Descriptor Combination using Firefly Algorithm

Size: px
Start display at page:

Download "Descriptor Combination using Firefly Algorithm"

Transcription

1 Descriptor Combination using Firefly Algorithm J. Matsuoka, A. Mansano, L. Afonso, J. Papa São Paulo State University - UNESP Department of Computing {jekemi, alex.fernandesm, luis.claudio, papa}@fc.unesp.br Abstract Accuracy rate in image classification can be enhanced by combining different types of descriptors as a way to improve the training set. Many works have proposed combining different feature vectors. However, this paper proposes a technique that models the descriptor combination as an optimization problem using evolutionary-based techniques which will compute parameters that increase the separability in the feature space. Among the evolutionary-based techniques in the literature, this work introduces and employs the Firefly algorithm for such purpose which has its performance compared against other evolutionary-based techniques and are assessed by the OPF classifier. Experiments have shown that descriptor combination does improve the accuracy rate for image classification and the proposed technique outperformed single descriptors in wellknown public datasets. I. INTRODUCTION Image classification can provide important information for several applications such as automatic diagnosis in medical systems and target recognition in remote sensing images. In order to perform the classification task, different types of features can be extracted such as shape, texture and color in order to describe images for further recognition. However, each database may require a different type of feature that best fits the sorts of images in that dataset in order to achieve reasonable results, or a set of descriptors can be used and a combination can be performed in that set. Many projects have proposed to combine features by assigning weights to them [1], [2], [3], while other works solve this problem using Linear Discriminant Analysis [4] and Principal Component Analysis [5]. Faria et. al [6] proposed a descriptor combination approach defining it as a pair (D I, I), where DI is an extractor algorithm, and D I the distance function associated to it. This approach combines distance values of single descriptors, resulting in a composite descriptor. It is worth mentioning that many works have evidenced the importance of this kind of modeling, giving the metric space is as much importance as feature space. This paper proposes an extension to Faria et. al s work [6] by introducing the Firefly algorithm for descriptor combination, which has never been employed before. This paper is organized as follows: Section II introduces the concept of descriptor combination. Section III presents the optimization algorithms employed in this work and Section IV describes the OPF classifier. The proposed methodology is presented in Section V and experimental results in Section VI. Finally, Section VII states the conclusions. II. DESCRIPTOR COMBINATION Definition 1. An image Î is a pair (D I, I), where: D I Z 2 is a finite set of pixels, and I : D I D is a function that assigns to each pixel p in D I a vector I(p) of values in some arbitrary space D (for example, D = IR 3 when a color in the RGB system is assigned to a pixel). Definition 2. A simple descriptor (briefly, descriptor) D is defined as a pair (ǫ D,δ D ), where: ǫ D : Î R n is a function, which extracts a feature vector vî from an image Î. δ D : R n R n R is a similarity function that computes the similarity between two images as the inverse of the distance between their corresponding feature vectors. Definition 3. A feature vector vî of an image Î is a point in R n space: vî = (v 1,v 2,...,v n ), where n is the dimension of the vector. Figure 1 illustrates the use of a simple descriptor D to compute the similarity between two images ÎA and ÎB. First, the extraction algorithm ǫ D is used to compute the feature vectors vîa and vîb associated with the images. Next, the similarity function δ D is used to determine the similarity value d between the images. D: Fig. 1. v I^ A ε D I^A d δ D v I^ B ε D I^B Simple descriptor. Definition 4. A composite descriptor ˆD is a pair (D,δ D ), where: D = {D 1,D 2,...,D k } is a set of k pre-defined simple descriptors.

