New computer technologies are applied to the classical material engineering two-body abrasive wear process. The computer simulation provides an interactive and visual representation of the wear process. The influence of grit size, grit tip radius and load (at constant workpiece hardness and tool path) on the wear rate, wear coefficient and wear surface topography is predicted. The simulation implements microcutting and microploughing with material displacement to the sides of the groove. The validation of the simulation is demonstrated by comparing with the previous modelling literature and with experiments. / Thesis (M.Sc.)-University of Natal,Durban, 2002.

This thesis examines the effectiveness of a specially constructed computer based support environment for the teaching of computer programming to novice programmers. In order to achieve this, the following distinct activities were pursued. Firstly, an in-depth investigation of programming misconceptions and techniques used for overcoming them was carried out. Secondly, the educational principles gained from this investigation were used to design and implement a computer based environment to support novice programmers learning the Pascal language. Finally, several statistical methods were used to compare students who made use of the support environment to those who did not and the results are discussed. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1996.

Genetic Algorithms (GAs) can be easily applied to many different problems since they make few assumptions about the application domain and perform relatively well. They can also be modified with some success for handling a particular problem. The travelling salesperson problem (TSP) is a famous NP-hard problem in combinatorial optimization. As a result it has no known polynomial time solution. The aim of this dissertation will be to investigate the application of a number of GAs to the TSP. These results will be compared with those of traditional solutions to the TSP and with the results of other applications of the GA to the TSP. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1996.

The school timetabling problem is a common optimization problem faced by many primary and secondary schools. Each school has its own set of requirements and constraints that are dependent on various factors such as the number of resources available and rules specified by the department of education for that country. There are two objectives in this study. In previous studies, genetic algorithms have only been used to solve a single type of school timetabling problem. The first objective of this study is to test the effectiveness of a genetic algorithm approach in solving more than one type of school timetabling problem. The second objective is to evaluate a genetic algorithm that uses an indirect representation (IGA) when solving the school timetabling problem. This IGA approach is then compared to the performance of a genetic algorithm that uses a direct representation (DGA). This approach has been covered in other domains such as job shop scheduling but has not been covered for the school timetabling problem. Both the DGA and IGA were tested on five school timetabling problems. Both the algorithms were initially developed based on findings in the literature. They were then improved iteratively based on their performance when tested on the problems. The processes of the genetic algorithms that were improved were the method of initial population creation, the selection methods and the genetic operators used. Both the DGA and the IGA were found to produce timetables that were competitive and in some cases superior to that of other methods such as simulated annealing and tabu search. It was found that different processes (i.e. the method of initial population creation, selection methods and genetic operators) were needed for each problem in order to produce the best results. When comparing the performance of the two approaches, the IGA outperformed the DGA for all of the tested school timetabling problems. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2013.

Event understanding systems, responsible for automatically generating human relatable event descriptions from video sequences, is an open problem in computer vision research that has many applications in the sports domain, such as indexing and retrieval systems for sports video. Background modelling and shot classification of broadcast video are important steps in event understanding in video sequences. Shot classification seeks to identify shots, i.e. the labelling of continuous frame sequences captured by a single camera action such as long shot, close-up and audience shot, while background modelling seeks to classify pixels in an image as foreground/background. Many features used for shot classification are built upon the background model therefore background modelling is an essential part of shot classification. This dissertation reports on an investigation into techniques and procedures for background modelling and classification of shots in broadcast soccer videos. Broadcast video refers to video which would typically be viewed by a person at home on their television set and imposes constraints that are often not considered in many approaches to event detection. In this work we analyse the performances of two background modelling techniques appropriate for broadcast video, the colour distance model and Gaussian mixture model. The performance of the background models depends on correctly set parameters. Some techniques offer better updating schemes and thus adapt better to the changing conditions of a game, some are shown to be more robust to changes in broadcast technique and are therefore of greater value in shot classification. Our results show the colour distance model slightly outperformed the Gaussian mixture model with both techniques performing similar to those found in literature. Many features useful for shot classification are proposed in the literature. This dissertation identifies these features and presents a detailed analysis and comparison of various features appropriate for shot classification in broadcast soccer video. Once a feature set is established, a classifier is required to determine a shot class based on the extracted features. We establish the best use of the feature set and decision tree parameters that result in the best performance and then use a combined feature set to train a neural network to classify shots. The combined feature set in conjunction with the neural network classifier proved effective in classifying shots and in some situations outperformed those techniques found in literature. / Thesis (M.Sc.)-University of KwaZulu-Natal, Durban, 2012.

