Representation of facial expression categories in continuous arousal-valence space: Feature and correlation

Representation of facial expressions using continuous dimensions has shown to be inherently more expressive and psychologically meaningful than using categorized emotions, and thus has gained increasing attention over recent years. Many sub-problems have arisen in this new field that remain only partially understood. A comparison of the regression performance of different texture and geometric features and investigation of the correlations between continuous dimensional axes and basic categorized emotions are two of these. This paper presents empirical studies addressing these problems, and it reports results from an evaluation of different methods for detecting spontaneous facial expressions within the arousal-valence dimensional space (AV). The evaluation compares the performance of texture features (SIFT, Gabor, LBP) against geometric features (FAP-based distances), and the fusion of the two. It also compares the prediction of arousal and valence, obtained using the best fusion method, to the corresponding ground truths. Spatial distribution, shift, similarity, and correlation are considered for the six basic categorized emotions (i.e. anger, disgust, fear, happiness, sadness, surprise). Using the NVIE database, results show that the fusion of LBP and FAP features performs the best. The results from the NVIE and FEEDTUM databases reveal novel findings about the correlations of arousal and valence dimensions to each of six basic emotion categories.

Towards pose-robust face recognition on video

This thesis investigates face recognition in video under the presence of large pose variations. It proposes a solution that performs simultaneous detection of facial landmarks and head poses across large pose variations, employs discriminative modelling of feature distributions of faces with varying poses, and applies fusion of multiple classifiers to pose-mismatch recognition. Experiments on several benchmark datasets have demonstrated that improved performance is achieved using the proposed solution.

How small business advisory program delivery methods (collective learning, tailored, and practice-based approaches) affect learning and innovation

Past studies relate small business advisory program effectiveness to advisory characteristics such as advisory intensity and scope. We contribute to existing literature by seeking to identify the impact of different advisory program methods of delivery on learning and subsequent firm innovation behavior. Our research is based on a survey of 257 Australian firms completing small business advisory programs in the three years preceding the research. We explore the range of small business advisory program delivery methods in which our surveyed firms participated and, with reference to the literature on organizational learning and innovation, we analyze predictors of firms' learning ability and innovativeness based on the identified delivery methods. First, we found that business advisory programs that involved high levels of collective learning and tailored approaches enhanced firms' perceptions of their learning of critical skills or capabilities. We also found that small business advisory programs that were delivered by using practice-based approaches enhanced firms' subsequent organizational innovation. We verified this finding by testing whether firms that have participated in small business advisory services subsequently demonstrate improved behavior in terms of organizational innovativeness, when compared with matched firms that have not participated in an advisory program.

Methods for extracting genomic DNA from whole blood samples : current perspectives

Deoxyribonucleic acid (DNA) extraction has considerably evolved since it was initially performed back in 1869. It is the first step required for many of the available downstream applications used in the field of molecular biology. Whole blood samples are one of the main sources used to obtain DNA, and there are many different protocols available to perform nucleic acid extraction on such samples. These methods vary from very basic manual protocols to more sophisticated methods included in automated DNA extraction protocols. Based on the wide range of available options, it would be ideal to determine the ones that perform best in terms of cost-effectiveness and time efficiency. We have reviewed DNA extraction history and the most commonly used methods for DNA extraction from whole blood samples, highlighting their individual advantages and disadvantages. We also searched current scientific literature to find studies comparing different nucleic acid extraction methods, to determine the best available choice. Based on our research, we have determined that there is not enough scientific evidence to support one particular DNA extraction method from whole blood samples. Choosing a suitable method is still a process that requires consideration of many different factors, and more research is needed to validate choices made at facilities around the world.

Thin films and nanostructures of niobium pentoxide : fundamental properties, synthesis methods and applications

As one of the transition metal oxides, niobium pentoxide (Nb2O5) offers a broad variety of properties that make it a potentially useful and highly applicable material in many different areas. In comparison to many other transition metal oxides, Nb2O5 has received relatively little attention, which presents a significant opportunity for future investigations aimed at fundamentally understanding this material and finding new and interesting applications for it. In this article, a general overview of Nb2O5 is presented which focuses on its fundamental properties, synthesis methods and recent applications, along with a discussion on future research directions relevant to this material.

Experimental investigations of improved dielectric condition monitoring methods on transformer oil-paper insulation system

Frequency Domain Spectroscopy (FDS) is successfully being used to assess the insulation condition of oil filled power transformers. However, it has to date only been implemented on de-energized transformers, which requires the transformers to be shut down for an extended period which can result in significant costs. To solve this issue, a method of implementing FDS under energized condition is proposed here. A chirp excitation waveform is used to replace the conventional sinusoidal waveform to reduce the measurement time in this method. Investigation of the dielectric response under the influence of a high voltage stress at power frequency is reported based on experimental results. To further understand the insulation ageing process, the geometric capacitance effect is removed to enhance the detection of the ageing signature. This enhancement enables the imaginary part of admittance to be used as a new indicator to assess the ageing status of the insulation.

