The purpose of this research was to enhance the methods towards the development of an efficient three dimensional face recognition system. More specifically, one of our aims was to investigate how the use of curvature of the diagonal profiles, extracted from 3D facial geometry models can help the neutral face recognition processes. Another aim was to use a gender classifier employed on 3D facial geometry in order to reduce the search space of the database on which facial recognition is performed. 3D facial geometry with facial expression possesses considerable challenges when it comes face recognition as identified by the communities involved in face recognition research. Thus, one aim of this study was to investigate the effects of the curvature-based method in face recognition under expression variations. Another aim was to develop techniques that can discriminate both expression-sensitive and expression-insensitive regions for ii face recognition based on non-neutral face geometry models. In the case of neutral face recognition, we developed a gender classification method using support vector machines based on the measurements of area and volume of selected regions of the face. This method reduced the search range of a database initially for a given image and hence reduces the computational time. Subsequently, in the characterisation of the face images, a minimum feature set of diagonal profiles, which we call T shape profiles, containing diacritic information were determined and extracted to characterise face models. We then used a method based on computing curvatures of selected facial regions to describe this feature set. In addition to the neutral face recognition, to solve the problem arising from data with facial expressions, initially, the curvature-based T shape profiles were employed and investigated for this purpose. For this purpose, the feature sets of the expression-invariant and expression-variant regions were determined respectively and described by geodesic distances and Euclidean distances. By using regression models the correlations between expressions and neutral feature sets were identified. This enabled us to discriminate expression-variant features and there was a gain in face recognition rate. The results of the study have indicated that our proposed curvature-based recognition, 3D gender classification of facial geometry and analysis of facial expressions, was capable of undertaking face recognition using a minimum set of features improving efficiency and computation.

The rate of adoption of renewable energy (RE) production and associated enterprises onfarms in the UK has been lower than expected suggesting that the UK government’s energy, agricultural and climate change objectives may not be achieved. The aim of this research is to investigate why this is the case by assessing the uptake, motivations, constraints and the factors affecting farmers’ RE investment intentions. Building on extant research literature (institutional theory, social cognition theory, theory of planned behaviour and the resource based view) a novel comprehensive and multidimensional model of entrepreneurial intentions was developed and tested using principal component, path and multivariate regression analysis techniques. Data were collected to test the model through a sample of 2000 farmers in the West Midlands Region of the UK. Of the 393 farmers who responded, 14% adopted RE enterprises, with half of adopters reporting slight to significant improvements in farm business performance in 2009. Solar panels were the most popular of the RE technologies available to farmers, compared to biomass related technologies. The study found that the most influential personal level factors contributing to the adoption of RE and associated technologies were cognitive such as the level of education. Of current 338 non-adopters, 66% might decide to invest in RE technologies over the next five years. For these potential adopters, the study shows that the type of tenure, educational attainment and the type of farm business diversification activity in which a farmer is engaged are the most significant personal and farm business situational factors which influence farmers’ RE investment intentions though contrary to expectation current non-adopters assessed the policy support framework more favourably than current adopters. The explanation of this seems to be connected with timing, in that two very positive and encouraging signals in relation to ii Feed in Tariffs (2010) and the Renewable Heat Incentive (2011) were underway or near introduction before this research took place. The study provides the first empirical evidence of the effects of the multidimensional measures of the country’s institutional profile on farmers’ RE investment intentions. Secondly, it clarifies the distinct role played by national formal and informal institutions on farmers’ investment intentions showing that informal institutions and not formal regulatory factors have a direct effect on farmers’ intentions to invest in RE enterprises. Thirdly, the investigation reveals that social acceptability of entrepreneurship in the RE sector is negatively related to investment intentions and moderates the efficacy of formal government policies in influencing entrepreneurial behaviour in the RE sector. The study concludes that any study that relies only on one type of institution will be making significant prediction mistakes. This study provides further support for cognitive based process models of intentions by showing strong significant positive effects of perceived self-efficacy and perceived desirability of RE enterprises on investment intentions. In fact, the study shows that farmers’ attitudes towards RE explain the highest amount of variance in investment intentions over and above the combined effect of external resource and institutional factors. The study illustrates that perceived self-efficacy and perceived desirability of RE enterprises mediate the effect of the rich set of exogenous variables investigated in this study on investment intentions and argues that policy makers need to focus on improving the regulatory, cognitive and normative institutional environments as a way to improve attitudes towards RE and consequently their intentions to invest in these enterprises.

