Yazar "Ülker, Erkan" seçeneğine göre listele
Listeleniyor 1 - 15 / 15
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe The Approximation of Interval Bezier Curves by a Fuzzy Rule-Based System(2009) Ülker, Erkan; Arslan, AhmetGenerally, we are interested in knots and control points as variables for finding a well Bezier model from many measuring data points. In this paper, we have selected good knots from given distributed data points on the basis of geometrical information such as curvature and bending of the curve. We have developed interval Bezier curves by estimating control points as an interval according to the knots which are selected by fuzzy rule basis. We have compared these curves and distributed data with the least squares error. We introduced, to the user the as interval Bezier curve, the one who has the minimum error toleration with variations in order to obtain most appropriate widths of Interval control points.Öğe Automatic Emotional Expression Analysis from Lip Shapes Using Artificial Neural Networks(IEEE, 2013) Akkoç, Betül; Ülker, Erkan; Arslan, AhmetIn this study, emotional expressions are automatically analyzed from lip form using artificial neural network. Firstly facial region and then mouth area were found for expressions analysis. After these steps, lip shape was found and the parameters that are used for the analysis were automatically obtained. Artificial neural network trained using these parameters. This network was used for expression analysis. Success rate of 86% is achieved for the seven classes mentioned (happiness, sadness, surprise, disgust, anger, fear and disdain) whereas a success rate of 97% is found when three classes (happiness, sadness and surprised) are used.Öğe Automatic emotional expression analysis from lip shapes using artificial neural networks(2013) Akkoç, Betül; Ülker, Erkan; Arslan, AhmetIn this study, emotional expressions are automatically analyzed from lip form using artificial neural network. Firstly facial region and then mouth area were found for expressions analysis. After these steps, lip shape was found and the parameters that are used for the analysis were automatically obtained. Artificial neural network trained using these parameters. This network was used for expression analysis. Success rate of 86% is achieved for the seven classes mentioned (happiness, sadness, surprise, disgust, anger, fear and disdain) whereas a success rate of 97% is found when three classes (happiness, sadness and surprised) are used. © 2013 IEEE.Öğe A Color Image Watermarking Scheme Based on Hybrid Classification Method Particle Swarm Optimization and K-Nearest Neighbor Algorithm(Elsevier Science Bv, 2010) Fındık, Oğuz; Babaoğlu, İsmail; Ülker, ErkanIn this paper a novel robust watermarking technique using particle swarm optimization and k-nearest neighbor algorithm is introduced to protect the intellectual property rights of color images in the spatial domain In the embedding process the color image is separated into non-overlapping blocks and each bit of the binary watermark is embedded into the individual blocks Then in order to extract the embedded watermark features are obtained from watermark embedded blocks using the symmetric cross-shape kernel These features are used to generate two centroids belonging to each binary (1 and 0) value of the watermark implementing particle swarm optimization Subsequently the embedded watermark is extracted by evaluating these centroids utilizing k-nearest neighbor algorithm. According to the test results embedded watermark is extracted successfully even if the watermarked image is exposed to various image processing attacksÖğe A Comparative Analysis of Metaheuristic Approaches for Multidimensional Two-Way Number Partitioning Problem(SPRINGER HEIDELBERG, 2018) Hacıbeyoğlu, Mehmet; Alaykıran, Kemal; Acılar, Ayşe Merve; Tongur, Vahit; Ülker, ErkanIn this study, a novel usage of four metaheuristic approaches Genetic algorithm (GA), Simulated annealing (SA), migrating bird optimization algorithm (MBO) and clonal selection algorithm (CSA) are applied to multidimensional two-way number partitioning problem (MDTWNPP). MDTWNPP is a classical combinatorial NP-hard optimization problem where a set of vectors have more than one coordinate is partitioned into two subsets. The main objective function of MDTWNPP is to minimize the maximum absolute difference between the sums per coordinate of elements. In order to solve this problem, GA is applied with greedy crossover and mutation operators. SA is improved with dual local search mechanism. MBO is specialized as multiple flock migrating birds optimization algorithms. CSA is applied with problem specific hyper mutation process. Furthermore, all instances are solved using an integer linear programming model which was previously presented in the literature. In the experiments, four metaheuristic approaches and integer linear programming model are used to solve 126 datasets with different sizes and coordinates. As a brief result, the GA and SA approaches designed for this problem outperformed all other heuristics and the integer programming model. Both the performance of GA and SA approaches are in a competitive manner where GA and SA yielded the best solution for 56 and 65 out of 125 datasets, respectively.