O uso de dados biométricos da face para verificação automática de identidade é um dos maiores desafios em sistemas de controle de acesso seguro. O processo é extremamente complexo e influenciado por muitos fatores relacionados à forma, posição, iluminação, rotação, translação, disfarce e oclusão de características faciais. Hoje existem muitas técnicas para se reconhecer uma face. Esse trabalho apresenta uma investigação buscando identificar uma face no banco de dados ORL com diferentes grupos de treinamento. É proposto um algoritmo para o reconhecimento de faces baseado na técnica de subespaço LDA (PCA + LDA) utilizando uma rede neural SOM para representar cada classe (face) na etapa de classificação/identificação. Aplicando o método do subespaço LDA busca-se extrair as características mais importantes na identificação das faces previamente conhecidas e presentes no banco de dados, criando um espaço dimensional menor e discriminante com relação ao espaço original. As redes SOM são responsáveis pela memorização das características de cada classe. O algoritmo oferece maior desempenho (taxas de reconhecimento entre 97% e 98%) com relação às adversidades e fontes de erros que prejudicam os métodos de reconhecimento de faces tradicionais. / The use of biometric technique for automatic personal identification is one of the biggest challenges in the security field. The process is complex because it is influenced by many factors related to the form, position, illumination, rotation, translation, disguise and occlusion of face characteristics. Now a days, there are many face recognition techniques. This work presents a methodology for searching a face in the ORL database with some different training sets. The algorithm for face recognition was based on sub-space LDA (PCA + LDA) technique using a SOM neural net to represent each class (face) in the stage of classification/identification. By applying the sub-space LDA method, we extract the most important characteristics in the identification of previously known faces that belong to the database, creating a reduced and more discriminated dimensional space than the original space. The SOM nets are responsible for the memorization of each class characteristic. The algorithm offers great performance (recognition rates between 97% and 98%) considering the adversities and sources of errors inherent to the traditional methods of face recognition.

O objetivo deste trabalho é investigar a relação entre a Temperatura da Superfície do Mar (TSM) no Oceano Atlântico Sul (OAS) e a precipitação no estado de São Paulo. Para tanto, adotaram-se as escalas mensal e sazonal. A técnica estatística multivariada, chamada Análise de Componentes Principais (ACP, ou EOF em inglês) foi aplicada para definir as áreas homogêneas do oceano (de \'A\' a \'L\') e as subregiões do continente (de 1 a 7). Além da definição das áreas e sub-regiões, a ACP identificou muito bem os padrões espaciais da precipitação e da TSM em todas as escalas adotadas: mensal contínua, mensal descontínua e sazonal. Na ACP da série temporal sazonal de verão, o primeiro e o segundo modos da precipitação evidenciam a distribuição espacial típica de verão (zonal) para o estado de São Paulo, e a variância explicada por esses dois modos juntos é de 59,79% do total. Esta distribuição espacial já havia sido observada nas análises iniciais a partir das médias sazonais da precipitação, padrão que se explica pelos fluxos de umidade que, nesta estação, devido ao sistema de monções da América do Sul, movem-se mais intensamente em direção ao continente. Já na ACP da série temporal sazonal de inverno, o primeiro e segundo modos da precipitação evidenciam a distribuição meridional, com variância explicada de 39,15 e 17,77%, respectivamente. Os sistemas frontais no inverno agem como responsáveis por uma parcela considerável do total precipitado, o que explica o padrão meridional da chuva. Os coeficientes de correlação linear significativos entre as áreas oceânicas e sub-regiões continentais homogêneas demonstraram que a variabilidade sazonal da precipitação está associada à TSM, sobretudo na porção central da área estudada (próximo