Multiple geophysical and geological data sets were compiled, reprocessed, and interpreted using state-of-the-art signal processing and modeling algorithms to characterize the complex post-Paleozoic geology that overlies the southwestern projection of the Fluorspar Area Fault Complex (FAFC) in western Kentucky. Specific data included 21.5 km of SH-wave seismic reflection, 1.5 km of P-wave seismic reflection, 2 km of electrical resistivity, vertical seismic profiles, Vp and Vs sonic-suspension logs, and 930 lithologic borehole logs. The resultant model indicates three general northeast–southwest-oriented fault zones pass through the study area as southwestern extensions of parts of the FAFC. These fault zones form two significant subparallel grabens with ancillary substructures. The geometry of the interpreted fault zones indicates that they have undergone episodic tectonic deformation since their first formation. Evidence of thickening and steeply dipping reflectors within Tertiary and Quaternary sediment in the downthrown blocks indicate syndepositional movement. Subtle thickening and lack of steeply dipping intraformational reflectors in the Cretaceous suggest a more quiescent period, with sediment deposition unconformably draping and filling the earlier Paleozoic structural surface. There is also evidence that the Tertiary and early Quaternary reactivation was associated with an extensional to compressional regional stress reversal, as manifested by the antiformal folds seen in the hanging wall reflectors and the potential small-amplitude force folds in the Quaternary alluvium, as well as a clear displacement inversion along the Metropolis-loess seismic horizon in two high-resolution reflection images. A surface shear-wave splitting experiment proved to be an efficient and effective tool for characterizing shallow subsurface azimuthally anisotropic geologic inclusions in low-impedance water-saturated sediment environments. The measured azimuthal anisotropy across a well-constrained N60ºE-striking fault exhibited a natural coordinate system that had a fast direction coincident with the fault strike and an orthogonal slow direction. This is also one indicator that faults inactive during significant geologic intervals (i.e., Holocene) do not "heal". Integrated shear-wave velocity models and electrical resistivity tomography profiles across the fault zones exhibit lower shear-wave velocities and resistivities within the deformation zones compared with values outside the boundaries. This is additional evidence that the deformed sediment does not reconsolidate or heal, but that the sediment particle configuration remains more loosely packed, providing an increase in the overall porosity (i.e., hydraulic conductivity). This can wholly or in large part explain the anomalous contaminant plume migration path that is coincident with the deformed zones of the regional gravel groundwater aquifer.

<p>The goal of this work was to contribute to the fed-batch process optimisation task by deriving parameters that have considerable impact on productivity as well as product quality The chosen parameters were I) the design of the glucose feed profile, II) the choice of induction strategy, with respect to the method of addition, and III) the time of the induction, with respect to the specific glucose consumption rate. </p><p>The present fed-batch experiments using the lacUV5-promoter, for production of b-galactosidase, have shown that a high glucose feed rate gives a specific production rate, q_p, that is twice as high, after induction, compared to a feed rate that is 2.5 times lower. The constant accumulation of lacZ-mRNA indicates that the translational capacity is initially limiting the synthesis machinery, but after four hours of maximum specific production and a corresponding drop in lacZ-mRNA production, the cultivation is likely to be transcription limited. The high feed-rate system resulted in high accumulation of β-galactosidase, corresponding to 40% of total cellular proteins.</p><p>By design of feed profiles in a fed-batch process the detrimental effects of overflow metabolism, giving acetic acid formation, can be avoided. However, the results show that a one-dose addition of isopropyl-β-D-galactopyranoside (IPTG), provokes a non-growth associated production of acetic acid. This response can be alleviated by; lowering the inducer concentration (in this case to below 165 μM), by further reducing the feed rate of glucose or by using alternative induction methods. The use of a stepwise addition or a feed of IPTG thus delayed and reduced the level of acetic acid accumulation. It was also shown that a small change in the time-point of induction lead to large variability, regarding both productivity and acetic acid accumulation, in a fed-batch cultivation, </p><p>In order to further investigate the protein quality two additional proteins were studied in fed-batch cultivations using high and low glucose feed. The aim was to prove the hypothesis that the feed related change in the rate of synthesis of the nascent polypeptide controls the product quality. For the two proteins: Zb-MalE (wt) and Zb-MalE31 (mutant), the transcription rate, in terms of amount of IPTG, and translation rate, in terms of changes in feed rate, influences the percentage of inclusion body formation and degradation of nascent polypeptide. The data show a higher rate of inclusion body formation for the model protein Zb-MalE31 during high feed rate cultivations, as well as at high levels of inducer. Furthermore, the rate of proteolysis was significantly higher for a high feed rate. The high feed rate thus results in a higher rate of synthesis but a lower corresponding quality, for the model proteins studied.</p><p>In the present investigation of fed-batch cultivations using several different expression vectors, it was found that the central alarmone guanosine tetraphosphate (ppGpp) was formed at both high and low feed rates upon induction. It could be shown, however, that by secretion of Zb-MalE to the periplasm, the stringent response could be avoided. This might be due to the decreased burden on the host where the secretion of product further seems to make the cell able to redirect the carbon flux from overflow metabolism, since no acetic acid was produced. The secretion also demonstrates that the growth arrest could be aborted, which is otherwise gained in the P_malKproduction system.</p><p>A novel fed-batch process based on the promoters for the universal stress proteins A and B (P_uspA, P_uspB) was designed to make use of these powerful promoters in an industrial production context. It was concluded that the process had to start from a high specific growth rate and induction was performed once a limiting feed started. This was done to purposely induce the stringent response and/or acetic acid accumulation since this was required for induction. In the suggested system, induction has to be performed and maintained at continuous substrate feeding, whilst avoiding exceeding the cellular capacity, since the stationary phase starvation alone did not lead to production. In conclusion, a new stress induction based production system was achieved resulting in high accumulations of product protein without any detected metabolic side effects.</p>

