Made available in DSpace on 2016-08-17T14:53:26Z (GMT). No. of bitstreams: 1 Leonardo Martins.pdf: 1400853 bytes, checksum: 3b6aa06e1c4b580a53150460124fdeaa (MD5) Previous issue date: 2007-12-07 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Breast cancer is a serious public health problem in several countries of the world. Computer-Aided Detection/Diagnosis systems (CAD/CADx) have been used with relative success in aid to health care professionals. The goal of such systems is not to replace the professional, but join forces in order to early detect the different types of cancer. The main contribution of this work is to present a methodology for detecting masses in digitized mammograms using the algorithm Growing Neural Gas for the segmentation of the image and Ripley’s K function to describe the texture of segmented objects. The classification of these objects is accomplished through a Support Vector Machine (SVM), which separates them into two groups: masses and non-masses. The methodology obtained 89,30% of accuracy and a rate of 0,93 false-positive per image. / O câncer de mama apresenta-se como um grave problema de saúde pública em vários países do mundo. Sistemas de Detecção e Diagnóstico baseados em computador (CAD/CADx) vêm sendo usados com relativo sucesso no auxílio aos profissionais de saúde. O objetivo de tais sistemas não é substituir o profissional, mas unir forças com o objetivo de detectar precocemente os diferentes tipos de câncer. A principal contribuição deste trabalho é apresentar uma metodologia para detecção de massas em imagens mamográficas digitais, utilizando para tanto o algoritmo Growing Neural Gas para a segmentação da imagem e a função K de Ripley para descrever a textura dos objetos segmentados. A classificação desses objetos é feita através de uma Máquina de Vetor de Suporte (Support Vector Machine - SVM), a qual separa os mesmos em dois grupos: massa e não-massa. A metodologia obteve 89,30% de acerto e uma taxa de 0,93 falso-positivos por imagem.

O uso de imagens provenientes de sensores remotos, tal como sistemas acoplados em aviões e satélites, é cada vez mais frequente, uma vez que permite o monitoramento continuo e periódico ao longo do tempo por meio de diversas observações de uma mesma região, por vezes ampla ou de difícil acesso. Essa ferramenta tem se mostrado importante e significativa em aplicações como o mapeamento de solo e fronteiras; acompanhamento de áreas de desmatamento, queimadas e de produção agrícola. Para gerar resultados interpretáveis ao usuário final, essas imagens devem ser processadas. Atualmente, o método de classificação por máxima verossimilhança é o mais empregado para classificação de imagens multiespectrais de sensores remotos, entretanto, por se tratar de uma técnica supervisionada, seus resultados dependem extensivamente da qualidade do conjunto de treinamento, utilizado para definir os parâmetros do método. A seleção de um bom conjunto de treinamento é um processo custoso e inviabiliza a automação da classificação para diversas imagens. O método de classificação por máxima verossimilhança é também paramétrico e portanto exitem algumas suposições quanto a distribuição dos dados que devem ser atendidas, caso contrário a aplicação do método pode gerar resultados ruins. Tendo em vista essas desvantagens do método da máxima verossimilhança, este trabalho propõe um novo método para a classificação de imagens multiespectrais provenientes de sensores remotos de forma que o procedimento seja autônomo, veloz e preciso, minimizando dessa forma os possíveis erros humanos inseridos em etapas intermediárias do processo, tal como a definição de conjuntos de treinamento. O método aqui proposto pertence ao conjunto das redes neurais artificiais (RNAs) e é denominado growing neural gas (GNG). Este método baseia-se no aprendizado não supervisionado de padrões \"naturais\" dentro de um conjunto de dados por meio da criação e adaptação de uma rede mínima de neurônios. Os resultados gerados a partir da classificação pela RNA foram comparados com os métodos mais utilizados na literatura atual, sendo eles o método da máxima verossimilhança e o método k-means. A partir da biblioteca espectral ASTER, mantida e criada parcialmente pela NASA, foram realizadas várias repetições do experimento, que consiste em classificar os dados de acordo com as diferentes classes existentes, e para cada uma destas repetições calculou-se uma medida de acurácia, denominada índice kappa, além do tempo de execução de cada método, de forma que suas médias foram comparadas via intervalo de confiança gerados por bootstrap não paramétrico. Também investigou-se como a análise de componentes principais (ACP), técnica utilizada para reduzir a dimensão dos dados e consequentemente o custo computacional, pode influenciar no desempenho dos métodos, tanto em sua qualidade de classificação quanto em relação ao tempo de execução. Os resultados mostram que o método proposto é superior nos dois aspectos estudados, acurácia e tempo de execução, para a maioria dos fatores aplicados. Mostra-se ainda um exemplo de aplicação prática em que uma imagem multiespectral de satélite não satisfaz as pré-suposições estabelecidas para o uso do método da máxima verossimilhança e verifica-se a diferença entre os métodos com relação a qualidade final da imagem classificada. / The use of images from remote sensors, such as coupled systems in airplanes and satellites, are increasingly being used because they allow continuous and periodic surveillance over time through several observations of some particular area, sometimes large or difficult to access. This sort of image has shown an important and meaningful participation in applications such as soil and borders mapping; surveillance of deforestation, forest fires and agricultural production areas. To generate interpretable results to the end user, these images must be processed. Currently, the maximum likelihood classification method is the most used for multispectral image classification of remote sensing, however, because it is a supervised technique, the results depend extensively on the quality of the training set, used to define the parameters of the method. Selecting a good training set is a costly process and prevents the automation of classification for different images. The maximum likelihood classification method is also parametric, and therefore, some assumptions about the distribution of the data must be met, otherwise the application of the method can generate bad results. In view of these disadvantages of the maximum likelihood method, this dissertation proposes a new, autonomous, fast and accurate method for multispectral remote sensing imagery classification thereby minimizing the possible human errors inserted at intermediate stages of the process, such as the definition of training sets. The method proposed here belongs to the set of artificial neural networks (ANN) and is called growing neural gas (GNG). This method is based on unsupervised learning of \"natural\" patterns in a dataset through the creation and adaptation of a minimum network of neurons. The results generated from the classification by ANN were compared with the most commonly used methods in the literature: the maximum likelihood method and the k-means method. From the spectral library Aster, maintained and made in part by NASA, several replications of the experiment were made, which is to classify the data according to different preestablished classes, and a measure of accuracy called kappa index was calculated for each of the replicates, in addition to the execution time of each method, so that their means were compared via confidence interval generated by nonparametric bootstrap. It was additionally investigated how principal component analysis (PCA), technique which reduces dimension of data and consequently the computational cost, can influence the performance of methods, both in its quality rating and runtime. The results show that the proposed method is superior in both aspects studied, accuracy and runtime, for the majority of applied factors. Furthermore, it is shown an example of a practical application in which a multispectral satellite image does not necessarily meet the established assumptions for using the maximum likelihood method, and there is a difference between the methods, regarding to its final classified image quality.