2 δ D is a similarity function which combines the similarity values obtained from each descriptor D i D, i = 1,2,...,k. Figure 2 illustrates the use a composite descriptor ˆD to compute the distance between images ÎA and ÎB. D: δ D1 d 1 d 2 d k δ D2 δ D δ Dk ε D ε 1 D ε 1 D ε 2 D... ε 2 D ε k D k Fig. 2. I ^ A d ^I B Composite descriptor. III. OPTIMIZATION ALGORITHMS In this section we briefly describe the evolutionary-based techniques for descriptor combination addressed in this paper: Firefly Algorithm, Harmony Search and Particle Swarm Optimization. A. Firefly Algorithm As most of the optimization algorithms, the Firefly algorithm (FFA) is another nature-inspired algorithm designed by Yang [7]. The Firefly algorithm is a metaheuristic algorithm based on Lévy flights and firefly s flashing light which is used to attract mating partners, preys and as a protective warning. Also, fireflies can imitate other fireflies flashing pattern in order to attract and eat them. Lévy flights represent the flight patterns of many animals and insects and is used in this algorithm to model the flight of fireflies on their search for an optimum solution. Thus, FFA explores how light intensity is used to attract other fireflies and Lévy flights for optimization problems as particles attract each other and move toward an optimum solution in the Particle Swarm Optimization algorithm. FFA is designed according to three rules: (i) fireflies are unisex so that they attract each other, (ii) attractiveness is proportional to their brightness, in which the less bright moves toward the brighter one, and (iii) the landscape of the objective function affects the brightness. Light intensity and attractiveness are formulated in which attractiveness can be associated to the firefly s brightness and the brightness can be associated to the solution of the problem. However, light intensity is relative to the distance between the firefly source and the firefly that is receiving the light. Yang formulates the light intensity I(r) varying according to the inverse square law: I(r) = I s r2, (1) in which I s is the source s intensity and r the distance. Attractiveness is proportional to the light intensity and defined as: β = β 0 e γr2, (2) where β 0 is the attractiveness at r = 0. So, when the firefly is near the optimum solution its brightness increases and attracts other fireflies which start moving toward the brighter firefly. Their movements in the search space are defined by the Lévy-Flight which is random walks in the space (Algorithm 1). Algorithm 1. FIREFLY ALGORITHM INPUT: OUTPUT: Set of fireflies. Best solution. 1. Generate the initial population of z fireflies. 2. Set light intensity I for each firefly a according to x. 3. For each firefly a = 1:z 4. For each firefly b = 1:z 5. If I a > I b,then 6. Firefly b moves toward a using the Levy-flight. 7. Evaluate the solution and update light intensity. 8. Rank fireflies according to their brightness 9. Return the best solution. In line 2, x stands for the location of the firefly in the space, and fireflies are ranked (Line 8) in order to find the best solution. The distance between two fireflies is their respective Cartesian distance and the movement of a fireflyato a brighter firefly b is: x a = x a +β 0 e γr2 ab (xb x a )+αsign[rand 1 ] Levy, (3) 2 where sign[rand 1 2 ] with rand [0,1] provides a random direction and Levy provides the lenght of the step. Also, the movement is based on the attraction (second term) and a randomization via Lévy-Flight where α is the randomization parameter. B. Harmony Search The Harmony Search algorithm is an evolutionary algorithm based on music composition, considering a process in which musicians improvise to create music [8]. Its main characteristics are the simple concept, a few parameters and speed to find a solution. The idea is to use the process of creating new songs in order to find solutions for optimization problems, in which solutions are modeled as harmonies and each parameter to be optimized is represented as a note. The final solution will be the best harmony computed by the algorithm. The main steps are: 1) Initialize algorithm parameters; 2) Initialize Harmony Memory (HM) with random values;

3 3) Improvise a new harmony to HM; 4) Updates the HM if the new harmony is better than the worst harmony in HM, inserts the new harmony in HM and removes the worst from it; 5) If the stop criteria has not been reached, return to step 3. In Step 3, a new harmony vector x =(x 1,x 2,..., x N ) is generated from the HM based on memory considerations, pitch adjustments, and randomization (music improvisation). As follows: (i) position ( x i ) and (ii) velocity ( v i ). First, we define the number of particles that will be initialized with random values for both velocity and position. Each particle is evaluated to update the local maximum through its fitness function. Finally, the best position is update with respect to the best of the local maximum particles. This process is repeated until some convergence criterion is reached, i.e. number of iterations. The given particle p i has the velocity and position equations at time step t, respectively, given by x j { x j x j φ j { } x j 1,xj 2,...,xj HMS with probability HMCR, with probability (1-HMCR), in which φ j denotes the range of values for variable j, for φ = (φ i,φ 2,...,φ N ). The HMCR is the probability of choosing one value from the historic values stored in the HM, and (1- HMCR) is the probability of randomly choosing one feasible value not limited to those stored in the HM. In order to make it clear, an HMCR= means that 70% of the notes (decision variables) to compose the new harmony h = ( x ) will be picked from HM, and the remaining ones will be randomly generated within the interval φ j. The pitch adjustment for each harmony is often used to improve solutions and to escape from local optimum. This mechanism concerns shifting the neighboring values of some decision variable in the harmony: x j x j +rb, (5) where x j is the note j that composes the new harmony vector, b is an arbitrary distance bandwidth for the continuous design variable, and r is a uniform distribution between 0 and 1. In this paper, we set b = 1. In Step 4, if the new harmony h is better than the worst harmony in the HM, the latter is replaced by this new harmony. Finally, in Step 5, the HS algorithm finishes when it satisfies the stopping criterion. Otherwise, Steps 3 and 4 are repeated in order to improvise a new harmony again. C. Particle Swarm Optimization The Particle Swarm Optimization (PSO) algorithm was proposed by [9] in order to find solutions for optimization problems. The algorithm is based on social behavior dynamics in which each solution for a given problem is represented by particles which fly in the search space. Particle Swarm Optimization (PSO) is an optimization algorithm based on social behavior dynamics [9], that finds a solution in a search space. Each solution is modeled as a particle in the swarm, which has a memory that stores its best local solution (local maxima) and the best global solution (global maxima). Each particle compares its value with the neighborhood based on a objective function, and decides if copies it in order to get the best local and global maxima. The swarm is modeled in a multidimensional space R N, in which each particle p i = ( x i, v i ) R N has two main features: (4) v j i (t+1) = wvj i (t)+c 1r 1 (ˆx i (t) x j i (t))+c 2r 2 ( g x j i (t)) (6) and x j i (t+1) = xj i (t)+vj i (t+1), (7) where w is the inertia weight that controls the interaction between particles, and r 1,r 2 [0,1] are random variables that give the stochastic idea to Particle Swarm Optimization. The variables c 1 and c 2 are used to guide the particles onto good directions. IV. OPTIMUM-PATH FOREST CLASSIFICATION The OPF classifier works by modeling the problem of pattern recognition as a graph partition in a given feature space. The nodes are represented by the feature vectors and the edges connect all pairs of them, defining a full connectedness graph. This kind of representation is straightforward, given that the graph does not need to be explicitly represented, allowing us to save memory. The partition of the graph is carried out by a competition process between some key samples (prototypes), which offer optimum paths to the remaining nodes of the graph. Each prototype sample defines its optimum-path tree (OPT), and the collection of all OPTs defines an optimumpath forest, which gives the name to the classifier [10], [11]. The OPF can be seen as a generalization of the well-known Dijkstra s algorithm to compute optimum paths from a source node to the remaining ones [12]. The main difference relies on the fact that OPF uses a set of source nodes (prototypes) with any smooth path-cost function [13]. In case of Dijkstra s algorithm, a function that summed the arc-weights along a path was applied. In regard to the supervised OPF version addressed here, we have used a function that gives the maximum arcweight along a path, as explained below. Let Z = Z 1 Z 2 Z 3 be a dataset labeled with a function λ, in which Z 1, Z 2 and Z 3 are, respectively, a training, validation and test sets. LetS Z 1 a set of prototype samples. Essentially, the OPF classifier creates a discrete optimum partition of the feature space such that any sample s Z 2 Z 3 can be classified according to this partition. This partition is an optimum path forest (OPF) computed in R n by the Image Foresting Transform (IFT) algorithm [13]. The OPF algorithm may be used with any smooth path-cost function which can group samples with similar properties [13]. Particularly, we used the path-cost function f max, which is