<p>This thesis gives a systematic description of theEnvironmental Load Profile (ELP), an environmental assessmenttool developed for the urban environment. The purpose of thework was to improve the stringency of the system boundaries andfunctional units of the tool. This was achieved by putting theELP structure in the context of Life Cycle Assessment (LCA)with a special emphasis on system boundaries. To create animproved scientific base for the ELP, a comparative study wasconducted using an evaluative framework for conceptual andanalytical approaches. Here, the ELP tool is compared with foursimilar environmental assessment tools for the builtenvironment.</p><p>Since, energy use in the operation phase is an importantfactor for the overall environmental performance of buildings,a sensitivity analysis was performed to investigate how theselection of heat and electricity mix affects the results of anenvironmental assessment of buildings. Four modes ofelectricity production and two modes of heat production wereapplied on three buildings with different technical systems intheir heat supply. The results show that the choice ofelectricity mix has a great influence on the outcome of anenvironmental assessment (EA) and it is suggested that both anaverage and marginal electricity mix should be applied inEA´s of the built environment. Further, it is argued thatconsequences of assumptions made should be explicitlycommunicated in the EA report, to allow the decision-makersrather than the analysts to make the final evaluation.</p><p>The ELP is primarily developed to follow up theenvironmental goal‘twice as good’and assess theenvironmental performance of Hammarby Sjöstad, a newcity-district in southern Stockholm. The city-district is builtas a continuation of the inner Stockholm and the first part ofthe project, called Sickla Udde, is nearly finished. The ELPtool was applied in a first case study to answer the questionof how far Sickla Udde has reached in achieving the goal. Theassessment indicates that compared to a reference districtbased on the technology used in 1990, the environmentalperformance of Sickla Udde has reached the goal‘twice asgood’for some environmental load categories and 30percent for others. Although these findings are preliminary,they indicate a development in the right direction. Measurestaken contributing to largest environmental improvements are: amore efficient energy production (improved district heating)and use (e.g. lower U-values in the buildings, energy efficientappliances, heat exchange of ventilation air) and improvedsewage treatment. The results also demonstrate that theenvironmental load from domestic transports can be of the samemagnitude as from the buildings situated within thecity-district. Hence, resources spent to decrease environmentalload in the planning process should primarily be divoted toimproving domestic transportation systems and on optimising theoperational phase of the buildings.</p><p><b>Keywords:</b>environmental assessment, urban district,environmental load profile, Hammarby Sjöstad, life cycleassessment, LCA, environmental management, builtenvironment</p>

<p>The company System Andersson is developing systems for controlling materials and production in industries. They want to add a new service, where parts of those systems are accessible through cell phones.</p><p>In this paper we are investigating which technology are best suited for this kind of application. And then we describe how we are developing a prototype fort this application, based upon that technology. After investigating technologies such as Java ME, Microsoft Silverlight, and web applications we settled upon building the system as a web application, customized for a mobile environment.</p><p>To build a web page which is readable by many different cell phones we use the markup-languages which are developed for cell phones. We put much effort in reading about the standards and recommendations for the mobile web, produced by the organizations W3C and OMA. It's also very important that one consider the limitations of cell phones, such as the small screens, their limited performance and their limited means of input and navigation. Finally we are very thorough in our testing on several cell phones created by different manufacturers and with different operating systems and web browsers.</p><p>On the server we are running ASP.NET, with a database using Microsoft SQL Server, since this was a request from the company. The development is made in C#.</p><p>We reach a fully operational application with the functions that the company has requested. We got functions for handling orders, changing jobs, supply management, listing employees, and more. We also got a class for generating and rendering of lists and tables for easy and customizable lists for small screens. The application has support for several users with a secure login function.</p><p>We have also created a system running as a service in Microsoft Windows, which is regularly sending messages, in the form of e-mails to the users of the mobile application. The messages contain user defined reminders, e.g. that the factory is running low on a certain material. The message system is well integrated with the rest of the system, in which one can list the active reminders, edit them or create new ones.</p><p>Overall this has been an interesting and successful project.</p>