Dissertation (PhD)--Stellenbosch University, 2004. / ENGLISH ABSTRACT: Pine sawdust-shavings (Pinus spp.) is at present a very popular soilless substrate in South African greenhouses. Growers use fresh pine sawdust-shavings as a substrate, which is biologically highly unstable. The greenhouse industry is looking at alternative organic substrates such as coco peat, which already went through a decomposition process and is more stable. A biological inactive substrate such as sand was included to compare microorganism activity with organic substrates. The main objective of this study was to compare the growth, yield and quality of hydroponically grown tomatoes in response to different growth mediums in combination with nitrogen source, irrigation frequency, period of substrate use and liming. In general the drainage water pH declined with an increase in NlLt+-N in the nutrient solution. Low pH values in the drainage water, especially when coco peat was used, had a detrimental effect on marketable yield. The drainage water pH of pine sawdustshavings increased during the growing season when 100 % N03--N was used. Due to the higher cation exchange capacity of coco peat, the drainage water electrical conductivity tends to increase more rapidly than with pine sawdust-shavings, during conditions with high temperatures and when insufficient irrigation volumes per irrigation cycle is applied. As expected the drainage water N03--N content decreased as the NlLt+-N content increased in the nutrient solution. Pine sawdust-shavings recorded a much lower N03--N and NlLt+-N content than sand and coco peat and thus supports the hypothesis that microbiological activity is higher in pine sawdustshavings, especially in the second season of substrate use. Coco peat produced the highest number of marketable fruit and yield per plant, followed by pine sawdustshavings and sand in the first season of substrate use. The number of marketable fruit and yield decreased with an increase in NlLt+-N content in the nutrient solution during production in warmer, summer conditions. Contrary to these fmdings, production in cooler, winter conditions recorded high yields when only N03--N or 80% N03--N : 20% NRt +-N was applied. The unmarketable yield increased with an increase in NlLt+-N in the nutrient solution. Visual evaluations showed that blossom-end rot (BER) was the main contributor to unmarketable yield. Increasing levels ofN03--N as nitrogen source in the nutrient solution, reduced weight loss and increased the loss of fruit firmness of tomatoes during storage. Increasing levels of N03 --N also increased fruit pH and reduced total titratable acidity. Coco peat produced fruit with a higher pH than pine sawdust-shavings. An increase in irrigation frequency affected fruit firmness negatively when coco peat was used as substrate. Different irrigation and fertigation practices are needed for different growth mediums and management needs to be adapted according to the growing season (winter vs. summer). / AFRIKAANSE OPSOMMING: 'n Mengsel van dennesaagsels en -skaafsels (Pinus spp.) word tans deur Suid- Afrikaanse kweekbuisprodusente gebruik as grondlose groeimedium. Hierdie groeimedium word nie vooraf gekomposteer nie en is dus biologies onstabiel. Die kweekbuisindustrie ondersoek tans die gebruik van alternatiewe, gekomposteerde en stabiele organiese groeimediums soos kokosveen. 'n Biologies onaktiewe groeimedium soos sand is ook ingesluit om met organiese groeimediums te kan vergelyk. Die hoof doelwit van die studie was om plantontwikkeling, opbrengs en kwaliteit van hidroponies geproduseerde tamaties te evalueer in verskillende groeimediums en in kombinasie met stikstofbron-verhouding, periode van groeimedium gebruik, besproeiingsfrekwensie en bekalking. Oor die algemeen het die pH in die dreinaat gedurende die groeiseisoen toegeneem soos die NH/-N inhoud verhoog het in die voedingsoplossing. Lae pH waardes in die dreinaat, veral waar kokosveen gebruik was, het 'n nadelige effek op bemarkbare opbrengs gehad. Die pH in die dreinaat van dennesaagsels en -skaafsels het gedurende die groeiseisoen toegeneem met die gebruik van 100% NO)--N in die voedingsoplossing. Die elektriese geleiding in die dreinaat van kokosveen neem vinniger toe gedurende toestande waarin hoë temperature en onder besproeiing voorkom, as in dreinaat van dennesaagsels en -skaafsels. Die NO)--N inhoud in die dreinaat het soos verwag afgeneem soos die NRt+-N inhoud in die voedingsoplossing toegeneem het. 'n Baie laer NO)--N en NRt+-N inhoud is by dennesaagsels en -skaafsels aangeteken wat dus die hipotese ondersteun dat mikrobiologiese aktiwiteit, veral in die tweede seisoen van gebruik, hoër is in dennesaagsels en -skaafsels as in sand en kokosveen. Kokosveen het die hoogste aantal bemarkbare vrugte en massa per plant geproduseer, gevolg deur dennesaagsels en -skaafsels en sand. Die aantal bemarkbare vrugte en opbrengs het verlaag met 'n verhoging in NRt+-N in die voedingsoplossing gedurende warm, somer toestande. In teenstelling met vorige resultate is gevind dat 100% NO)-- N of 80% NO)--N : 20% NRt+-N hoë opbrengste gelewer het gedurende koeler, winter toestande. Die onbemarkbare opbrengs het verhoog met hoër NRt+-N vlakke. Visuele waarnemings het aangedui dat blom-end verrotting die grootste bydrae tot onbemarkbare opbrengs gelewer het. 'n Verhoging in NO)--N vlakke het massaverlies beperk en die verlies in fermheid verhoog gedurende opberging. Hoër NO)--N vlakke het ook die pH van vrugte verhoog en die totale titreerbare suur verlaag. Kokosveen het vrugte met 'n hoër pH as dennesaagsels en -skaafsels geproduseer. 'n Toename in besproeiingsfrekwensie het vrug fermheid negatief beïnvloed wanneer kokosveen as groeimedium gebruik was. Verskillende besproeiings- en voedingspraktyke word benodig vir verskillende groeimediums en bestuur van die groeimediums moet aangepas word by klimaatstoestande gedurende die spesifieke produksieseisoen.