Machine learning for activity recognition : hip versus wrist data

Problem addressed Wrist-worn accelerometers are associated with greater compliance. However, validated algorithms for predicting activity type from wrist-worn accelerometer data are lacking. This study compared the activity recognition rates of an activity classifier trained on acceleration signal collected on the wrist and hip. Methodology 52 children and adolescents (mean age 13.7 +/- 3.1 year) completed 12 activity trials that were categorized into 7 activity classes: lying down, sitting, standing, walking, running, basketball, and dancing. During each trial, participants wore an ActiGraph GT3X+ tri-axial accelerometer on the right hip and the non-dominant wrist. Features were extracted from 10-s windows and inputted into a regularized logistic regression model using R (Glmnet + L1). Results Classification accuracy for the hip and wrist was 91.0% +/- 3.1% and 88.4% +/- 3.0%, respectively. The hip model exhibited excellent classification accuracy for sitting (91.3%), standing (95.8%), walking (95.8%), and running (96.8%); acceptable classification accuracy for lying down (88.3%) and basketball (81.9%); and modest accuracy for dance (64.1%). The wrist model exhibited excellent classification accuracy for sitting (93.0%), standing (91.7%), and walking (95.8%); acceptable classification accuracy for basketball (86.0%); and modest accuracy for running (78.8%), lying down (74.6%) and dance (69.4%). Potential Impact Both the hip and wrist algorithms achieved acceptable classification accuracy, allowing researchers to use either placement for activity recognition.

Prediction of activity type in preschool children using machine learning techniques

Objectives Recent research has shown that machine learning techniques can accurately predict activity classes from accelerometer data in adolescents and adults. The purpose of this study is to develop and test machine learning models for predicting activity type in preschool-aged children. Design Participants completed 12 standardised activity trials (TV, reading, tablet game, quiet play, art, treasure hunt, cleaning up, active game, obstacle course, bicycle riding) over two laboratory visits. Methods Eleven children aged 3–6 years (mean age = 4.8 ± 0.87; 55% girls) completed the activity trials while wearing an ActiGraph GT3X+ accelerometer on the right hip. Activities were categorised into five activity classes: sedentary activities, light activities, moderate to vigorous activities, walking, and running. A standard feed-forward Artificial Neural Network and a Deep Learning Ensemble Network were trained on features in the accelerometer data used in previous investigations (10th, 25th, 50th, 75th and 90th percentiles and the lag-one autocorrelation). Results Overall recognition accuracy for the standard feed forward Artificial Neural Network was 69.7%. Recognition accuracy for sedentary activities, light activities and games, moderate-to-vigorous activities, walking, and running was 82%, 79%, 64%, 36% and 46%, respectively. In comparison, overall recognition accuracy for the Deep Learning Ensemble Network was 82.6%. For sedentary activities, light activities and games, moderate-to-vigorous activities, walking, and running recognition accuracy was 84%, 91%, 79%, 73% and 73%, respectively. Conclusions Ensemble machine learning approaches such as Deep Learning Ensemble Network can accurately predict activity type from accelerometer data in preschool children.

Differential effects of refractive blur on day and night-time driving performance

Purpose To investigate the effect of different levels of refractive blur on real-world driving performance measured under day and nighttime conditions. Methods Participants included 12 visually normal, young adults (mean age = 25.8 ± 5.2 years) who drove an instrumented research vehicle around a 4 km closed road circuit with three different levels of binocular spherical refractive blur (+0.50 diopter sphere [DS], +1.00 DS, +2.00 DS) compared with a baseline condition. The subjects wore optimal spherocylinder correction and the additional blur lenses were mounted in modified full-field goggles; the order of testing of the blur conditions was randomized. Driving performance was assessed in two different sessions under day and nighttime conditions and included measures of road signs recognized, hazard detection and avoidance, gap detection, lane-keeping, sign recognition distance, speed, and time to complete the course. Results Refractive blur and time of day had significant effects on driving performance (P < 0.05), where increasing blur and nighttime driving reduced performance on all driving tasks except gap judgment and lane keeping. There was also a significant interaction between blur and time of day (P < 0.05), such that the effects of blur were exacerbated under nighttime driving conditions; performance differences were evident even for +0.50 DS blur relative to baseline for some measures. Conclusions The effects of blur were greatest under nighttime conditions, even for levels of binocular refractive blur as low as +0.50 DS. These results emphasize the importance of accurate and up-to-date refractive correction of even low levels of refractive error when driving at night.