ao litoral sul e sudeste do Brasil), nas estações seca (abril a setembro) e chuvosa (outubro a março). Além das variáveis principais (precipitação e TSM), os campos de Radiação de Onda Longa Emergente (ROLE), Pressão no nível do Mar (PNM), e Divergência do ar em 250 e 850hPa auxiliaram na identificação dos padrões atmosféricos e na análise dos eventos extremos. As estações consideradas atípicas, ou seja, as que apresentaram média muito acima ou muito abaixo do desvio padrão, foram separadas em quatro classes: verão chuvoso, verão seco, inverno chuvoso e inverno seco. A partir da análise desses eventos, conclui-se então que a TSM do OAS pode alterar consideravelmente o padrão sazonal de precipitação no estado de São Paulo, principalmente se estiver atrelada a padrões atmosféricos. No inverno, anomalias negativas (positivas) de TSM estão associadas com a precipitação acima (abaixo) da média climatológica nesse estado. Devido a bloqueios atmosféricos no OAS no verão, a diminuição da passagem de sistemas frontais pode intensificar anomalias positivas no oceano, as quais estão associadas aos períodos chuvosos no sul do país e de seca no estado de São Paulo. / The purpose of this paper is to study the relationship between the Sea Surface Temperature (SST) in the South Atlantic Ocean (SAO) and the precipitation in São Paulo state. So, the monthly and seasonal scales were adopted. The Principal Component Analysis multivariate statistic technic (PCA, and also called EOF) was applied to define the oceans homogeneous areas (from A to L) and the continents sub-regions (from 1 to 7). Besides defining the areas and sub-regions, the PCA has identified very well the spatial patterns of the precipitation and SST in all the adopted scales: continuous monthly, discontinuous monthly and seasonal. In the PCA of the summer seasonal temporal series, the first and the second precipitations mode evidence the summer-typical spatial distribution (zonal) for São Paulo state, and the variance explained by these two modes together is 59,79% of the total. This spatial distribution was already noted in the first analysis from the seasonal precipitation averages, and such a pattern is explained by the flow of humidity, which in the summer, owing to the South America monsoon systems, move more intensely toward the continent. On the other hand, in the PCA of the winter seasonal temporal series, the first and the second precipitations modes evidence the meridional distribution, with a variance explained from 39,15 to 17,77%, respectively. The frontal systems in the winter act as responsibles for a substantial portion of the total of precipitation, which explains the rains meridional pattern. The significative linear correlation coefficients between the oceanic areas and the homogeneous continental sub-regions demonstrate that the precipitations seasonal variability is associated to the SST, mostly in the central portion of the studied area (next to the southern and southeastern Brazilian coast), in the drought season (from April to September) and rainy season (from October to March). Besides the principal variables (precipitation and SST), the fields of Outgoing Long wave Radiation (OLR), Sea-level Pressure (SLP) and air Divergence in 250 and 850hPa has helped to identify the atmospheric patterns and to analyze the extreme events. The atypical seasons, in other words, those which presented an average too high or too low compared to the standard deviation, were divided in four categories: rainy summer, drought summer, rainy winter and drought winter. With the analysis of these events, it is possible to conclude that the SST of the SAO can change substantially the precipitation seasonal pattern in São Paulo state, mainly if it is linked to atmospheric patterns. In the winter, SSTs negative (positive) anomalies are associated to the overage (underage) climatologic precipitation in this state. Owing to atmospheric blockades on the SAO in the summer, the decreasing transit of frontal systems can intensify positive anomalies in the ocean, which are associated to the rainy periods in the Brazilian South and drought periods in the São Paulo state.