4 computed as follows: { 0 if s S, f max ( s ) = + otherwise f max (π s,t ) = max{f max (π),d(s,t)}, (8) in which d(s,t) means the distance between samples s and t, and a path π is defined as a sequence of adjacent samples. In such a way, we have that f max (π) computes the maximum distance between adjacent samples in π, when π is not a trivial path. The OPF algorithm works with training and testing phases. In the former step, the competition process begins with the prototypes computation. We are interested into finding the elements that fall on the boundary of the classes with different labels. For that purpose, we can compute a Minimum Spanning Tree (MST) over the original graph and then mark as prototypes the connected elements with different labels. Figure 3b displays the MST with the prototypes at the boundary. After that, we can begin the competition process between prototypes in order to build the optimum-path forest, as displayed in Figure 3c. The classification phase is conducted by taking a sample from the test set (blue triangle in Figure 3d) and connecting it to all training samples. The distance to all training nodes are computed and used to weight the edges. Finally, each training node offers to the test sample a cost given by a path-cost function (maximum arc-weight along a path - Equation 8), and the training node that has offered the minimum path-cost will conquer the test sample. This procedure is shown in Figure 3e (a) (b) 0.9 V. DESCRIPTOR COMBINATION USING FIREFLY ALGORITHM Mansano et. al [14] have proposed an extension to Faria s et al. work [6] by introducing a set of parameters β = (β 1,β 2,...,β M ) in order to allow a greater variability of arithmetic computations, which was limited by the linear formulation proposed by Faria et al. [6] (Eq. 9). Equation 10 presents the proposed formulation. δ D = δ D = N α i δ Di, (9) i=1 N i=1 α i δ βi D i, (10) in which 2 α i,β i 2, β i R. The optimization algorithms are responsible for computing the best values for α and β that maximize the accuracy rate over the validation set which is our objective function. As it is known, each agent (particle, harmony or firefly) represents a different solution which changes at the end of each iteration. Each pair (α, β) provides a different training set which is assessed using the validation set. The process of computing new solutions and validation each of them repeats until the number of iterations is reached and the best training set of all computed is saved in order to run the tests. The purpose of validation the training set is to obtain the set that may provide the highest accuracy rates for a given problem. Thus, the methodology uses three sets: (i) training set which will become the final training set, (ii) validation set used to evaluate the descriptor combination parameters at each iteration of the optimization algorithms, and (iii) testing set which is used to assess the final training set. Then, the whole process can be split in two phases: (i) the design phase where the final training set is modeled by means of optimization algorithms which compute the best parameters for the set of descriptors, and (ii) classification phase in which each sample is classified using the training model computed in the previous phase. Figure 4 shows the methodology proposed by Mansano et. al [14] (c) (d) 0.6 (e) Fig. 3. OPF pipeline: (a) complete graph, (b) MST and prototypes bounded, (c) optimum-path forest generated at the final of training step, (d) classification process and (e) the triangle sample is associated to the white circle class. The values above the nodes are their costs after training, and the values above the edges stand for the distance between their corresponding nodes. Fig. 4. Proposed methodology for descriptor combination. Extracted from Mansano et. al [14]

