Electrical and Electronic Engineering - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 125
-
ItemImplementation of marker training exercises to improve marking reliability and consistency(Australasian Association for Engineering Education, 2018)CONTEXT: One of the challenges present in teaching a large engineering subject is that of achieving marking consistency of assessments across multiple markers. Several measures of standardising markers exist, such as calibrated review, and are commonly used in the humanities, particularly for assessments that could be prone to a wide variation in marks such as essays. The application of such methods, in an engineering context, is somewhat less documented but of particular importance in the case of reflective writing. This study contrasts the implementation of several different methods of using marker training exercises prior to the actual assessment marking and provides an analysis of the results in order to minimise the effect of multiple marker irregularities and to provide effective high-quality formative feedback on a piece of reflective writing. PURPOSE: This paper presents several different methods of marker training exercises, run prior to the actual assessment marking, and provides analyses to determine the effect of each in terms of minimising marking inconsistency among multiple markers on a piece of reflective writing. APPROACH: In all three marker training exercises, markers are given samples of a piece of reflective writing, of differing quality, along with a rubric outlining the marking criteria for the piece of writing and exemplars for indicative marking standards. Each of the methods employed differ in how the reference standard was set and how feedback was delivered to the markers. Statistics comparing the marking results across markers from before the introduction of the training exercises and between each of the three training methods were analysed to investigate marking reliability and consistency. RESULTS: A significant reduction in the spread of the marker means has been achieved through the introduction of the marker training, indicating an improvement in consistency. Some differences in results between the alternative methods employed has also been observed. CONCLUSIONS: Marking consistency can be improved with the introduction of a marker training exercise prior to the actual assessment marking. Different methods of implementing the marking training exercise, and how the feedback is provided, can have an effect of the amount of improvement in terms of consistency and reliability.
-
ItemProbabilistic detection of crowd events on riemannian manifolds(IEEE, 2015-01-12)Event detection in crowded scenarios becomes complex due to articulated human movements, occlusions and complexities involved in tracking individual humans. In this work, we focus on crowd event (activity) detection and classification. We focus on active crowd (continuously moving crowd) events. First, event primitives such as motion, action, activity and behaviour are defined. Furthermore, a distinction is made among event detection, action recognition and abnormal event detection. Further, event detection and classification are defined on Riemannian Manifolds that yields six different probabilities of the event occurring. Using a new probabilistic approach, an automated event detection algorithm is proposed that temporally segments the event using a novel framework. The results indicate that the proposed approach delivers superior performance in selected cases and similar results in other cases, whilst the detection model delay allows operation in near real-time. The Performance Evaluation of Tracking and Surveillance (PETS) 2009 dataset was used for evaluation. Existing crowd event detection approaches used supervised approach, whereas we eschew semi-supervised approach.
-
ItemDetection of anomalous crowd behaviour using hyperspherical clustering(IEEE, 2015-01-12)Analysis of crowd behaviour in public places is an indispensable tool for video surveillance. Automated detection of anomalous crowd behaviour is a critical problem with the increase in human population. Anomalous events may include a person loitering about a place for unusual amounts of time; people running and causing panic; the size of a group of people growing over time etc. In this work, to detect anomalous events and objects, two types of feature coding has been proposed: spatial features and spatio-temporal features. Spatial features comprises of contrast, correlation, energy and homogeneity, which are derived from Gray Level Co-occurrence Matrix (GLCM). Spatio-temporal feature includes the time spent by an object at different locations in the scene. Hyperspherical clustering has been employed to detect the anomalies. Spatial features revealed the anomalous frames by using contrast and homogeneity measures. Loitering behaviour of the people were detected as anomalous objects using the spatio-temporal coding.
-
ItemClassification of Convulsive Psychogenic Non-epileptic Seizures Using Histogram of Oriented Motion of Accelerometry Signals(IEEE, 2015)A seizure is caused due to sudden surge of electrical activity within the brain. There is another class of seizures called psychogenic non-epileptic seizure (PNES) that mimics epilepsy, but is caused due to underlying psychology. The diagnosis of PNES is done using video-electroencephalography monitoring (VEM), which is a resource intensive process. Recently, accelerometers have been shown to be effective in classification of epileptic and non-epileptic seizures. In this work, we propose a novel feature called histogram of oriented motion (HOOM) extracted from accelerometer signals for classification of convulsive PNES. An automated algorithm based on HOOM is proposed. The algorithm showed a high sensitivity of (93.33%) and an overall accuracy of (80%) in classifying convulsive PNES.