Submitted by Gabriela Lopes (gmachadolopesufpel@gmail.com) on 2017-04-24T11:58:28Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) dissertaçaofinalll.pdf: 1151891 bytes, checksum: 0b347b32b7f2948018d266fd346e3ed7 (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2017-05-04T19:03:28Z (GMT) No. of bitstreams: 2 dissertaçaofinalll.pdf: 1151891 bytes, checksum: 0b347b32b7f2948018d266fd346e3ed7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-05-04T19:03:28Z (GMT). No. of bitstreams: 2 dissertaçaofinalll.pdf: 1151891 bytes, checksum: 0b347b32b7f2948018d266fd346e3ed7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-03-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / O mel é um alimento natural e nutritivo, com elevado teor de carboidratos e água, contendo também quantidades menores de diversos ácidos orgânicos, enzimas, vitaminas, hormônios, flavonoides, proteínas, compostos aminados e outros elementos orgânicos e inorgânicos. Compostos fenólicos e carotenoides, mesmo estando em baixa concentração, conferem efeito anti-inflamatório, antioxidante e antimicrobiano ao produto, aumentando a imunidade e outras atividades fisiológicas do organismo. O teor e diversidade dos constituintes do mel são dependentes das condições regionais da área forrageada pelas abelhas. Assim, este trabalho teve por objetivo a determinação do perfil físico químico, de compostos bioativos e da atividade antioxidante de méis oriundos de mesorregiões do estado do Rio Grande do Sul. Para tanto, foram analisadas vinte e uma amostras de diferentes produtores do Sudoeste e Sudeste Rio-Grandense, quanto ao teor de cinzas, pH, acidez total, umidade, sólidos solúveis totais, cor, condutividade elétrica e açúcares, através de metodologias oficiais; além das avaliações de compostos fenólicos, flavonoides, carotenoides e atividade antioxidante, por espectrofotometria. Para a maioria das amostras, os parâmetros físico-químicos determinados se apresentaram dentro dos padrões brasileiros e internacionais. Os conteúdos de compostos fenólicos e de flavonoides das amostras avaliadas neste estudo reafirmam a qualidade dos méis. A análise de componentes principais permitiu agrupar amostras de acordo com sua origem geográfica, em dois grandes grupos, utilizando apenas seis parâmetros. Mesmo que a flora e as condições climáticas da zona geográfica estudada sejam semelhantes, este trabalho demonstra que pode ser possível classificar o mel por origem geográfica, o que contribuiria para conhecimento da rastreabilidade físico química deste produto. / Honey is a natural and nutritious food with a high content of carbohydrates and water, also containing smaller amounts of various organic acids, enzymes, vitamins, hormones, flavonoids, proteins, amino compounds and other organic and inorganic elements. Phenolic compounds and carotenoids, even in low concentration, confer anti-inflammatory, anti-oxidant and antimicrobial effect to the product, increasing the immunity and other physiological activities of the organism. The content and diversity of the constituents of honey are dependent on the regional conditions of the area foraged by the bees. The objective of this work was to determine the chemical physical profile, bioactive compounds and antioxidant activity of honey from mesoregions in the state of Rio Grande do Sul. Twenty one samples from different producers were analyzed of the Southwest and Southeast Rio-Grandense, as regards ash content, pH, total acidity, moisture, total soluble solids, color, electrical conductivity and sugars, through official methodologies; besides the evaluations of phenolic compounds, flavonoids, carotenoids and antioxidant activity, by spectrophotometry. For most of the samples, the physical and chemical parameters determined were within the Brazilian and International standards. The contents of phenolic compounds and flavonoids of the samples evaluated in this study reaffirm the quality of the honey. The analysis of main components allowed to group samples according to their geographical origin, in two large groups, using only six parameters. Even though the flora and climatic conditions of the geographical area studied are similar, this work demonstrates that should be possible to classify honey by geographical origin, which would contributes to the knowledge of the physical chemical traceability of this product.