Necrotizing enterocolitis (NEC) is a costly and deadly disease in neonates. Composite risk for NEC is poorly understood and consensus has not been established on the relevance of risk factors. This two-phase study attempted to validate and test a neonatal NEC risk index, GutCheck(NEC). Phase I used an E-Delphi methodology in which experts (n=35) rated the relevance of 64 potential NEC risk factors. Items were retained if they achieved predefined levels of expert consensus or stability. After three rounds, 43 items were retained (CVI=.77). Qualitative analysis revealed two broad themes: individual characteristics of vulnerability and the impact of contextual variation within the NICU on NEC risk. In Phase II, the predictive validity of GutCheck(NEC) was evaluated using a sample from the Pediatrix BabySteps Clinical Data Warehouse (CDW). The sample included infants born<1500 grams, before 36 weeks, and without congenital anomalies or spontaneous intestinal perforation (N=58,818, of which n=35,005 for empiric derivation and n=23,813 for empiric validation). Backward stepwise likelihood-ratio method regression was used to reduce the number of predictive factors in GutCheck(NEC) to 11 and derive empiric weights. Items in the final GutCheck(NEC) were gestational age, history of a transfusion, NICU-specific NEC risk, late onset sepsis, multiple infections, hypotension treated with Inotropic medications, Black or Hispanic race, outborn status, metabolic acidosis, human milk feeding on both day 7 and day 14 (reduces risk) and probiotics (reduces risk).Discrimination was fair in the case-control sample (AUC=.67, 95% CI .61-.73) but better in the validation set (AUC=.76, 95% CI .75-.78) and best for surgical NEC (AUC=.84, 95% CI .82-.84) and infants who died from NEC (AUC=.83, 95% CI .81-.85). A GutCheck(NEC) score of 33 (range 0-58) yielded a sensitivity of .78 and a specificity of .74 in the validation set. Intra-individual reliability was acceptable (ICC (19) =.97, p<.001). Future research is needed to repeat this procedure in infants between 1500 and 2500 grams, complete psychometric testing, and explore unit variation in NEC rates using a comprehensive approach.

Microalgae has shown to be an ideal choice for biofuel industry. Algae has high oil productivity, a short growth cycle and survives in a wide variety of water sources including high salinity and waste water. For this project, four different species of marine microalgae were screened based on oil content. They were Dunaliella tertiolecta (CCMP364), Nannochloropsis gaditana (CCMP527), Tetraselmis sp (CCMP 908) and Nannochloropsis salina (CCMP1776). Experimental results showed that CCMP 527 and 1776 strains had higher lipid content and better fatty acids profile than the other two. Further investigations were carried on CCMP 527 in order to maximize biomass productivity and lipid content. Nutrients, salinity, pH, temperature, light intensity and aging of the culture can all affect both lipid content and fatty acid profile and were investigated. Nutrient stress is the easiest way to manipulate lipid composition and increase lipid content. Hence, various carbon and nitrogen sources were investigated to determine the range and amount of substrates that may be feasible for cultivation. For supplying lipid for biodiesel production, the optimum culture conditions for strain Nannochloropsis gaditana are using CO₂ enriched air bubbling, f/2-Si medium, pH control, and nitrate as the nitrogen source. Use of other fertilizers is feasible as well, however, the nitrogen source greatly affects lipid productivity, but trace amounts of organics in ground water do not.A model which predicts cell growth, nitrogen concentration, and lipid yield in batch systems is developed that is applicable for low nitrogen conditions. Plus, a sensitivity analysis of three major parameters was done to validate how variations in these key parameters affect simulation results. The fatty acid profile as a function of time was shown not to vary from mid-exponential to stationary phase. The model describes reactor behavior well, therefore it can be applied to the genus of Nannochloropsis to predict biomass yield and lipid accumulation, and be a useful tool to optimize and compare bioreactor systems for the biofuel industry. In addition, effects of nitrate and urea under repletion condition on microalgae growth, lipid yield and fatty acids profile for microalgae Nannochloropsis gaditana were investigated. Replacing nitrate by urea didn't show positive influence on lipid content and yield compared to normal medium. The major fatty acids for these two mediums were palmitic acid (C16:0) and palmitioleic acid (C16:1). Nannochloropsis gaditana still shows to be ideal candidate for biodiesel production using urea or nitrate enriched agriculture wastewater.