In this thesis a number of mathematical models are investigated with the aid of the modelling package Mathematica. Some of the models are of a mechanical nature and some of the models are laboratories that have been constructed for the purpose of assisting researchers in a particular field. In the early sections of the thesis mechanical models are investigated. After the equations of motion for the model have been presented, Mathematica is employed to generate solutions which are then used to drive animations of the model. The frames of the animations are graphical snapshots of the model in motion. Mathematica proves to be an ideal tool for this type of modelling since it combines algebraic, numeric and graphics capabilities on one platform. In the later sections of this thesis, Mathematica laboratories are created for investigating models in two different fields. The first laboratory is a collection of routines for performing Phase-Plane analysis of planar autonomous systems of ordinary differential equations. A model of a mathematical concept called a bifurcation is investigated and an animation of this mathematical event is produced. The second laboratory is intended to help researchers in the tomography field. A standard filtered back-projection algorithm for reconstructing images from their projections is implemented. In the final section of the thesis an indication of how the tomography laboratory could be used is presented. Wavelet theory is used to construct a new filter that could be used in filtered back-projection tomography. / Thesis (Ph.D.)-University of Natal, Durban, 1994.

This study considers the meteor-burst communication (MBC) environment at three levels. At the lowest level, the trails themselves are studied and analysed. Then individual links are studied in order to determine the data throughput and wait time that might be expected at various data rates. Finally, at the top level, MBC networks are studied in order to provide information on the effects of routing strategies, topologies, and connectivity in such networks. A significant amount of theoretical work has been done in the classification of meteor trails, and the analysis of the throughput potential of the channel. At the same time the issues of wait time on MBC links, and MBC network strategies, have been largely ignored. The work presented here is based on data captured on actual monitoring links, and is intended to provide both an observational comparison to theoretical predictions in the well-researched areas, and a source of base information for the others. Chapter 1 of this thesis gives an overview of the field of meteor-burst communications. Prior work in the field is discussed, as are the advantages and disadvantages of the channel, and current application areas. Chapter 2 describes work done on the classification of observed meteor trails into distinctive 'families'. The rule-based system designed for this task is discussed as well as the eventual classification schema produced, which is far more comprehensive and consistent than previously proposed schemas. Chapter 3 deals with the throughput potential of the channel, based on the observed trails. A comparison to predicted results, both as regards fixed and adaptive data-rates, is made with some notable differences between predicted v results and observed results highlighted. The trail families with the largest contribution to the throughput capacity of the channel are identified. Chapter 4 deals with wait time in meteor-burst communications. The data rates at which wait time is minimised in the links used are found, and compared to the rates at which throughput was optimised. These are found to be very different, as indeed are the contributions of the various trail families at these rates. Chapter 5 describes a software system designed to analyse the effect of routing strategies in MBC networks, and presents initial results derived from this system. Certain features of the channel, in particular its sporadic nature, are shown to have significant effects on network performance. Chapter 6 continues the presentation of network results, specifically concentrating on the effect of topologies and connectivity within MBC networks. Chapter 7 concludes the thesis, highlighting suggested areas for further research as well as summarising the more important results presented. / Thesis (Ph.D.)-University of Natal, Durban, 1991.