5 VI. EXPERIMENTAL RESULTS Experiments used two well-known public datasets: Corel 1 : It contains 3,906 images labeled in 85 classes, and the number of images per class ranges from 7 to 98 images. Free Photo 2 : A subset of this dataset was employed containing 3,426 images labeled in 9 classes, and the number of images per class ranges from 70 to 854 images. Different types of descriptors were computed for each dataset [15]: Color Autocorrelogram (ACC): descriptor of color representing the space color in 64 bins and 4 values of distance. Border/Interior pixel Classification (BIC): it is a descriptor of color in which color space is represented by 64 bins. Color Coherent Vector (CCV): descriptor of color composed by a combination of two histograms of 64 bins total. Global Color Histogram (GCH): descriptor of color in which the color space was split in 64 bins. Homogeneous Texture Descriptor (HTD): descriptor of texture Local Activity Spectrum (LAS): descriptor of texture in which components are quantized in 4 bins resulting in a histogram of 256 bins. Quantized Compound Change Histogram (QCCH): descriptor of texture composed by 40 bins. There were computed the ACC, BIC, GCH, LAS and QCCH descriptors for Corel dataset; and BIC, CCV, LAS, GCH and HTD for Free Photo dataset [16] The set of descriptors is split in: training, validation and testing sets. Experiments using a single descriptor used just the training and testing sets, and experiments involving combination is performed using the 3 sets where the validation set is employed to evaluate the solution given by the optimization algorithms. The accuracy rate is computed according to Eq. 11. n i=1 Acc = ta i, (11) n in which ta i stands for the accuracy rate of class i and n represents the number of classes. Table I shows the average accuracy rate of 5 rounds of tests for each descriptor and each database using the OPF classifier using a single descriptor. Single descriptor experiments used a training set of 30% of the entire dataset and 50% of the dataset for testing which were randomly generated in each round of test. The purpose of this first experiment is to compare the gain when descriptor combination with optimization algorithms is performed. Results show different performances for each descriptor in each Descriptor/Dataset Corel Free Photo ACC(HTD) 74.27%± %±1.31 LAS 59.75%± %±0.65 BIC 72.67%± %±0.96 GCH 65.82%± %±1.25 QCCH(CCV) 57.38%± %±0.61 TABLE I SINGLE DESCRIPTOR EFFICIENCY. database where most of them had a better performance in the Free Photo database. Experiments combining different descriptors included optimization algorithms which were employed separately in order to assess each algorithm s performance. In this phase, the validation set is included in order to evaluate and improve the training set prior the test step. The parameters of each algorithm were set as shown in Table II where there were used 250 agents (fireflies, particles and harmonies), 100 iterations and parameters with similar functions were set with the same value (e.g. HMCR from HS and w from PSO) in order to make a fair comparison. Technique Parameters FFA α = 0.2, β = 1.0, γ = 1.0 HS HMCR:, PAR: PSO w =, c 1 = 1.6, c 2 = TABLE II PARAMETERS USED IN THE EXPERIMENTS. Since the purpose is to improve accuracy rate in image classification, the OPF classifier was employed to assess each combination computed by each technique and accuracy rates for combinations of 2 and 5 descriptors, in order words, the accuracy rate is the objective function to be maximized. Table III presents the mean accuracy rates for combination of 2 (LAS and GCH, empirically chosen) and 5 descriptors for each technique and dataset. Descriptor/Dataset Corel Free Photo FFA-LAS+GCH 65.82%± %±0.83 HS-LAS+GCH 67.94%± %±3 PSO-LAS+GCH 68.04%± %±1 FFA-ALL 76.02%± %±1.06 HS-ALL 75.34%± %±0.91 PSO-ALL 76.77%± %±2 TABLE III COMPOSITE DESCRIPTOR EFFICIENCY ACCORDING TO THE PROPOSED APPROACH. Comparing the combination of 2 and 5 descriptors, it is visible an improvement of over 8% in Corel database and almost 6% in the Free Photo, considering the best accuracy rates in each scenario. Among the optimization algorithms, PSO has achieved the best accuracy rates in 3 out of 4 situations, although all techniques had similar results. However, comparing performances of single descriptor and descriptor combination using 5 descriptors has shown a slight