-
ItemAn Improved Approach to Crowd Event Detection by Reducing Data Dimensions(SPRINGER-VERLAG BERLIN, 2016)Crowd monitoring is a critical application in video surveillance. Crowd events such as running, walking, merging, splitting, dispersion, and evacuation inform crowd management about the behavior of groups of people. For an effective crowd management, detection of crowd events provides an early sign of the behavior of the people. However, crowd event detection using videos is a highly challenging task because of several challenges such as non-rigid human body motions, occlusions, unavailability of distinguishing features due to occlusions, unpredictability in people movements, and other. In addition, the video itself is a high-dimensional data and analyzing to detect events becomes further complicated. One way of tackling the huge volume of video data is to represent a video using low-dimensional equivalent. However, reducing the video data size needs to consider the complex data structure and events embedded in a video. To this extent, we focus on detection of crowd events using the Isometric Mapping (ISOMAP) and Support Vector Machine (SVM). The ISOMAP is used to construct the low-dimensional representation of the feature vectors, and then an SVM is used for training and classification. The proposed approach uses Haar wavelets to extract Gray Level Coefficient Matrix (GLCM). Later, the approach extracts four statistical features (contrast, correlating, energy, and homogeneity) at different levels of Haar wavelet decomposition. Experiment results suggest that the proposed approach is shown to perform better when compared with existing approaches.
-
ItemAnomalous Crowd Event Analysis Using Isometric Mapping(SPRINGER-VERLAG BERLIN, 2016)Anomalous event detection is one of the important applications in crowd monitoring. The detection of anomalous crowd events requires featurematrix to capture the spatio-temporal information to localize the events and detect the outliers. However, feature matrices often become computationally expensive with large number of features becomes critical for large-scale and real-time video analytics. In this work, we present a fast approach to detect anomalous crowd events and frames. First, to detect anomalous crowd events, the motion features are captured using the optical flow and a feature matrix of motion information is constructed and then subjected to nonlinear dimensionality reduction (NDR) using the Isometric Mapping (ISOMAP). Next, to detect anomalous crowd frames, the method uses four statistical features by dividing the frames into blocks and then calculating the statistical features for the blocks where objects were present. The main focus of this study is to understand the effect of large feature matrix size on detecting the anomalies with respect to computational time. Experiments were conducted on two datasets: (1) Performance Evaluation of Tracking and Surveillance (PETS) 2009 and (2) Melbourne Cricket Ground (MCG) 2011. Experiment results suggest that the ISOMAP NDR reduces the computation time significantly, more than ten times, to detect anomalous crowd events and frames. In addition, the experiment revealed that the ISOMAP provided an upper bound on the computational time.
-
ItemMoving Horizon Estimation for Linear Cascade Systems(IEEE, 2018-01-01)A structured approach to the problem of state estimation for cascaded linear sub-systems is studied in terms of minimizing a measure of the error relative to a model over a moving horizon of past system input and output observations. A quadratic programming formulation of this optimization problem is considered and two approaches are explored. One approach involves solving the Karush-Kuhn-Tucker conditions directly, and the other is based on the alternating direction method of multipliers. In both cases, the problem structure can be exploited to yield distributed computations in the following sense: Construction of the estimate for each sub-system component of the state involves information pertaining to the two immediate neighbours only. Numerical simulations based on model data from an automated irrigation channel are used to investigate and compare the computational burden of the two approaches.
-
ItemStructured moving horizon estimation for linear system chains(IEEE, 2019-06-01)Computational aspects of moving horizon state estimation are studied for a class of chain networks with bidirectional coupling in the linear state dynamics, and measured outputs. Moving horizon estimation involves solving a quadratic program to minimize the estimation error relative to a model over a fixed window of past input-output observations. By exploiting the spatial structure of a chain, two algorithms for solving this quadratic program are considered. Both algorithms can be distributed in the sense that the computations associated with each sub-system component of the state depend only on information associated with the immediate neighbours. The algorithms differ in the way that the linear Karush-Kuhn-Tucker conditions for optimality are solved. Computational and information dependency overheads are analyzed. Numerical results are presented for a 1-D mass-spring-damper chain.
-
ItemAnomalous Behavior Detection in Crowded Scenes Using Clustering and Spatio-Temporal Features(Springer, 2016)important problem in real-life applications. Detection of anomalous behaviors such as people standing statically and loitering around a place are the focus of this paper. In order to detect anomalous events and objects, ViBe was used for background modeling and object detection at first. Then, a Kalman filter and Hungarian cost algorithm were implemented for tracking and generating trajectories of people. Next, spatio-temporal features were extracted and represented. Finally, hyperspherical clustering was used for anomaly detection in an unsupervised manner. We investigate three different approaches to extracting and representing spatio-temporal features, and we demonstrate the effectiveness of our proposed feature representation on a standard benchmark dataset and a real-life video surveillance environment.
-
ItemNon-Protruding Hazard Detection for the Aged Vision-Impaired(IEEE, 2016)Usage of the traditional white cane by the elderly with vision impairment is inefficient as many are also reliant on ambulatory aids such as wheelchairs and walking frames. The fall occurrence when using ambulatory aids is higher, contributed by non-protruding hazards such as potholes and drop-offs. Currently available technology for blind navigation, predominantly based on proximity sensing, is not designed to detect non protruding hazards. We address this critical need by developing a new optical laser system that combines innovative approaches in optical laser projection, vision-sensing, pattern recognition, and machine learning. Here, we present an overview of the system, including a new feature descriptor termed Histogram of Intersections, and results from our proof-of-concept demonstration.