Books by Leonardo Emmendörfer
Papers by Leonardo Emmendörfer
Blur and other undesirable visual artifacts are often produced from simple interpolation methods ... more Blur and other undesirable visual artifacts are often produced from simple interpolation methods when applied to magnification from low-resolution into higher resolution images. This work evaluates the effect of existing sharpening filters when applied to images obtained from interpolation in order to determine whether image quality can be improved. Additionally, a novel sharpening filter is proposed which uses information from the interpolation process to prevent the occurrence of blur. The novel filter is shown to achieve better results when compared to other sharpening filters from the literature.
Springer eBooks, 2013
Inference of Genetic Regulatory Networks from sparse and noisy expression data is still a challen... more Inference of Genetic Regulatory Networks from sparse and noisy expression data is still a challenge nowadays. In this work we use an Estimation of Distribution Algorithm to infer Genetic Regulatory Networks. In order to evaluate the algorithm we apply it to three types of data: (i) data simulated from a multivariate Gaussian distribution, (ii) data simulated from a realistic simulator, GeneNetWeaver and (iii) data from flow cytometry experiments. The proposed inference method shows a performance comparable with traditional inference algorithms in terms of the network reconstruction accuracy.
Neste trabalho, uma formulação nodal é proposta para o tratamento de uma classe de problemas de t... more Neste trabalho, uma formulação nodal é proposta para o tratamento de uma classe de problemas de transporte de nêutrons, em geometria cartesiana bidimensional. Pelo processo de integração, equações unidimensionais são obtidas, reescrevendo o modelo em termos de quantidades médias. A resolução das equações integradas é feita usando uma versão do método de Ordenadas Discretas Analítico (ADO), onde também são obtidas soluções explicitas, analíticas em termos das variáveis espaciais, através de um código de fácil implementação. Pode-se destacar também como vantagens desta formulação a versatilidade na escolha da quadratura e o baixo custo computacional, uma vez que esquemas iterativos não são necessários, tampouco a subdivisão do domínio em células. Para lidar como os termos do contorno que surgem no processo, utiliza-se aqui uma representação por constantes, de forma que as equações nas variáveis x e y são tratadas por um sistema acoplado. Resultados obtidos por esta formulação são apresentados, bem como alguns perfis de fluxo escalar. Palavras-chave: Ordenadas discretas, métodos nodais, problema de transporte de nêutrons bidimensional. In this paper, a nodal formulation is proposed for the treatment of a class of neutron transport problems in two-dimensional Cartesian geometry. By the integration process, one-dimensional equations are obtained, rewriting the model in terms of average quantities. The resolution of the integrated equations is made using a version of the Analytical Discrete Ordinate method (ADO), where also be obtained explicit solutions, analytical in terms of spatial variables, through an easy implementation code. It can also highlight as advantages of this formulation the versatility of the quadrature choice and the low computational cost, since iterative schemes are not needed either subdividing the domain in cells. To deal with the contour terms that arise in the process, is used here a representation by constants, so that the equations in the variables x and y are treated by a coupled system. Results obtained by this formulation are presented, as well as some profiles of scalar flux.
Energy, Dec 1, 2015
Abstract The SA (Simulated Annealing) is a meta-heuristic to combinatorial and optimization probl... more Abstract The SA (Simulated Annealing) is a meta-heuristic to combinatorial and optimization problems. The main purpose here is to investigate several cooling schedules (main parameter of SA) combined with Constructal Design for geometric optimization of an isothermal Y-shaped cavity intruded into a solid conducting wall with internal heat generation. This shape was optimized with Exhaustive Search and Genetic Algorithm in previous studies of literature. The problem has four degrees of freedom ( H / L , L 1 / L 0 , t 1 / t 0 , α) and two constraints: the cavity and auxiliary fraction cavity areas (ϕ and ψ). Five different cooling schedules are evaluated: Fast, Exponential, Boltz, BoltzExp and ConstExp1. The first three methods are default, while the others are hybrid proposed here. Firstly, it is evaluated the percentage of simulations in which the global optimal point is achieved for the case with square solid wall. Algorithms hybrid and Fast lead to the highest and lowest reliabilities to find global optimal geometries. Posteriorly, a complete optimization of the Y-shaped cavity evaluating the effect of constraints ϕ and ψ over four times optimized Y-shaped cavity are performed. Results are compared with those previously obtained with GA (Genetic Algorithm). The combination of the best cooling schedules improves even more the achievement of the global optimal shapes.