Made available in DSpace on 2014-06-12T17:40:54Z (GMT). No. of bitstreams: 2 arquivo6273_1.pdf: 4227824 bytes, checksum: fca68f64de842a88997be5a7b7300ac7 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2007 / A Bacia do Rio Jaboatão, que juntamente com a Bacia do Rio Pirapama formam o grupo de pequenos rios litorâneos GL-2, situa-se na região da mata sul do Estado de Pernambuco. A bacia compreende parte dos municípios de Vitória de Santo Antão, Jaboatão dos Guararapes, Moreno, São Lourenço da Mata, Cabo de Santo Agostinho e Recife. A quase inexistência de infra-estrutura sanitária na área de abrangência da bacia do rio Jaboatão, o lançamento de efluentes industrias inadequadamente tratados, a disposição inadequada dos resíduos sólidos e o uso indiscriminado de agrotóxicos e fertilizantes químicos, principalmente motivados pelas culturas temporárias de cana de açúcar, faz com que a qualidade da água do rio esteja comprometida em diversos locais ao longo do seu curso. A elaboração da dissertação teve como objetivo a avaliação da qualidade da água da Bacia do Rio Jaboatão com o intuito de subsidiar a futura realização do enquadramento dos corpos de água em classes de uso preponderante e propor um enquadramento preliminar, como hipótese de trabalho para o enquadramento definitivo. O trabalho utilizou os resultados do programa de monitoramento da Agência Estadual de Meio Ambiente e Recurso Hídricos CPRH relativos aos anos de 1997 a 2005. Os resultados do monitoramento foram comparados com os valores dos limites das classes definidas pela resolução CONAMA 357/05. Foi utilizada a analise multivariada de dados através da técnica de projeção em componentes principais (PCA), e calculado o índice de qualidade da água (IQA) nesses dados, como também foi elaborado mapa em AutoCad Map, contendo informações de uso do solo, fontes poluidoras, usuários de água, pontos de coleta da CPRH, estações fluviométricas e pluviométricas. Como resultado verifica-se que os parâmetros monitorados encontram-se na sua maioria fora do limite da classe 2, o IQA variou de aceitável a péssimo ao longo do rio e a analise multivariada dos dados mostrou que os parâmetros que mais influenciam a má qualidade do rio são o OD, a DBO e o Coliforme fecal, o que evidencia o pouco ou a falta de investimento em políticas de controle da poluição nesta bacia

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / Com a finalidade de avaliar a qualidade das Ãguas das principais lagoas pertencentes à bacia hidrogrÃfica do rio Cocà (Porangabussu, Opaia, Itaperaoba, Maria Vieira, Messejana, Sapiranga, Maraponga e o Lago JacareÃ). Realizaram-se coletas de Ãgua em trÃs seÃÃes, no perÃodo de agosto/2006 e julho/2007, mensalmente. As variÃveis temperatura, turbidez, pH, CE, DBO5, DQO, OD, SST, SDT, nitrogÃnio amoniacal total, nitrito, nitrato, fÃsforo, alcalinidade, dureza e CTT foram analisadas. Os dados obtidos foram validados, padronizados (Escores Z) e submetidos, ao Programa The Unscrambler versÃo 7.5, para PCA visando reduzir os dados, facilitando na determinaÃÃo da homogeneidade: entre as lagoas e entre os pontos de amostragem de cada lagoa e tambÃm avaliar a interferÃncia da sazonalidade na qualidade da Ãgua dos ecossistemas, facilitando a compreensÃo. Para a homogeneidade entre os ecossistemas apenas a Lagoa de Sapiranga e o Lago Jacareà apresentaram-se heterogÃneos Ãs demais. Provavelmente devido à influÃncia da marà imposta a Lagoa de Sapiranga e ao pequeno aporte hÃdrico do Lago JacareÃ. A homogeneidade das outras lagoas ocorreu em funÃÃo do aporte de matÃria orgÃnica (MO), usos e ocupaÃÃo de suas Ãreas de influÃncias e o efeito de sazonalidade ser comum a todas. A respeito dos fatores que determinam a qualidade da Ãgua dos ecossistemas, existe um padrÃo de poluiÃÃo por parte dos ecossistemas, sempre determinados pelo aporte de (MO), representados pelos esgotos domÃsticos e poluiÃÃo difusa. Essa degradaÃÃo à acentuada em dois perÃodos distintos: nas primeiras chuvas e no perÃodo de estiagem, confirmando a sazonalidade caracterÃstica da regiÃo Nordeste. Assim, à interessante propor que as coletas poderiam ser realizadas semestralmente, ao invÃs de mensal. TambÃm sÃo necessÃrias medidas que controle o aporte de (MO), principalmente, esgotos domÃsticos, jà que a DBO5, DQO, nitrito, nitrogÃnio amoniacal total e fÃsforo total comprometeram significativamente nos resultados propostos pelas componentes principais. Com relaÃÃo ao IET calculado, observa-se que 100% dos pontos de amostragem das lagoas foram classificados como hipereutrÃficos. Analisando o comportamento das variÃveis: DBO5, pH, OD e fÃsforo total com a Res. CONAMA n 357/05 observa-se que quase na totalidade dos casos os parÃmetros estiveram fora dos padrÃes