12 pp.

The study consisted of two experiments conducted over two growing seasons. Urea ammonium nitrate was used as a source of N at rates of 50, 75, 100 and 150% of levels estimated to be ideal for maximum yield of cotton (Gossvpium hirsutum L.). The nitrogen fertilizer was applied through a drip irrigation system. The yield of seed cotton, flowering pattern, boll set, plant N uptake, and dry matter production were studied in relation to four N fertilizer rates and two plant populations in the 1984 study. Yield of seed cotton, plant N uptake and dry matter production were studied in relation to four N rates, three seeding rates, and three cotton cultivars in the 1985 study. Petiole nitrate patterns were studied both seasons. The effect of N applications on seed cotton yield was dependent upon the initial soil N and the yield possibility. In this study the lower rate of N appeared to be sufficient for the yields obtained. Thinning resulted in reduction of the total number of flowers and significantly decreased yield, but percent boll set was not affected. Nitrogen additions significantly increased plant N uptake and dry matter production as well as petiole NO₃-N levels during the growing season. The N need of cotton under drip irrigation was determined throughout the growing season by using petiole analysis. The levels of petiole NO₃-N for N sufficiency and deficiency which are accepted under furrow irrigation cotton were shown to be applicable for drip irrigated cotton. Yield of DPL-775 and DPL-90 cotton cultivars was significantly higher than that for DPL-41 cotton cultivar in 1985.

No description available.

This thesis aims to provide an insight into the legal rules governing viticulture and winemaking, and their historical development. Furthermore, the current legal and factual situation as well as the current issues associated with the legislation in these fields are more closely examined in the thesis. An anonymous questionnaire had been created and sent to a selected group of small and medium-size viniculturists and wine producers in the Slovacko wine subregion in order to supplement the research and its aim of presenting the current legislation from de lege ferenda point of view.

L'amélioration de l'efficacité alimentaire est un enjeu économique et environnemental important pour la production porcine. L'aliment est le premier poste de dépenses en élevage porcin, et le restera vraisemblablement au vu de l'évolution du coût des matières premières. Historiquement, l'efficacité alimentaire a été améliorée en sélectionnant, avec succès, pour une diminution de l'indice de consommation (IC), qui représente l'efficacité de conversion de l'aliment en gain de poids. Dans un contexte d'évolution des objectifs de sélection, de nouvelles stratégies d'amélioration de l'efficacité alimentaire sont recherchées. Ce travail de thèse propose d'en explorer certaines en utilisant des données enregistrées en stations publiques de contrôle de performance entre 2000 et 2009 dans les quatre races porcines en sélection collective en France, deux races maternelles (Large White femelle et Landrace) et deux races paternelles (Large White mâle et Piétrain). L'approche privilégiée s'appuie sur la consommation moyenne journalière résiduelle (CMJR), qui est un critère d'efficacité alimentaire indépendant des performances des animaux, afin d'estimer ses relations génétiques avec les performances de production, ainsi qu'avec les quantités d'azote et de phosphore excrétées. Dans un premier temps, l'étude d'une population expérimentale Large White a montré que les corrélations génétiques entre sexes pour la CMJR étaient proches de 1, permettant par la suite de comparer les estimations de paramètres obtenues sur des femelles Piétrain à celles obtenues sur les mâles castrés des trois autres races. Nous montrons que ce caractère a une héritabilité (h²) comprise entre 0,21 et 0,33 selon les races, qu'il est un peu moins héritable que l'IC (h² entre 0,30 et 0,40), que ces deux caractères présentent des corrélations génétiques favorables (entre 0,52 et 0,85), et que