6 difference of over 2% for Corel database and almost 1% for Free Photo database. The low gain might be caused due to the usage of a descriptor that has low performance in such configuration of datasets. VII. CONCLUSIONS Many approaches have proposed feature combination in image classification. This paper has presented techniques based on social dynamics to perform descriptor combination using optimization algorithms and Firefly Algorithm is introduced in this context, which is defined by a pair of feature extraction algorithm and a distance function associated with it. The approach consists in a mathematical formulation of a composite descriptor, which is obtained by combining descriptors using PSO, HS and FFA, and the OPF to perform the tests. In this case, the accuracy rate of OPF in a validation set is used as an objective function to be maximized by these techniques. Experiments with color and texture features have shown that combining different descriptors can outperform the sensibility in image classification. [13] A.X. Falcão, J. Stolfi, and R.A. Lotufo, The image foresting transform theory, algorithms, and applications, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, no. 1, pp , [14] A. Mansano, J. A. Matsuoka, L. C S Afonso, J. P. Papa, F. Faria, and R. da S Torres, Improving image classification through descriptor combination, in Graphics, Patterns and Images (SIBGRAPI), th SIBGRAPI Conference on, 2012, pp [15] F. A. Faria, J. A. dos Santos, A. Rocha, and R. S. Torres, Automatic classifier fusion for produce recognition., in SIBGRAPI. 2012, pp , IEEE Computer Society. [16] O.A.B. Penatti, E. Valle, and R.S. Torres, Comparative study of global color and texture descriptors for web image retrieval, J. of Visual Communication and Image Representation, VIII. ACKNOWLEDGMENTS The authors thank FAPESP (Procs. #2011/ , #2012/ and #2009/ ) and CNPq grant #303182/ REFERENCES [1] P. Gehler and S. Nowozin, On feature combination for multiclass object classification, in Proceedings of the 12th International Conference on Computer Vision, 2009, pp [2] J. Hou, B.-P. Zhang, N.-M. Qi, and Y. Yang, Evaluating feature combination in object classification, in Proceedings of the 7th International Conference on Visual Computing, Las Vegas, NV, USA, 2011, pp , Springer-Verlag. [3] D. Okanohara and J. Tsujii, Learning combination features with L1 regularization, in Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the ACL, Stroudsburg, PA, USA, 2009, pp [4] X. Liu, L. Zhang, M. Li, H. Zhang, and D.Wang, Boosting image classification with LDA-based feature combination for digital photograph management, Pattern Recognition, vol. 38, no. 6, pp , June [5] F. Yan and X. Yanming, Image classification based on multi-feature combination and PCA-RBaggSVM, in Proceedings of the IEEE International Conference on Progress in Informatics and Computing, 2010, vol. 2, pp [6] F.A. Faria, J.P. Papa, R.S. Torres, and A.X. Falcão, Multimodal pattern recognition through particle swarm optimization, in Proceedings of the 17th International Conference on Systems, Signals and Image Processing, Rio de Janeiro, Brazil, 2010, pp [7] Xin-She Yang, Firefly algorithm, lévy flights and global optimization, in SGAI Conf., 2009, pp [8] Z. Geem, Novel derivative of harmony search algorithm for discrete design variables, Applied Mathematics and Computation, vol. 199, no. 1, pp , May [9] J. Kennedy and R.C. Eberhart, Swarm Intelligence, M. Kaufman, [10] J. P. Papa, A. X. Falcão, and C. T. N. Suzuki, Supervised pattern classification based on optimum-path forest, International Journal of Imaging Systems and Technology, vol. 19, no. 2, pp , [11] J. P. Papa, A. X. Falcão, V. H. C. Albuquerque, and J. M. R. S. Tavares, Efficient supervised optimum-path forest classification for large datasets, Pattern Recognition, vol. 45, no. 1, pp , [12] E. W. Dijkstra, A note on two problems in connexion with graphs, Numerische Mathematik, vol. 1, pp , 1959.

The movement of the dimmer firefly i towards the brighter firefly j in terms of the dimmer one s updated location is determined by the following equat

The movement of the dimmer firefly i towards the brighter firefly j in terms of the dimmer one s updated location is determined by the following equat An Improved Firefly Algorithm for Optimization Problems Amarita Ritthipakdee 1, Arit Thammano, Nol Premasathian 3, and Bunyarit Uyyanonvara 4 Abstract Optimization problem is one of the most difficult

More information

A Comparative Study of the Application of Swarm Intelligence in Kruppa-Based Camera Auto- Calibration

A Comparative Study of the Application of Swarm Intelligence in Kruppa-Based Camera Auto- Calibration ISSN 2229-5518 56 A Comparative Study of the Application of Swarm Intelligence in Kruppa-Based Camera Auto- Calibration Ahmad Fariz Hasan, Ali Abuassal, Mutaz Khairalla, Amar Faiz Zainal Abidin, Mohd Fairus

More information

A Native Approach to Cell to Switch Assignment Using Firefly Algorithm

A Native Approach to Cell to Switch Assignment Using Firefly Algorithm International Journal of Engineering Inventions ISSN: 2278-7461, www.ijeijournal.com Volume 1, Issue 2(September 2012) PP: 17-22 A Native Approach to Cell to Switch Assignment Using Firefly Algorithm Apoorva

More information

Particle Swarm Optimization applied to Pattern Recognition

Particle Swarm Optimization applied to Pattern Recognition Particle Swarm Optimization applied to Pattern Recognition by Abel Mengistu Advisor: Dr. Raheel Ahmad CS Senior Research 2011 Manchester College May, 2011-1 - Table of Contents Introduction... - 3 - Objectives...