legais. Sendo assim, estes ecossistemas precisam està numa perspectiva de melhoria da qualidade, com uma eficiente intervenÃÃo do poder pÃblico para que a comunidade possa usufrui destes ambientes / This study was to evaluate the quality water of the main lake ecosystems belonging the watershed the river Cocà - CE (Lagoon Porangabussu, Opaia, Itaperaoba, Maria Vieira, Sapiranga, Maraponga and Lake JacareÃ). Were collected from water in three sections of ecosystems, the period of August/2006 and July/2007, monthly. The variable temperature, turbidity, pH, conductivity, BOD5, COD, DO, TSS, TDS, total ammonia, nitrite, nitrate, phosphorus, alkalinity, hardness and CTT were analyzed. The data were validated, standardized (Z scores) and referred to the program The Unscrambler version 7.5, the Principal Component Analysis - PCA to reduce the data, facilitating the determination of homogeneity: between lagoons and between sampling points in each lagoon. Also evaluate the role of seasonality in water quality of ecosystems. The principal component analysis (PCA) was satisfactory in the reduction of information from the database, making them easier to understand. For the homogeneity between ecosystems only Sapiranga lagoon and Lake Jacareà presented as the other heterogeneous. This is possible mainly due to the influence of the imposed tidal lagoon Sapiranga and lake Jacareà your little fluid intake. The homogeneity is possible according to the input of organic matter, use and occupancy of their areas of influence and the effect of seasonality is common to all. Regarding the factors that determine the quality of water ecosystems, we can say that there is a pattern of pollution by ecosystems. Always determined by the input of organic matter, represented by domestic sewage and diffuse pollution. This degradation is enhanced by two distinct periods: the first rains and the dry season, determining the power of the seasonality that is characteristic of the Northeast. Thus, it is interesting to propose that instead of the collections are made at a monthly frequency, could be performed every six months. Also needed are measures to control the input of organic matter, mainly domestic sewage, since the parameters BOD5, COD, total ammonia and nitrite committed significantly the results offered by the principal components. With regard to the EIT calculated, it is noted that 100% of the sampling ponds were classified as hypereutrophic. Analyzing the variables: BOD5, pH, DO and total phosphorus with CONAMA n  357/05 is observed that in almost all cases the parameters were outside the legal standards. Thus, the necessity causes these ecosystems are part of a view to improving the quality of its waters by an efficient public intervention and presence in the community who enjoys these environments.

Seizures containing large numbers of units of narcotics, goods dangerous to health and doping are often sent to the Swedish National Forensic Centre (NFC). Only a fraction of these capsules or tablets can be analyzed, therefore the samples need to represent the whole seizure. If the samples show content variations, Near Infrared (NIR) spectroscopy in combination with hyperspectral imaging has been shown to be a promising tool to gauge the homogeneity in the seizures based on chemical content. The objective of this thesis was to further develop and then validate a method for the separation of pregabalin and gabapentin capsules using NIR hyperspectral imaging and Principal Component Analysis (PCA). Capsules containing different amounts of pregabalin and gabapentin were prepared and analyzed. Additionally, authentic seizures were analyzed to confirm that the method fulfilled its purpose. The result of this study showed that use of hyperspectral data in the wavelength range 1650-1750 nm gave the best differentiation between pregabalin and gabapentin capsules. Capsules containing the ratio 70-30 % gabapentin and pregabalin could be separated distinctively from capsules containing pure gabapentin. Multiple authentic seizures could be separated into groups correctly depending on the capsules or tablets content.

The goal of this thesis is to use machine learning methods for datasets of scanned faces and to create a program that allows to explore and edit faces represented as triangle meshes with a number of controls. Firstly we had to reduce dimension of triangle meshes by PCA and then we tried to predict shape of meshes according to physical properties like weight, height, age and BMI. The modeled faces can be used in animation or games.