si la CMJR est par définition phénotypiquement indépendante des caractères de croissance et de composition corporelle, elle l'est aussi au niveau génétique (corrélations inférieures à 0,16 en valeur absolue quelle que soit la race). En revanche, la CMJR, comme l'IC, a des relations génétiques antagonistes avec les caractères de qualité de la viande. Cependant, nous montrons que la CMJR, contrairement à l'IC, n'est pas affectée par le génotype halothane en ségrégation chez le Piétrain. Enfin, des corrélations génétiques très élevées avec l'IC (0,97 et plus), et plus modérées avec la CMJR (entre 0,38 et 0,86) ont été trouvées pour l'excrétion d'azote et de phosphore, suggérant la réduction attendue des rejets en réponse à la sélection sur l'efficacité alimentaire. Ces corrélations étaient globalement plus élevées dans les races paternelles que dans les races maternelles. Enfin, l'étude des cinétiques de croissance et d'ingestion des animaux en fonction de leur niveau d'efficacité alimentaire a montré qu'une sélection pour des animaux plus efficaces ne peut être réalisée qu'en adaptant la formulation des aliments, notamment leur concentration en acides aminés, de façon à permettre la pleine expression de leur potentiel génétique. / Improvement of feed efficiency is a main economic and environmental goal for pig production. Feed is the first component of cost production for pig farmers, and this will remain the case due to the evolution of feedstuffs cost. Historically, feed efficiency has been improved through selection, with success, for a decrease of feed conversion ratio (FCR) representing efficiency of feed conversion into body weight gain. In a context of evolving breeding objectives, new strategies of feed efficiency improvement are investigated. This PhD work has studied some of them using data collected in the French central test stations between 2000 and 2009 for the four main pigs breeds used in collective selection, i.e. two dam breeds (Large White dam line and Landrace) and two sire breeds (Large White sire line and Piétrain). The main approach retained here deals with residual feed intake (RFI), which is a criterion of feed efficiency independent from animal's performance level, in order to estimate its genetic relationships with usual production traits, and also with nitrogen and phosphorus quantities excreted by the animals. In a first step, a study of an experimental Large White population showed that genetic correlations between the three sexes for RFI were very close to 1 for RFI. This result allowed the comparisons of the results obtained on Piétrain females with those obtained on castrated males in the three other breeds. We showed that RFI had an heritability (h²) ranging between 0.21 and 0.33 depending on the breed, slightly less heritable than FCR (h² between 0.30 and 0.40) and with a favourable genetic correlation between them (between 0.52 and 0.85). The RFI is, by definition, phenotypically independent from growth rate and carcass composition traits, and this was also the case at the genetic level (genetic correlations lower than 0.16 in absolute values whatever the breed). However, RFI, like FCR, did not display favourable genetic correlations with meat quality traits. Contrary to FCR, RFI was shown to be unaffected by the halothane genotype segregating in the Piétrain breed. Very high genetic correlations with FCR (0.97 and higher), and more moderate genetic correlations with RFI (between 0.38 and 0.86) were found for nitrogen and phosphorus excreted quantities, which indicates that the reduction of excreted quantities is expected in response to selection for a better feed efficiency. These genetic correlations were globally higher in sire lines than in dam lines. Finally, the study of growth and feed intake curves of animals classified according to their feed efficiency level (either RFI or FCR) suggested that breeding for more efficient animals could not be conducted without an adaptation of feed formulation (e.g. for amino acids concentration), in order to allow the full expression of their genetic potential.