The Simulated Annealing is a widely used metaheuristic to combinatorial and optimization problems... more The Simulated Annealing is a widely used metaheuristic to combinatorial and optimization problems studied since 1983, when it was first proposed. This algorithm has a particular parameter, the cooling schedule (specifically the decrease of temperature) which is still defined for each problem empirically. This study has the aim to find the best cooling schedule(s) for geometric optimization of a heat transfer problem. More specifically it is performed the geometrical evaluation of an isothermal Yshaped cavity intruded into conducting solid wall with internal heat generation. In this research, Constructal Design is employed to determine the constraints and objectives for geometric evaluation of this problem. The cooling schedule(s) recommended here will contribute for future applications of Simulated Annealing heuristic in association with Constructal Design to seek for the optimal shapes of complex geometries in heat transfer problems, as well as, evaluate the influence of geometric parameters over thermal performance of the problem. Six different methods of decrease of temperature are evaluated in twenty samples of thirty simulations for each cooling schedule. From these simulations, it was extracted the percentage of convergence to the global optimal point. Results showed that the Fast cooling schedule led to the worst performance in the seek for the global optimal shapes, while the three hybrid cooling schedules proposed here are the most recommended to find the optimal geometrical configurations in heat transfer problems in association with Constructal Design (BoltzExp, ConstExp1 and ConstExp2).
Jornal Brasileiro De Pneumologia, May 3, 2023
Engenharia térmica, Jun 30, 2015
The problem study here is concerned with the geometrical evaluation of an isothermal Y-shaped cav... more The problem study here is concerned with the geometrical evaluation of an isothermal Y-shaped cavity intruded into conducting solid wall with internal heat generation. The cavity acts as a sink of the heat generated into the solid. The main purpose here is to minimize the maximal excess of temperature (θmax) in the solid. Constructal Design, which is based on the objective and constraints principle, is employed to evaluate the geometries of Y-shaped cavity. Meanwhile, Simulated Annealing (SA) algorithm is employed as optimization method to seek for the best shapes. To validate the SA methodology, the results obtained with SA are compared with those achieved with Genetic Algorithm (GA) and Exaustive Search (ES) in recent studies of literature. The comparison between the optimization methods (SA, GA and ES) showed that Simulated Annealing is highly effective in the search for the optimal shapes of the studied case.
Nowadays, applications have an increasingly degree of interactivity between the computer and the ... more Nowadays, applications have an increasingly degree of interactivity between the computer and the users. A great part of them are multimedia applications, and among them are the computer music system. It is the eld that combines musical techniques with computational algorithms. One interesting application of computer music is the Computer-Aided Algorithmic Composition. The composition may be an entire song or just a part of it. This paper presents the Orpheu architecture, which uses the Markov decision process to predict, in real time, an accompaniment for a solo. Also, a learning mechanism was developed to allow the system to learn how musician performs his solos. To validate the system experiments were performed and measured the correctness of the generated accompaniment. In a real scenario the system achieves an average hit rate about 90%.
