks

Calls from 14 species of bat were classified to genus and species using discriminant function analysis (DFA), support vector machines (SVM) and ensembles of neural networks (ENN). Both SVMs and ENNs outperformed DFA for every species while ENNs (mean identification rate – 97%) consistently outperformed SVMs (mean identification rate – 87%). Correct classification rates produced by the ENNs varied from 91% to 100%; calls from six species were correctly identified with 100% accuracy. Calls from the five species of Myotis, a genus whose species are considered difficult to distinguish acoustically, had correct identification rates that varied from 91 – 100%. Five parameters were most important for classifying calls correctly while seven others contributed little to classification performance.

The contribution of major depression to the global burden of ischemic heart disease : a comparative risk assessment

Background Cardiovascular disease and mental health both hold enormous public health importance, both ranking highly in results of the recent Global Burden of Disease Study 2010 (GBD 2010). For the first time, the GBD 2010 has systematically and quantitatively assessed major depression as an independent risk factor for the development of ischemic heart disease (IHD) using comparative risk assessment methodology. Methods A pooled relative risk (RR) was calculated from studies identified through a systematic review with strict inclusion criteria designed to provide evidence of independent risk factor status. Accepted case definitions of depression include diagnosis by a clinician or by non-clinician raters adhering to Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases (ICD) classifications. We therefore refer to the exposure in this paper as major depression as opposed to the DSM-IV category of major depressive disorder (MDD). The population attributable fraction (PAF) was calculated using the pooled RR estimate. Attributable burden was calculated by multiplying the PAF by the underlying burden of IHD estimated as part of GBD 2010. Results The pooled relative risk of developing IHD in those with major depression was 1.56 (95% CI 1.30 to 1.87). Globally there were almost 4 million estimated IHD disability-adjusted life years (DALYs), which can be attributed to major depression in 2010; 3.5 million years of life lost and 250,000 years of life lived with a disability. These findings highlight a previously underestimated mortality component of the burden of major depression. As a proportion of overall IHD burden, 2.95% (95% CI 1.48 to 4.46%) of IHD DALYs were estimated to be attributable to MDD in 2010. Eastern Europe and North Africa/Middle East demonstrate the highest proportion with Asia Pacific, high income representing the lowest. Conclusions The present work comprises the most robust systematic review of its kind to date. The key finding that major depression may be responsible for approximately 3% of global IHD DALYs warrants assessment for depression in patients at high risk of developing IHD or at risk of a repeat IHD event.

Pollen image classification using the classifynder system: Algorithm comparison and a case study on New Zealand honey

We describe an investigation into how Massey University’s Pollen Classifynder can accelerate the understanding of pollen and its role in nature. The Classifynder is an imaging microscopy system that can locate, image and classify slide based pollen samples. Given the laboriousness of purely manual image acquisition and identification it is vital to exploit assistive technologies like the Classifynder to enable acquisition and analysis of pollen samples. It is also vital that we understand the strengths and limitations of automated systems so that they can be used (and improved) to compliment the strengths and weaknesses of human analysts to the greatest extent possible. This article reviews some of our experiences with the Classifynder system and our exploration of alternative classifier models to enhance both accuracy and interpretability. Our experiments in the pollen analysis problem domain have been based on samples from the Australian National University’s pollen reference collection (2,890 grains, 15 species) and images bundled with the Classifynder system (400 grains, 4 species). These samples have been represented using the Classifynder image feature set.We additionally work through a real world case study where we assess the ability of the system to determine the pollen make-up of samples of New Zealand honey. In addition to the Classifynder’s native neural network classifier, we have evaluated linear discriminant, support vector machine, decision tree and random forest classifiers on these data with encouraging results. Our hope is that our findings will help enhance the performance of future releases of the Classifynder and other systems for accelerating the acquisition and analysis of pollen samples.

A comparison of classification algorithms within the Classifynder pollen imaging system

We describe an investigation into how Massey University's Pollen Classifynder can accelerate the understanding of pollen and its role in nature. The Classifynder is an imaging microscopy system that can locate, image and classify slide based pollen samples. Given the laboriousness of purely manual image acquisition and identification it is vital to exploit assistive technologies like the Classifynder to enable acquisition and analysis of pollen samples. It is also vital that we understand the strengths and limitations of automated systems so that they can be used (and improved) to compliment the strengths and weaknesses of human analysts to the greatest extent possible. This article reviews some of our experiences with the Classifynder system and our exploration of alternative classifier models to enhance both accuracy and interpretability. Our experiments in the pollen analysis problem domain have been based on samples from the Australian National University's pollen reference collection (2890 grains, 15 species) and images bundled with the Classifynder system (400 grains, 4 species). These samples have been represented using the Classifynder image feature set. In addition to the Classifynder's native neural network classifier, we have evaluated linear discriminant, support vector machine, decision tree and random forest classifiers on these data with encouraging results. Our hope is that our findings will help enhance the performance of future releases of the Classifynder and other systems for accelerating the acquisition and analysis of pollen samples. © 2013 AIP Publishing LLC.

Classification of inventory data suit for managing Australian railway bridges

With a focus to optimising the life cycle performance of Australian Railway bridges, new bridge classification and environmental classification systems are proposed. The new bridge classification system is mainly to facilitate the implementation of novel Bridge Management System (BMS) which optimise the life cycle cost both at project level and network level while environment classification is mainly to improve accuracy of Remaining Service Potential (RSP) module of the proposed BMS. In fact, limited capacity of the existing BMS to trigger the maintenance intervention point is an indirect result of inadequacies of the existing bridge and environmental classification systems. The proposed bridge classification system permits to identify the intervention points based on percentage deterioration of individual elements and maintenance cost, while allowing performance based rating technique to implement for maintenance optimisation and prioritisation. Simultaneously, the proposed environment classification system will enhance the accuracy of prediction of deterioration of steel components.

Mining resource profiles from event logs

Human resources are often responsible for the execution of business processes. In order to evaluate resource performance and identify best practices as well as opportunities for improvement, managers need objective information about resource behaviours. Companies often use information systems to support their processes and these systems record information about process execution in event logs. We present a framework for analysing and evaluating resource behaviour through mining such event logs. The framework provides a method for extracting descriptive information about resource skills, utilisation, preferences, productivity and collaboration patterns; a method for analysing relationships between different resource behaviours and outcomes; and a method for evaluating the overall resource productivity, tracking its changes over time and comparing it with the productivity of other resources. To demonstrate the applicability of our framework we apply it to analyse behaviours of employees in an Australian company and evaluate its usefulness by a survey among managers in industry.