This dissertation deal with problems of fast growing trees (FGT). The aim is to create transparent summary of views and problems, which the public perceives in connection with growing FGT. On the basis of discovered knowledge is to map conditions in Trebic region which have an influence on creating supplier customerś relations, focusing on growing and processing FGT. Visits, excursions, workshops and interview to gain some information was used for making this dissertation. This work provides summary of views appealing entities (producers FGT, mayors, farmers) and evaluating conditions which may affect cooperation among appealed entities from Trebic region. The discovered knowledge prove that the ability to make more difficult relations run into the factors: economic, social, mental, legislation, and technical conditions.

The aim of this master thesis was the finding out of potential of growing the soybeans in the area with not so suitable growing conditions. The next aim was to perform the experiment with application of leaf fertilizer EGT Fulhum which increases root system volume. In the years 2016 and 2017 was in the altitude 395 m established the field experiment with the variety Amandine and since the year 2017 was also joint the experiment with the very early variety Abeline. Within these cultivars of soybean were evaluated these parameters: Seed yield, the 1000 seed weight, seed oiliness, plant height, number of pods on one plant, number of the early branches and weight of roots. The seed yield was very variable from the point of view of the year. The yield in the year 2016 was above-average, the control variant reached the yield 2,93 tons per hectare and variant treated by Fulhum 2, 78 tons per hectare. One of the studied parameter was the nitrogen content and based on the results of this parameter was noticeable reduction of nitrogen content of the treated variant in the comparison with control variant. In the further year were two compared two varieties - Amandine and Abeline. Within the both varieties were involved to the experiment the control variant and the variant treated by plant auxiliary substance. In the year 2017 was confirmed the influence of the crop area establishment on the yield quality parameters. The progress of weather confirmed the high requirements of soybean on the good rainfall conditions. Seed yield of the variety Amandine was 1,07 tons per hectare in the case of the treated variant, by control variant it was 0,88 tons per hectare. The difference of the seed yield between both variants of the variety Abeline was very little, it was 0,90 tons per hectare by the treated variant and 0,93 tons per hectare by the control variant. The application of the plant auxiliary substance didn't cause, on average, the improvement of the studied parameters so the using of this substance, for the improvement of these parameters almost doesn't make any sense.