Ocean and Coastal Research, 2022
Understanding how wave climate variability and its trends change over time are crucial analyses r... more Understanding how wave climate variability and its trends change over time are crucial analyses required to mitigate potential wave-induced impacts and adapt coastal areas to such effects. The long-term trends and breakpoints of the wave energy flux (WEF) and its relationship with teleconnection patterns in southern Brazil were studied using ERA-5 wave reanalysis with validation using a waverider. We determined that the interannual mean WEF (WEFm) and extreme WEF (WEF98) that reaches the southern Brazil have increased over the past four decades, with a increment of 0.063 and 0.17 kW/m/year, respectively 0.63 and 0.29% per year. By the Muggeo method we determined that subperiods with increasing WEFm trends are related to the SSE and SSW quadrants and that these are also the most energetic ones and with the highest annual increment rates of WEFm and WEF98. Our results also suggest that the positive trends observed in interannual WEF values are likely related to the long term transition of cold to warm Atlantic Multidecadal Oscillation (AMO) phase in the western South Atlantic Ocean. From a coastal risk perspective, it is important to understand the relationship between climatic indices and the wave climate to support long-term coastal management policies.
2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Jul 18, 2022
Several generalizations of the Choquet integral have been applied in the Fuzzy Reasoning Method (... more Several generalizations of the Choquet integral have been applied in the Fuzzy Reasoning Method (FRM) of Fuzzy Rule-Based Classification Systems (FRBCS's) to improve its performance. Additionally, to achieve that goal, researchers have searched for new ways to provide more flexibility to those generalizations, by restricting the requirements of the functions being used in their constructions and relaxing the monotonicity of the integral. This is the case of CT-integrals, CC-integrals, CF-integrals, CF1F2-integrals and dCF-integrals, which obtained good performance in classification algorithms, more specifically, in the fuzzy association rule-based classification method for high-dimensional problems (FARC-HD). Thereafter, with the introduction of Choquet integrals based on restricted dissimilarity functions (RDFs) in place of the standard difference, a new generalization was made possible: the d-XChoquet (d-XC) integrals, which are ordered directional increasing functions and, depending on the adopted RDF, may also be a pre-aggregation function. Those integrals were applied in multi-criteria decision making problems and also in a motor-imagery brain computer interface framework. In the present paper, we introduce a new FRM based on the d-XC integral family, analyzing its performance by applying it to 33 different datasets from the literature.
IEEE Transactions on Fuzzy Systems, Dec 1, 2022
Engenharia térmica, Jun 30, 2017
In this work it is presented a numerical study about geometrical evaluation of heat transfer in s... more In this work it is presented a numerical study about geometrical evaluation of heat transfer in solids with volumetric heat generation and complex intruded cavity for cooling the wall by means of Constructal Design. Several cavities with varied shapes have been evaluated in literature, such as IT T-and H-shaped cavities. The purpose here is to evaluate a complex cavity that combines different elemental shapes. More precisely, the resultant cavity is a merge between a H-shaped cavity and a I-shaped one, forming a ramified geometry, which is more expected in the flow between a point and volume in systems with high magnitude. The main purpose is to minimize two times the maximal temperature in the solid domain (T max). Here two degrees of freedom are defined for the cavity: H 1 /L 1 (ratio between the second branch thickness and its length) and H 0 /L 0 (ratio between the third branch height and its thickness) and the area of cavity is the constraint. For each geometrical configuration the heat diffusion equation is solved with the Finite Element Method (FEM). Results showed that differences of until five times between the optimal shapes and the worst ones are achieved, showing the importance of application of Constructal Design in the problem. Moreover, the best performance is achieved when the vertical branches are fully intruded in the solid domain for intermediate lengths of horizontal branch.
HAL (Le Centre pour la Communication Scientifique Directe), Oct 30, 2015
Cutting and packing problems appear in many industrial applications, such as textile, automotive ... more Cutting and packing problems appear in many industrial applications, such as textile, automotive and transportation. This paper presents a constructive heuristic developed for cutting and packing problems, restricted to the case where a number of non-identical ellipses in an irregular polygon. This problem is related to the gem cut from crude material. The 3D Gems technology proposed in [1] aims to maximize the volumetric use of the gem, taking into account the final market value expected, respecting the industrial standards. The optimization of the ellipses inside the polygon is performed by the Hooke e Jeeves method [2]-[4]. Packed ellipses are tangent to the vertexes of a master ellipse, initially positioned at the center of the polygon. Superposition restrictions are imposed and checked at each stage. R 2 transformations are applied, which raise parametric ellipse equations. Results illustrate that the proposed heuristic can be applied to polygons with varying shapes, either convex or concave, due to the type of search which is performed in the interior of the polygon. The constructive heuristic was able to find the optimal solutions for the polygons tested.