Current two dimensional face recognition methods rely on visible photometric or geometric attributes that are present in the intensity image. In many of these approaches a technique called Principal Component Analysis (PCA) is extensively used. PCA extracts the maximum intensity variations from the set of input images in the form of "eigen" faces which are used as a feature vector. In these approaches the intensity images used were mostly that of the subject's frontal face, which yielded promising results after doing PCA. These approaches however fail in the presence of facial expression, unstable lighting conditions and artifacts such as make-up, glasses etc. Thus, it is desirable to establish a new biometric source that will be least affected bythe afore mentioned factors. This study describes a face recognition method that is designed based on the consideration of anatomical and biomechanical characteristics of facial tissues. During facial expressions such as smile, frown, anger etc, various muscles get activated in tandem. A strain pattern inferred from a face expression can reveal an individual's signature associated with the underlying anatomical structure, and thus has the potential for face recognition. In this study, the strain is computed by measuring the displacement of a point on the face that results from a facial expression such as opening the mouth. The information provided by the change in the depth value for the face across the open and close mouth frames does not provide any information required for computing the strain maps, because the strain map depends on the relative displacements of two points on the face, which remains same with rigid motions of the face such as rotation and translation. Hence the information in the 2D spaceis sufficient to compute strain since the depth is assumed constant. The approach used to calculate strain computes the strain distribution directly using the mathematical definition of strain as the derivative of displacement in 2D space (XY plane). The strain values obtained are converted to gray scale intensity images, which are used as inputs for the intensity based PCA analysis. Experiments were conducted using 62 subjects. The data set comprised of two pairs of images for a subject: closed mouth and open mouth under bright and low light. Analysis of CMC and ROC curves indicate that the proposed strain map biometric is a promising new biometric that has the potential to improve the performance of current face recognition method. In summary, the contribution of this thesis is twofold: 1. Facial strain map proves to be promising new biometric. 2. Strain map helps increase the identification rate when used in conjunction with intensity based biometric as a multi-classifier.

<p>This thesis is a part of a project where a new method, the base descriptor approach, is studied. The purpose of this method is to reduce drift and extract vital information from electronic tongue measurements. Reference solutions, called descriptors, are measured and the measurements are used to find base descriptors. A base descriptor is, in this thesis, a regression vector for prediction of the property that the descriptor represent. The property is in this case the concentration of a chemical substance in the descriptor solution. Measurements from test samples, in this case fruit juices, are projected onto the base descriptors to extract vital and drift-stable information from the test samples. </p><p>The base descriptors are used to determine the concentrations of the descriptors'chemical substances in the juices and thereby also to classify the different juices. It is assumed that the measurements of samples of juices and descriptors drift the same way. This assumption has to be true in order for the base descriptor approach to work. The base descriptors are calculated by multivariate regression methods like partial least squares regression (PLSR) and principal component regression (PCR). </p><p>Only two of the descriptors tested in this thesis worked as basis for base descriptors. The base descriptors'predictions of the concentrations of chemical substances in the juices are hard to evaluate since the true concentrations are unknown. Comparing the projections of juice measurements onto the base descriptors with a classification model on the juice measurements performed by principal component analysis (PCA), there is no significant difference in drift of the juice measurements in the results of the two methods. The base descriptors, however, separates the juices for classification somewhat better than the classification of juices performed by PCA.</p>

<p>The purpose of this thesis was to examine the demand for cross directional control of basis weight on a board machine. To analyse the demand, changes made by the control system are studied. The significant changes were expected to be present when a major event occurred on the machine. The events classified as major were changes in basis weight, of grade or of coating blade. Break of board and stoppage of the machine were also included. These events can be seen as large disturbances to the machine. </p><p>In order to identify the disturbances a methodology had to be developed. The methodology developed is to analyse the output from a model with the actuators of the control system as input and measurement of basis weight as output. The analysis of this output was done using the multivariate method of principal component analysis. </p><p>The data analysed in this thesis was collected on-line from a board machine operating within the Stora Enso group. Over a period of 3 months, a total of 47 sets of data were collected, each set representing 12-14 hours. The data analysis shows that the variations in the control system are greater than the variation in the measured basis weight. This is a strong indication that the control system is needed and in order to find disturbances in the cross directional profile it is not enough only to analyse the final product, the control signals also have to be analysed. </p><p>The large disturbances do not necessarily emerge from the major events as assumed. Other causes might havelarger impact to the process then first believed. </p><p>One of the major obstacles in trying to explain the variations is that the basis weight is controlled by using the centre layer of the board but measured on the final product. This leads to the fact that the errors seen by the measuring system can result from anything on the machine and be compensated by basis weight in the centre layer of the board.</p>