An evaluation of crowd counting methods, features and regression models

Existing crowd counting algorithms rely on holistic, local or histogram based features to capture crowd properties. Regression is then employed to estimate the crowd size. Insufficient testing across multiple datasets has made it difficult to compare and contrast different methodologies. This paper presents an evaluation across multiple datasets to compare holistic, local and histogram based methods, and to compare various image features and regression models. A K-fold cross validation protocol is followed to evaluate the performance across five public datasets: UCSD, PETS 2009, Fudan, Mall and Grand Central datasets. Image features are categorised into five types: size, shape, edges, keypoints and textures. The regression models evaluated are: Gaussian process regression (GPR), linear regression, K nearest neighbours (KNN) and neural networks (NN). The results demonstrate that local features outperform equivalent holistic and histogram based features; optimal performance is observed using all image features except for textures; and that GPR outperforms linear, KNN and NN regression

From policy borrowing to implementation : an illuminative evaluation of learning and teaching in higher education in Australia (2002 to 2008)

This study documents and theorises the consequences of the 2003 Australian Government Reform Package focussed on learning and teaching in Higher Education during the period 2002 to 2008. This is achieved through the perspective of program evaluation and the methodology of illuminative evaluation. The findings suggest that the three national initiatives of that time, Learning and Teaching Performance Fund (LTPF), Australian Learning and Teaching Council (ALTC), and Australian Universities Quality Agency (AUQA), were successful in repositioning learning and teaching as a core activity in universities. However, there were unintended consequences brought about by international policy borrowing, when the short-lived nature of LTPF suggests a legacy of quality compliance rather than one of quality enrichment.

The development and evaluation of a medical imaging training immersive environment

Introduction A novel realistic 3D virtual reality (VR) application has been developed to allow medical imaging students at Queensland University of Technology to practice radiographic techniques independently outside the usual radiography laboratory. Methods A flexible agile development methodology was used to create the software rapidly and effectively. A 3D gaming environment and realistic models were used to engender presence in the software while tutor-determined gold standards enabled students to compare their performance and learn in a problem-based learning pedagogy. Results Students reported high levels of satisfaction and perceived value and the software enabled up to 40 concurrent users to prepare for clinical practice. Student feedback also indicated that they found 3D to be of limited value in the desktop version compared to the usual 2D approach. A randomised comparison between groups receiving software-based and traditional practice measured performance in a formative role play with real equipment. The results of this work indicated superior performance with the equipment for the VR trained students (P = 0.0366) and confirmed the value of VR for enhancing 3D equipment-based problem-solving skills. Conclusions Students practising projection techniques virtually performed better at role play assessments than students practising in a traditional radiography laboratory only. The application particularly helped with 3D equipment configuration, suggesting that teaching 3D problem solving is an ideal use of such medical equipment simulators. Ongoing development work aims to establish the role of VR software in preparing students for clinical practice with a range of medical imaging equipment.

Heat strain evaluation of overt and covert body armour in a hot and humid environment

The aim of this study was to elucidate the thermophysiological effects of wearing lightweight non-military overt and covert personal body armour (PBA) in a hot and humid environment. Eight healthy males walked on a treadmill for 120 min at 22% of their heart rate reserve in a climate chamber simulating 31 °C (60%RH) wearing either no armour (control), overt or covert PBA in addition to a security guard uniform, in a randomised controlled crossover design. No significant difference between conditions at the end of each trial was observed in core temperature, heart rate or skin temperature (P > 0.05). Covert PBA produced a significantly greater amount of body mass change (−1.81 ± 0.44%) compared to control (−1.07 ± 0.38%, P = 0.009) and overt conditions (−1.27 ± 0.44%, P = 0.025). Although a greater change in body mass was observed after the covert PBA trial; based on the physiological outcome measures recorded, the heat strain encountered while wearing lightweight, non-military overt or covert PBA was negligible compared to no PBA. Practitioner summary The wearing of bullet proof vests or body armour is a requirement of personnel engaged in a wide range of occupations including police, security, customs and even journalists in theatres of war. This randomised controlled crossover study is the first to examine the thermophysiological effects of wearing lightweight non-military overt and covert personal body armour (PBA) in a hot and humid environment. We conclude that the heat strain encountered while wearing both overt and covert lightweight, non-military PBA was negligible compared to no PBA.