2019 8th Brazilian Conference on Intelligent Systems (BRACIS)
Hierarchical agglomerative clustering (HAC) is among the most widely adopted algorithms in unsupe... more Hierarchical agglomerative clustering (HAC) is among the most widely adopted algorithms in unsupervised learning. The choice of linkage criteria highly influences the results of HAC. This paper presents and evaluates two novel linkage criteria for HAC. The evaluation is performed on 12 datasets from the literature. Novel criteria are compared to three reference methods from the literature: single linkage, complete linkage, and average linkage.
Procedings do XV Simpósio Brasileiro de Automação Inteligente, 2021
Fuzzy modeling is frequently used to deal with the problems involving approximate reasoning, such... more Fuzzy modeling is frequently used to deal with the problems involving approximate reasoning, such as classification problems. However, fuzzy membership functions defined in terms of real functions sometimes can not reflect the uncertainty of the domain specialists. Also, when considering the use of fuzzy quantities, we are executing operations through sets defined by real numbers. In this case, the propagation of errors can become important and affect the end result. This problem finds its solution within the concept of interval-valued fuzzy sets. This paper presents a fuzzy reasoning mechanism to be used in interval-valued fuzzy rule based classification systems (IVFRBCSs). For that, we consider different interval-valued overlap indices, constructed using interval-valued overlap functions, developing confidence and support measures, which are generally used to evaluate the degree of certainty or interest of a given association rule. By considering several interval-valued overlap indices, we obtain an ensemble of IVFRBCS. Then penalty functions are used as a consensus method for the decision-making related to the selection of the best class. Resumo: A modelagem fuzzyé frequentemente usada para lidar com problemas que envolvem raciocínio aproximado, como problemas de classificação. No entanto, funções de pertinência difusas definidas em termos de funções reais nem sempre refletem a incerteza de especialistas. Além disso, ao considerar o uso de valores fuzzy, estamos executando operações através de conjuntos definidos por números reais. Neste caso, a propagação de erros pode se tornar importante e afetar o resultado final. Este problema encontra sua solução dentro do conceito de conjuntos fuzzy intervalares. Este artigo apresenta um mecanismo de raciocínio fuzzy a ser usado em sistemas de classificação baseados em regras fuzzy intervalares (IVFRBCSs). Para isso, consideramos diferentesíndices de overlap intervalares desenvolvendo medidas de confiança e suporte, que geralmente são utilizadas para avaliar o grau de certeza ou interesse de uma determinada regra de associação. Considerando váriosíndices de overlap intervalares, obtemos um conjunto de IVFRBCS. Em seguida, as funções de penalidade (penalty) são usadas como um método de consenso para a tomada de decisão relacionadaà seleção da melhor classe.
Artificial Intelligence and Soft Computing, 2021
Intelligent Systems, 2021
Revista de Gestão Social e Ambiental
Purpose: Given the importance of monitoring and managing land cover, especially in countries with... more Purpose: Given the importance of monitoring and managing land cover, especially in countries with continental proportions, such as Brazil. This research aimed to compare two remote sensing image classifier algorithms. Method/design/approach: The article compared the Gaussian Mixture Model and Random Forest classification algorithms, using Landsat 8 image, which was classified in a supervised way, in the Dezetsaka plugin of QGIS. The analysis of the performance of each model was performed using the Kappa index and Total Accuracy. Results and conclusion: The results showed that the Random Forest algorithm was more efficient than the Gaussian Mixture Model. Taking the Kappa Index (K) and Total Accuracy (po), the models obtained the following performances in the classification of classes: the Random Forest Model (K= 0.94 and po= 96.31) and the Gaussian Mixture Model obtained (K=0.85 and po=90.60). Research implications: The results can support the choice of classification method b...
Uploads
Books by Leonardo Emmendörfer
Papers by Leonardo Emmendörfer