Advances in sensor technology and space science have resulted in the availability of vast quantities of high quality earth observation data. This data can be used for monitoring the earth and to enhance our understanding of natural processes. Sensor Web researchers are working on constructing a worldwide computing infrastructure that enables dynamic sharing and analysis of complex heterogeneous earth observation data sets. Key challenges that are currently being investigated include data integration; service discovery, reuse and composition; semantic interoperability; and system dynamism. Two emerging technologies that have shown promise in dealing with these challenges are ontologies and software agents. This research investigates how these technologies can be integrated into an Ontology Driven Multi-Agent System (ODMAS) for the Sensor Web. The research proposes an ODMAS framework and an implemented middleware platform, i.e. the Sensor Web Agent Platform (SWAP). SWAP deals with ontology construction, ontology use, and agent based design, implementation and deployment. It provides a semantic infrastructure, an abstract architecture, an internal agent architecture and a Multi-Agent System (MAS) middleware platform. Distinguishing features include: the incorporation of Bayesian Networks to represent and reason about uncertain knowledge; ontologies to describe system entities such as agent services, interaction protocols and agent workflows; and a flexible adapter based MAS platform that facilitates agent development, execution and deployment. SWAP aims to guide and ease the design, development and deployment of dynamic alerting and monitoring applications. The efficacy of SWAP is demonstrated by two satellite image processing applications, viz. wildfire detection and monitoring informal settlement. This approach can provide significant benefits to a wide range of Sensor Web users. These include: developers for deploying agents and agent based applications; end users for accessing, managing and visualising information provided by real time monitoring applications, and scientists who can use the Sensor Web as a scientific computing platform to facilitate knowledge sharing and discovery. An Ontology Driven Multi-Agent Sensor Web has the potential to forever change the way in which geospatial data and knowledge is accessed and used. This research describes this far reaching vision, identifies key challenges and provides a first step towards the vision. / Thesis (Ph.D.)-University of KwaZulu-Natal, 2009.

Spoken language is used by people to communicate naturally with one another. A simplistic view of the communication process is as follows. Person A wishes to communicate an idea to person B. The idea, initiated in the mind/brain of person A is encoded into speech signals by means of the person A's speech production mechanism, the vocal apparata in the vocal tract. Various kinds of noise may interfere with the speech signals as they travel to person B. The resulting signal is captured by person B's speech receiving mechanism, the ear. It is then analysed and decoded into a meaningful message by the brain of person B. This thesis concerns itself with the investigation of and attempt to automate the receiving and decoding of English sentences using a machine - that is to perform the task of person B in the above scenario using a computer. The aim is not only to produce a sequence of phonetic sounds, but to look at the problems of building in the 'mind of the machine', a picture of the meanings, intentions, absurdities and realities of the spoken message. The various models, algorithms and techniques of speech recognition and speech understanding systems are examined. Speech signals are captured and digitised by hardware. The digital samples are analysed and the important distinguishing features of all speech sounds are identified. These are then used to classify speech sounds in subsequent spoken words. The way speech sounds are joined together to form syllables and words introduces difficult problems to the automatic recognition process. Speech sounds are blurred, overlapped or left out due to the effects of coarticulation. Finally, natural language processing issues, such as the importance of syntax (the structure) and semantics (the meaning) of sentences, are studied. A system to control and unite all the above processing is considered. The blackboard expert system model of the widely reported HEARSAY-II speech recognition system is reviewed as the system with the best potential for the above tasks. / Thesis (M.Sc.)-University of KwaZulu-Natal, 1992.

Transport logistics systems typically evolve as networks over time, which may result in system rigidity and cause changes to become expensive and time consuming. In this study a logistics model, named TranSwarm, was developed to simulate sugarcane harvesting, transport and mill-yard activities for a mill supply area. The aim was to simulate produce flow, and allow individual working entities to make decisions, driven by rules and protocols, based on their micro-environments. Noodsberg mill was selected as a case study because of low current levels of synchronization. Growers were assumed to operate independent harvesting and transport systems causing inconsistent convergences at the mill. This diverse and fragmented system provided a suitable environment to construct a model that would consider interactions between individual growers and their respective transport systems. Ideally, by assessing the micro-decisions of individuals and how they influence the larger holistic supply chain, TranSwarm quantifies the impacts of different types of transport practices, such as staggering shift changes, transport scheduling, core sampling and consortium-based logistics. TranSwarm is visual, mechanistic and represents key entities, such as roads, farm groupings and the mill. The system uses discrete events to create a dynamic and stochastic environment from which observations and conclusions can be drawn. This approach potentially allows stakeholders to identify key components and interactions that may jeopardize overall efficiency and to use the system to test new working protocols and logistics rules for improving the supply chain. / Thesis (M.Sc.)-University of KwaZulu-Natal, 2008.