With the proliferation of mobile devices and wireless connectivity technologies, mobile social communities offer novel opportunities for targeted marketing by service or product providers. Unfortunately, marketers are still unable to realize the full potential of these markets due to their inability to effectively target right audiences. This thesis presents a novel middleware for identifying spontaneous social communities (SSCs) of mobile users in ad hoc networks in order to facilitate marketers' advertisements. The contributions of the presented work are two fold; the first is a novel model for SSCs that captures their unique dynamic nature, in terms of community structure and interest in different \textit{hot-topics} over time. These time-varying interests are represented through an inferred \textit{community profile prototype} that reflects dominant characteristics of community members. This prototype is then employed to facilitate the identification of new potential members. The selected community prototypes are also used by marketers to identify the right communities for their services or products promotions. The second contribution of this paper is novel distributed techniques for efficient calculation of the community prototypes and identification of potential community links. In contrast to traditional models of detecting fixed and mobile social networks that rely on pre-existing friendships among its members to predict new ones, the proposed model focuses on measuring the degree of similarity between the new user's profile and the profiles of members of each community in order to predict new users' relationships in the community. The adopted model of SSCs can foster many existing and new socially-aware applications such as recommender systems for social events and tools for collaborative work. It is also an ideal target for business-oriented applications such as short-message-service (SMS) advertisement messages, podcasting news feeds in addition to location/context-aware services. The performance of the proposed work was evaluated using the NetLogo platform where obtained experimental results demonstrate the achieved high degree of stability in the resulting communities in addition to the effectiveness of the proposed middleware in terms of the reduction in the number of routing messages required for advertisements.

Motivation: Cardiovascular disease (CVD) is one of the leading causes of mortality and morbidity in South Africa and worldwide. Various investigations have confirmed the hypothesis that elevated plasma homocysteine (Hcy) levels may be linked to vascular disease, and it has become clear that hyperhomocysteinemia is an independent risk factor for atherosclerosis and atherothrombosis. Extensive research on the influence of vitamin supplementation leading to the lowering of homocysteine levels has been done, but extensive research on the effect of physical activity on high homocysteine levels is lacking. The interaction of vitamin supplementation in combination with physical activity has also not been investigated. If a conditioning exercise programme can demonstrate a lowering effect on elevated homocysteine levels, it will confirm the importance of physical activity as a less expensive alternative for a better lifestyle that can also continue to lower morbidity and mortality rates. Objective: This study examined the effect of a conditioning program, vitamin supplement and a combination of both on Hcy levels in men with coronary heart disease risk factors. Methods: In a randomized, placebo-controlled, blinded crossover study, 84 men matched for physical activity (PA) levels, age and risk factors were randomly assigned to one of 4 groups [A = physical conditioning, 20-30 min; 70-80% (THR), 8 = physical conditioning + supplement, C = supplement (12,5 ug vitamin 812; 200 ug folic acid) or D = control). Groups A, B, and C were crossed over according to the Latin square design. Total plasma homocysteine, maximal oxygen consumption (V02max) and body composition (BMI & Fat %) were measured before and after each 12-week intervention period. A 6-week washout period separated the crossovers. Results: The experimental and control groups presented similar baseline characteristics and the profile analysis of the V02max values and Hcy concentrations indicated positive results (multivariate p-value <0.0001), due to the fact that the four groups repeated measurements, presented different patterns. A phase effect for the V02max values and a phase and interaction effect for the Hcy concentrations were indicated, though all the subjects were requested to maintain their normal daily routine (eating pattern, PA levels and alcohol consumption) for the duration of the study. The lack of compliance to the conditioning programme makes it impossible to draw conclusions for V02max values. The poor compliance lead to a small sample size that eventually leads to less statistical power. Conclusion: This study found that a 12-week conditioning programme had no effect on Hcy concentrations. The results of this study make it impossible, due to poor compliance, to suggest that the effect of increased PA on homocysteine may play an important role in the prevention and treatment of CVD. It is, therefore, recommended that more studies should be conducted to further investigate the effect of PA and vitamin supplements on tHcy levels. / Thesis (M.Sc. (Human Movement Science))--North-West University, Potchefstroom Campus, 2006.