Öğe A Comparison of Feature Selection Models Utilizing Binary Particle Swarm Optimization and Genetic Algorithm in Determining Coronary Artery Disease Using Support Vector Machine(Pergamon-Elsevier Science Ltd, 2010) Babaoğlu, İsmail; Fındık, Oğuz; Ülker, ErkanThe aim of this study is to search the efficiency of binary particle swarm optimization (BPSO) and genetic algorithm (CA) techniques as feature selection models on determination of coronary artery disease (CAD) existence based upon exercise stress testing (EST) data. Also, increasing the classification performance of the classifier is another aim. The dataset having 23 features was obtained from patients who had performed EST and coronary angiography. Support vector machine (SVM) with k-fold cross-validation method is used as the classifier system of CAD existence in both BPSO and CA feature selection techniques. Classification results of feature selection technique using BPSO and CA are compared with each other and also with the results of the whole features using simple SVM model. The results show that feature selection technique using BPSO is more successful than feature selection technique using CA on determining CAD. Also with the new dataset composed by feature selection technique using BPSO, this study reached more accurate values of success on CAD existence research with more little complexity of classifier system and more little classification time compared with whole features used SVM.Öğe A Digital Robust Image Watermarking Against Desynchronization Attacks(Academic Journals, 2010) Fındık, Oğuz; Babaoğlu, İsmail; Ülker, ErkanIn this paper, a robust image watermarking technique using support vector regression (SVR) and particle swarm optimization is introduced to protect intellectual property rights of the gray images in discrete cosine transform domain against a variety of desynchronization attacks. After the division of the original image to 8 x 8 non-overlapping blocks, frequency coefficients of each block are found using discrete cosine transform. Positions of the inputs and output, among the low frequency coefficients which have the significant characteristics of the image, which are used to train SVR are obtained by using particle swarm optimization technique. After SVR is trained using the obtained positions of the inputs and output, watermark embedding and extracting processes are implemented using the trained SVR. Experiments implemented using the optimized coefficients selected among low frequency coefficients show that our watermarking technique has better watermark extracting success after the desynchronization attacks.Öğe Effects of Discretization on Determination of Coronary Artery Disease Using Support Vector Machine(2009) Babaoğlu, İsmail; Fındık, Oğuz; Ülker, ErkanIn this paper, the effect of discretization on determination of coronary artery disease using exercise stress test data by support vector machine classification method is investigated. The study dataset is obtained from cardiology department of Meram faculty of medicine including 480 patients having 23 features. Four classification models are composed. In the first model, the data is classified simply by normalizing it into [-1,1] range. In the second, third and fourth models, the data is classified by employing entropy-MDL, equal width and equal frequency discretization methods on it respectively. Support vector machine is used as the classifier for all classification models. The results show that classification performance of the model implemented by entropy-MDL discretization has the best value.