More information

International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 12, December ISSN

International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 12, December ISSN Optimum Path Forest Approach for Image Retrieval Based on Relevance Feedback: A Survey. Dhawale Shrikant Pandurang, Bela Joglekar Research Scholar,Dept. of Information Technology, Assistant Professor,

More information

Comparing Classification Performances between Neural Networks and Particle Swarm Optimization for Traffic Sign Recognition

Comparing Classification Performances between Neural Networks and Particle Swarm Optimization for Traffic Sign Recognition Comparing Classification Performances between Neural Networks and Particle Swarm Optimization for Traffic Sign Recognition THONGCHAI SURINWARANGKOON, SUPOT NITSUWAT, ELVIN J. MOORE Department of Information

More information

A Gaussian Firefly Algorithm

A Gaussian Firefly Algorithm A Gaussian Firefly Algorithm Sh. M. Farahani, A. A. Abshouri, B. Nasiri and M. R. Meybodi Abstract Firefly algorithm is one of the evolutionary optimization algorithms, and is inspired by fireflies behavior

More information

Experimental Study on Bound Handling Techniques for Multi-Objective Particle Swarm Optimization

Experimental Study on Bound Handling Techniques for Multi-Objective Particle Swarm Optimization Experimental Study on Bound Handling Techniques for Multi-Objective Particle Swarm Optimization adfa, p. 1, 2011. Springer-Verlag Berlin Heidelberg 2011 Devang Agarwal and Deepak Sharma Department of Mechanical

More information

A *69>H>N6 #DJGC6A DG C<>C::G>C<,8>:C8:H /DA 'D 2:6G, ()-"&"3 -"(' ( +-" " " % '.+ % ' -0(+$,

A *69>H>N6 #DJGC6A DG C<>C::G>C<,8>:C8:H /DA 'D 2:6G, ()-&3 -(' ( +-   % '.+ % ' -0(+$, The structure is a very important aspect in neural network design, it is not only impossible to determine an optimal structure for a given problem, it is even impossible to prove that a given structure

More information

A Wrapper Approach for Feature Selection based on Bat Algorithm and Optimum-Path Forest

A Wrapper Approach for Feature Selection based on Bat Algorithm and Optimum-Path Forest 1 2 A Wrapper Approach for Feature Selection based on Bat Algorithm and Optimum-Path Forest 3 4 5 6 7 8 9 10 11 12 13 Douglas Rodrigues a, Luís A. M. Pereira a, Rodrigo Y. M. Nakamura a, Kelton A. P. Costa

More information

NEW TRENDS IN MUSICAL GENRE CLASSIFICATION USING OPTIMUM-PATH FOREST

NEW TRENDS IN MUSICAL GENRE CLASSIFICATION USING OPTIMUM-PATH FOREST 12th International Society for Music Information Retrieval Conference (ISMIR 2011) NEW TRENDS IN MUSICAL GENRE CLASSIFICATION USING OPTIMUM-PATH FOREST C. Marques, I. R. Guilherme UNESP - Univ Estadual

More information

AN EFFICIENT BATIK IMAGE RETRIEVAL SYSTEM BASED ON COLOR AND TEXTURE FEATURES

AN EFFICIENT BATIK IMAGE RETRIEVAL SYSTEM BASED ON COLOR AND TEXTURE FEATURES AN EFFICIENT BATIK IMAGE RETRIEVAL SYSTEM BASED ON COLOR AND TEXTURE FEATURES 1 RIMA TRI WAHYUNINGRUM, 2 INDAH AGUSTIEN SIRADJUDDIN 1, 2 Department of Informatics Engineering, University of Trunojoyo Madura,

More information

ARMA MODEL SELECTION USING PARTICLE SWARM OPTIMIZATION AND AIC CRITERIA. Mark S. Voss a b. and Xin Feng.

ARMA MODEL SELECTION USING PARTICLE SWARM OPTIMIZATION AND AIC CRITERIA. Mark S. Voss a b. and Xin Feng. Copyright 2002 IFAC 5th Triennial World Congress, Barcelona, Spain ARMA MODEL SELECTION USING PARTICLE SWARM OPTIMIZATION AND AIC CRITERIA Mark S. Voss a b and Xin Feng a Department of Civil and Environmental

More information

On the Training of Artificial Neural Networks with Radial Basis Function Using Optimum-Path Forest Clustering

On the Training of Artificial Neural Networks with Radial Basis Function Using Optimum-Path Forest Clustering 204 22nd International Conference on Pattern Recognition On the Training of Artificial Neural Networks with Radial Basis Function Using Optimum-Path Forest Clustering Gustavo H Rosa, Kelton A P Costa,

More information

আ ঐ হ ফ. An Enhanced Harmony Search Method for Bangla Handwritten Character Recognition Using Region Sampling

আ ঐ হ ফ. An Enhanced Harmony Search Method for Bangla Handwritten Character Recognition Using Region Sampling An Enhanced Harmony Search Method for Bangla Handwritten Character Recognition Using Region Sampling Ritesh Sarkhel 1, Amit K Saha 1, Nibaran Das 1 1 Computer Science and Engineering Department, Jadavpur

More information

Synthesis of Thinned Planar Concentric Circular Antenna Array using Evolutionary Algorithms

Synthesis of Thinned Planar Concentric Circular Antenna Array using Evolutionary Algorithms IOSR Journal of Electronics and Communication Engineering (IOSR-JECE) e-issn: 2278-2834,p- ISSN: 2278-8735.Volume 10, Issue 2, Ver. II (Mar - Apr.2015), PP 57-62 www.iosrjournals.org Synthesis of Thinned

More information

Comparison of Some Evolutionary Algorithms for Approximate Solutions of Optimal Control Problems

Comparison of Some Evolutionary Algorithms for Approximate Solutions of Optimal Control Problems Australian Journal of Basic and Applied Sciences, 4(8): 3366-3382, 21 ISSN 1991-8178 Comparison of Some Evolutionary Algorithms for Approximate Solutions of Optimal Control Problems Akbar H. Borzabadi,