Seeing pedestrians at night : effect of driver age and visual abilities

Purpose To quantify the effects of driver age on night-time pedestrian conspicuity, and to determine whether individual differences in visual performance can predict drivers' ability to recognise pedestrians at night. Methods Participants were 32 visually normal drivers (20 younger: M = 24.4 years ± 6.4 years; 12 older: M = 72.0 years ± 5.0 years). Visual performance was measured in a laboratory-based testing session including visual acuity, contrast sensitivity, motion sensitivity and the useful field of view. Night-time pedestrian recognition distances were recorded while participants drove an instrumented vehicle along a closed road course at night; to increase the workload of drivers, auditory and visual distracter tasks were presented for some of the laps. Pedestrians walked in place, sideways to the oncoming vehicles, and wore either a standard high visibility reflective vest or reflective tape positioned on the movable joints (biological motion). Results Driver age and pedestrian clothing significantly (p < 0.05) affected the distance at which the drivers first responded to the pedestrians. Older drivers recognised pedestrians at approximately half the distance of the younger drivers and pedestrians were recognised more often and at longer distances when they wore a biological motion reflective clothing configuration than when they wore a reflective vest. Motion sensitivity was an independent predictor of pedestrian recognition distance, even when controlling for driver age. Conclusions The night-time pedestrian recognition capacity of older drivers was significantly worse than that of younger drivers. The distance at which drivers first recognised pedestrians at night was best predicted by a test of motion sensitivity.

A new approach to spatial data interpolation using higher-order statistics

Interpolation techniques for spatial data have been applied frequently in various fields of geosciences. Although most conventional interpolation methods assume that it is sufficient to use first- and second-order statistics to characterize random fields, researchers have now realized that these methods cannot always provide reliable interpolation results, since geological and environmental phenomena tend to be very complex, presenting non-Gaussian distribution and/or non-linear inter-variable relationship. This paper proposes a new approach to the interpolation of spatial data, which can be applied with great flexibility. Suitable cross-variable higher-order spatial statistics are developed to measure the spatial relationship between the random variable at an unsampled location and those in its neighbourhood. Given the computed cross-variable higher-order spatial statistics, the conditional probability density function (CPDF) is approximated via polynomial expansions, which is then utilized to determine the interpolated value at the unsampled location as an expectation. In addition, the uncertainty associated with the interpolation is quantified by constructing prediction intervals of interpolated values. The proposed method is applied to a mineral deposit dataset, and the results demonstrate that it outperforms kriging methods in uncertainty quantification. The introduction of the cross-variable higher-order spatial statistics noticeably improves the quality of the interpolation since it enriches the information that can be extracted from the observed data, and this benefit is substantial when working with data that are sparse or have non-trivial dependence structures.

Using qualitative mixed methods to study small health care organizations while maximising trustworthiness and authenticity

Background The primary health care sector delivers the majority of health care in western countries through small, community-based organizations. However, research into these healthcare organizations is limited by the time constraints and pressure facing them, and the concern by staff that research is peripheral to their work. We developed Q-RARA—Qualitative Rapid Appraisal, Rigorous Analysis—to study small, primary health care organizations in a way that is efficient, acceptable to participants and methodologically rigorous. Methods Q-RARA comprises a site visit, semi-structured interviews, structured and unstructured observations, photographs, floor plans, and social scanning data. Data were collected over the course of one day per site and the qualitative analysis was integrated and iterative. Results We found Q-RARA to be acceptable to participants and effective in collecting data on organizational function in multiple sites without disrupting the practice, while maintaining a balance between speed and trustworthiness. Conclusions The Q-RARA approach is capable of providing a richly textured, rigorous understanding of the processes of the primary care practice while also allowing researchers to develop an organizational perspective. For these reasons the approach is recommended for use in small-scale organizations both within and outside the primary health care sector.

Rethinking the value of simulation methods in the information systems research field : a call for reconstructing contribution for a broader audience

The impact of simulation methods for social research in the Information Systems (IS) research field remains low. A concern is our field is inadequately leveraging the unique strengths of simulation methods. Although this low impact is frequently attributed to methodological complexity, we offer an alternative explanation – the poor construction of research value. We argue a more intuitive value construction, better connected to the knowledge base, will facilitate increased value and broader appreciation. Meta-analysis of studies published in IS journals over the last decade evidences the low impact. To facilitate value construction, we synthesize four common types of simulation research contribution: Analyzer, Tester, Descriptor, and Theorizer. To illustrate, we employ the proposed typology to describe how each type of value is structured in simulation research and connect each type to instances from IS literature, thereby making these value types and their construction visible and readily accessible to the general IS community.

Visual place recognition for persistent robot navigation in changing environments

This thesis demonstrates that robots can learn about how the world changes, and can use this information to recognise where they are, even when the appearance of the environment has changed a great deal. The ability to localise in highly dynamic environments using vision only is a key tool for achieving long-term, autonomous navigation in unstructured outdoor environments. The proposed learning algorithms are designed to be unsupervised, and can be generated by the robot online in response to its observations of the world, without requiring information from a human operator or other external source.