Öğe Improved Socıal Spıder Algorıthm For Mınımızıng Molecular Potentıal Energy Functıon(Selçuk Üniversitesi Mühendislik Fakültesi, 2020) Ülker, ErkanThe social spider algorithm (SSA) is a new heuristic algorithm created on spider behaviors to solve continuous optimization problems. In this study, SSA is used in order to minimize a simplified model of the energy function of the molecule. The Molecular potential energy function problem is one of the most important real-life problems. The Molecular potential energy function problem attempts to predict the 3D structure of a protein. SSA is developed by various techniques (Crossover-mutation and Gbest convergence-silent spider techniques) and SSA is called Improved SSA (ISSA). By these techniques, the exploration and exploitation capabilities of SSA in the continuous search space are improved. The general performances of SSA and ISSA are tested on low-scaled and large-scaled thirteen benchmark functions and obtained results are compared with each other. Wilcoxon signed-rank test is applied to SSA and ISSA results. Then, the general performance of the SSA and ISSA is tested on a simplified model of the molecule for different dimensions. Also, the performance of the ISSA is compared to various state-of-art algorithms in the literature. The results showed the superiority of the performance of ISSA.Öğe A Marriage Honey Bee Optimization Approach to the Asymmetric Traveling Salesman Problem(SILA SCIENCE, 2012) Çelik, Yüksel; Ülker, ErkanIn the travelling salesman problem (TSP), a travelling salesman completes a tour of “n” number of cities by stopping once in each city and completes the tour by returning to his starting point, while minimizing the distance and the cost. The asymmetric travelling salesman problem (ATSP) is the problem in which the cost of travel from city A to B is different from that from B to A. Marriage in Honey Bee Optimisation (MBO) is a meta-heuristic procedure inspired by the mating and insemination process of honey bees. In this study, we seek to use an MBO algorithm for an optimal solution to the ATSP problem, which has previously been solved by different methods. The results of the MBO algorithm for ATSP are compared with Genetic Algorithm (GA), another meta-heuristic method.Öğe A Marriage in Honey Bee Ooptimisation Approach to the Asymmetric Travelling Salesman Problem(2012) Çelik, Yüksel; Ülker, ErkanIn the travelling salesman problem (TSP), a travelling salesman completes a tour of "n" number of cities by stopping once in each city and completes the tour by returning to his starting point, while minimizing the distance and the cost. The asymmetric travelling salesman problem (ATSP) is the problem in which the cost of travel from city A to B is different from that from B to A. Marriage in Honey Bee Optimisation (MBO) is a meta-heuristic procedure inspired by the mating and insemination process of honey bees. In this study, we seek to use an MBO algorithm for an optimal solution to the ATSP problem, which has previously been solved by different methods. The results of the MBO algorithm for ATSP are compared with Genetic Algorithm (GA), another meta-heuristic method.Öğe A New Approach to Rapid 3d Modelling of Rock Mass Structure(Ice Publ, 2010) Turanboy, Alparslan; Ülker, ErkanThe prediction of rock mass behaviour is an important task in many engineering projects, as the behaviour of rock masses can be controlled by the presence of discontinuities. Being able to map the structure of a rock mass is crucial to understanding its potential behaviour. This understanding can positively impact on the safety and efficiency of an engineering project. In this research, rock masses were modelled and analysed using linear mathematical transformations and isometric perspective methods to achieve meaningful three-dimensional results. The rock mass fracture representation is based on explicit modelling of rock faces. The developed model can improve safety and productivity through its application in the determination and analysis of rock mass structure for geological and geotechnical assessment. Based on the methods explained here, a software system was developed for analysing the geometric characteristics of discontinuities in a rock mass. In this model, discontinuities in a rock mass can be visualised both individually and in combination, and cross-sections can be generated at any desired location. In addition, intersection lines between discontinuities can be generated as dip direction vectors. The natural structure attained by using this developed model agrees well with field measurements.