More information

A semi-supervised learning algorithm for relevance feedback and collaborative image retrieval

A semi-supervised learning algorithm for relevance feedback and collaborative image retrieval Pedronette et al. EURASIP Journal on Image and Video Processing (25) 25:27 DOI 6/s364-5-8-6 RESEARCH Open Access A semi-supervised learning algorithm for relevance feedback and collaborative image retrieval

More information

Experiments with Firefly Algorithm

Experiments with Firefly Algorithm Experiments with Firefly Algorithm Rogério B. Francisco 1,2, M. Fernanda P. Costa 2, Ana Maria A. C. Rocha 3 1 Escola Superior de Tecnologia e Gestão de Felgueiras, 4610-156 Felgueiras, Portugal rbf@estgf.ipp.pt

More information

Cellular Learning Automata-Based Color Image Segmentation using Adaptive Chains

Cellular Learning Automata-Based Color Image Segmentation using Adaptive Chains Cellular Learning Automata-Based Color Image Segmentation using Adaptive Chains Ahmad Ali Abin, Mehran Fotouhi, Shohreh Kasaei, Senior Member, IEEE Sharif University of Technology, Tehran, Iran abin@ce.sharif.edu,

More information

Modified Particle Swarm Optimization

Modified Particle Swarm Optimization Modified Particle Swarm Optimization Swati Agrawal 1, R.P. Shimpi 2 1 Aerospace Engineering Department, IIT Bombay, Mumbai, India, swati.agrawal@iitb.ac.in 2 Aerospace Engineering Department, IIT Bombay,

More information

Firefly Algorithm to Solve Two Dimensional Bin Packing Problem

Firefly Algorithm to Solve Two Dimensional Bin Packing Problem Firefly Algorithm to Solve Two Dimensional Bin Packing Problem Kratika Chandra, Sudhir Singh Department of Computer Science and Engineering, U.P.T.U. Kanpur Institute of Technology, Kanpur, India. Abstract

More information

A New Feature Local Binary Patterns (FLBP) Method

A New Feature Local Binary Patterns (FLBP) Method A New Feature Local Binary Patterns (FLBP) Method Jiayu Gu and Chengjun Liu The Department of Computer Science, New Jersey Institute of Technology, Newark, NJ 07102, USA Abstract - This paper presents

More information

Feature weighting using particle swarm optimization for learning vector quantization classifier

Feature weighting using particle swarm optimization for learning vector quantization classifier Journal of Physics: Conference Series PAPER OPEN ACCESS Feature weighting using particle swarm optimization for learning vector quantization classifier To cite this article: A Dongoran et al 2018 J. Phys.:

More information

Clustering and Classification

Clustering and Classification Clustering and Classification Visual Informatics Laboratory - Institute of Computing - University of Campinas afalcao@ic.unicamp.br Introduction In Sciences and Engineering, many problems require an optimum

More information

Hybrid Particle Swarm-Based-Simulated Annealing Optimization Techniques

Hybrid Particle Swarm-Based-Simulated Annealing Optimization Techniques Hybrid Particle Swarm-Based-Simulated Annealing Optimization Techniques Nasser Sadati Abstract Particle Swarm Optimization (PSO) algorithms recently invented as intelligent optimizers with several highly

More information

Improving the Efficiency of Fast Using Semantic Similarity Algorithm

Improving the Efficiency of Fast Using Semantic Similarity Algorithm International Journal of Scientific and Research Publications, Volume 4, Issue 1, January 2014 1 Improving the Efficiency of Fast Using Semantic Similarity Algorithm D.KARTHIKA 1, S. DIVAKAR 2 Final year

More information

Automatic Colorization of Grayscale Images

Automatic Colorization of Grayscale Images Automatic Colorization of Grayscale Images Austin Sousa Rasoul Kabirzadeh Patrick Blaes Department of Electrical Engineering, Stanford University 1 Introduction ere exists a wealth of photographic images,

More information

Application of Improved Discrete Particle Swarm Optimization in Logistics Distribution Routing Problem

Application of Improved Discrete Particle Swarm Optimization in Logistics Distribution Routing Problem Available online at www.sciencedirect.com Procedia Engineering 15 (2011) 3673 3677 Advanced in Control Engineeringand Information Science Application of Improved Discrete Particle Swarm Optimization in

More information

Meta- Heuristic based Optimization Algorithms: A Comparative Study of Genetic Algorithm and Particle Swarm Optimization

Meta- Heuristic based Optimization Algorithms: A Comparative Study of Genetic Algorithm and Particle Swarm Optimization 2017 2 nd International Electrical Engineering Conference (IEEC 2017) May. 19 th -20 th, 2017 at IEP Centre, Karachi, Pakistan Meta- Heuristic based Optimization Algorithms: A Comparative Study of Genetic

More information

Three-Dimensional Off-Line Path Planning for Unmanned Aerial Vehicle Using Modified Particle Swarm Optimization