Öğe Nurbs Curve Fitting Using Artificial Immune System(Icic International, 2012) Ülker, ErkanNon-Uniform Rational B-spline (NURBS) is an industrial standard for Computer Aided Design (CAD) model data representation. For constructing an CAD model from a physical part by curve modeling and dimensional measure, the NURBS design often results in a multi-objective optimization (MOO) problem which cannot be handled as such by traditional single objective optimization algorithms. For large data, this problem needs to be dealt with non-deterministic optimization algorithms achieving global optimum and at the same time getting to the desired solution in an iterative fashion. In order to find a good NURBS model from large number of data, generally the knots, control points and weights are respected as variables. In this paper, the minimization of the fitting error is aimed in order to find a smooth curve and the optimization of the NURBS weights and the knot vector for curve fitting is worked. The heuristic of Artificial Immune System (AIS) was used as a new methodology. The best model was searched among the candidate models by using the Akaike's Information Criteria (AIC). Numerical examples were given in order to show the efficiency of our method.Öğe Watermarking Schema Using an Artificial Immune System in Spatial Domain(2009) Fındık, Oğuz; Babaoğlu, İsmail; Ülker, ErkanIn this paper, we propose a robust watermarking method for image copyright protection in spatial domain based on artificial immune system (AIS). Our method optimizes robustness and imperceptibility which are known to be inversely proportional to each other. The robustness of our watermark method was tested extensively against attacks by lossy JPEG compression. Performance of proposed method was evaluated by comparing with the genetic watermarking method and also the randomly chosen block based watermark method. The experiments indicate that our method is better at robustness over genetic watermarking method. We have shown that our method improves the robustness and imperceptibility comparing with the randomly chosen block based watermarking.Öğe Yapay zeka teknikleri kullanılarak yüzey modelleme(Selçuk Üniversitesi Fen Bilimleri Enstitüsü, 2007-07-20) Ülker, Erkan; Arslan, AhmetGerçek dünyadaki çoğu nesnenin ve yüzeyin bilgisayar modelinin üretilmesinde farklı yöntemler kullanılmaktadır. Bunlardan biri de uzaydaki dağıtık veri noktalarından (nokta bulutu) 3B yüzeylerin temsilidir. Noktalar; fotogrametrik bir metot, bir lazer tarayıcı aygıt ya da diğer yüzey ölçüm teknikleri kullanılarak elde edilebilir. Yüzey modeli gerçekleştirilmeden önce eğri modelleme işlemleri gerçekleştirilmelidir. Nokta sayısının fazla olması yüksek dereceli formüllerle temsillerin gerçekleştirilmesini zorunlu kılmaktadır. Daha sonra eğrilerin harmanlanması ile yüzeyler modellenir. Nokta bulutlarından veri temsili için endüstri standardı olmuş yüzey türlerinden birisi parametrik (Bezier, B-spline ve Non- Uniform Rational B-spline (NURBS)) yüzeylerdir. Çok sayıda veriden iyi bir parametrik yüzey modeli bulmak için genelde değişkenler olarak düğümler, parametrizasyon, kontrol noktaları ve ağırlıklarla ilgilenilir. Bu çalışmada yapay zeka teknikleri açısından parametrik yüzey değişkenlerinin optimizasyonu problemine yaklaşılmıştır. Önce bilinmiyen bir eğri üzerinde uzandığı yada bu eğriye yakın geçtiği varsayılan düzensiz noktalar kümesinden ilgili eğriyi tahmin eden parametrik değişkenlerin tahmini gerçekleştirilmiştir. Sonra bilinmiyen bir gerçek nesneye yada yüzeye ait olduğu varsayılan düzensiz noktalar kümesinden tahmin edilen eğrilerin harmanlanması ile yüzey tahmini yapılmıştır. Tezde düzgün bir yüzey/eğri bulmak için uydurma hatasının minimizasyonu hedeflenmiştir. Yüzeyin kontrol noktalarının tahmininde kullanılan matematiksel metotlara alternatif olan yeni bir yöntem bulanık mantık kullanılarak önerilmiştir. Eğri veya yüzey noktası ile ilk ve son noktalar girişler olarak kabul edilip ara kontrol noktaları bulanık çıkarım mekanizması ile tahmin edilmiştir. Eğri ve yüzey parametrizasyonunda seçilen noktalara göre parametrizasyon gerçekleştirme işleminin gerçekleştirilmesinde Yapay Sinir Ağları yeni bir yaklaşım olarak kullanılmıştır. Ağın girişleri iki parametre değeri iken çıkışları üç boyutlu nokta verileridir. Düğüm ve ağırlık optimizasyonu işleminde de Genetik Algoritma ve Yapay Bağışıklık Sistemi yeni bir metodoloji olarak kullanılmıştır. Düğüm olarak noktaların seçilip seçilmemelerine göre kromozom ve antikorlar üretilmiştir. Uygunluk fonksiyonu ve duyarlılık tanımlanarak yüzeyi temsil niteliği en fazla olan noktaların düğüm olarak seçilmeleri sağlanmıştır. Hata değerlerinin hesaplanmasında ve karşılaştırılmasında r-kare ve etkin değer hataları hesaplanmıştır. Düğüm ve ağırlık optimizasyonunda Akaike'nin Bilgi Kriteri (AIC) kullanılarak aday modeller arasında en iyi model aranmıştır. Metodların etkinliğini göstermek için sayısal örnekler verilmiştir. Deneysel çalışmaların sonucunda parametrik yüzey modellemede yapay zeka tekniklerinin global optimumu bulmada ve uydurma hatasını minimize etmede literatürdeki metotlara nazaran iyi sonuçlar verdiği, verilerdeki belirsizlik ve gürültüyü azalttığı, hesaplamsal karmaşıklıktan kurtardığı ve daha düzgün yüzeyler ürettiği ispatlanmıştır.