Three-Dimensional Off-Line Path Planning for Unmanned Aerial Vehicle Using Modified Particle Swarm Optimization Three-Dimensional Off-Line Path Planning for Unmanned Aerial Vehicle Using Modified Particle Swarm Optimization Lana Dalawr Jalal Abstract This paper addresses the problem of offline path planning for

More information

CHAPTER 6 ORTHOGONAL PARTICLE SWARM OPTIMIZATION

CHAPTER 6 ORTHOGONAL PARTICLE SWARM OPTIMIZATION 131 CHAPTER 6 ORTHOGONAL PARTICLE SWARM OPTIMIZATION 6.1 INTRODUCTION The Orthogonal arrays are helpful in guiding the heuristic algorithms to obtain a good solution when applied to NP-hard problems. This

More information

Fundamentals of Digital Image Processing

Fundamentals of Digital Image Processing \L\.6 Gw.i Fundamentals of Digital Image Processing A Practical Approach with Examples in Matlab Chris Solomon School of Physical Sciences, University of Kent, Canterbury, UK Toby Breckon School of Engineering,

More information

Traffic Signal Control Based On Fuzzy Artificial Neural Networks With Particle Swarm Optimization

Traffic Signal Control Based On Fuzzy Artificial Neural Networks With Particle Swarm Optimization Traffic Signal Control Based On Fuzzy Artificial Neural Networks With Particle Swarm Optimization J.Venkatesh 1, B.Chiranjeevulu 2 1 PG Student, Dept. of ECE, Viswanadha Institute of Technology And Management,

More information

Particle Swarm Optimization Artificial Bee Colony Chain (PSOABCC): A Hybrid Meteahuristic Algorithm

Particle Swarm Optimization Artificial Bee Colony Chain (PSOABCC): A Hybrid Meteahuristic Algorithm Particle Swarm Optimization Artificial Bee Colony Chain (PSOABCC): A Hybrid Meteahuristic Algorithm Oğuz Altun Department of Computer Engineering Yildiz Technical University Istanbul, Turkey oaltun@yildiz.edu.tr

More information

International Journal of Digital Application & Contemporary research Website: (Volume 1, Issue 7, February 2013)

International Journal of Digital Application & Contemporary research Website:   (Volume 1, Issue 7, February 2013) Performance Analysis of GA and PSO over Economic Load Dispatch Problem Sakshi Rajpoot sakshirajpoot1988@gmail.com Dr. Sandeep Bhongade sandeepbhongade@rediffmail.com Abstract Economic Load dispatch problem

More information

Efficient Image Compression of Medical Images Using the Wavelet Transform and Fuzzy c-means Clustering on Regions of Interest.

Efficient Image Compression of Medical Images Using the Wavelet Transform and Fuzzy c-means Clustering on Regions of Interest. Efficient Image Compression of Medical Images Using the Wavelet Transform and Fuzzy c-means Clustering on Regions of Interest. D.A. Karras, S.A. Karkanis and D. E. Maroulis University of Piraeus, Dept.

More information

An improved PID neural network controller for long time delay systems using particle swarm optimization algorithm

An improved PID neural network controller for long time delay systems using particle swarm optimization algorithm An improved PID neural network controller for long time delay systems using particle swarm optimization algorithm A. Lari, A. Khosravi and A. Alfi Faculty of Electrical and Computer Engineering, Noushirvani

More information

Improving accuracy and speed of Optimum-Path Forest classifier using combination of disjoint training subsets

Improving accuracy and speed of Optimum-Path Forest classifier using combination of disjoint training subsets Improving accuracy and speed of Optimum-Path Forest classifier using combination of disjoint training subsets Moacir P. Ponti-Jr. 1 and João P. Papa 2 1 Institute of Mathematical and Computer Sciences,

More information

Effect of the PSO Topologies on the Performance of the PSO-ELM

Effect of the PSO Topologies on the Performance of the PSO-ELM 2012 Brazilian Symposium on Neural Networks Effect of the PSO Topologies on the Performance of the PSO-ELM Elliackin M. N. Figueiredo and Teresa B. Ludermir Center of Informatics Federal University of

More information

A Comparative Study on Nature Inspired Algorithms with Firefly Algorithm

A Comparative Study on Nature Inspired Algorithms with Firefly Algorithm International Journal of Engineering and Technology Volume 4 No. 10, October, 2014 A Comparative Study on Nature Inspired Algorithms with Firefly Algorithm M. K. A. Ariyaratne, T. G. I. Fernando Department

More information

An Approach to Polygonal Approximation of Digital CurvesBasedonDiscreteParticleSwarmAlgorithm

An Approach to Polygonal Approximation of Digital CurvesBasedonDiscreteParticleSwarmAlgorithm Journal of Universal Computer Science, vol. 13, no. 10 (2007), 1449-1461 submitted: 12/6/06, accepted: 24/10/06, appeared: 28/10/07 J.UCS An Approach to Polygonal Approximation of Digital CurvesBasedonDiscreteParticleSwarmAlgorithm

More information

A Modified PSO Technique for the Coordination Problem in Presence of DG

A Modified PSO Technique for the Coordination Problem in Presence of DG A Modified PSO Technique for the Coordination Problem in Presence of DG M. El-Saadawi A. Hassan M. Saeed Dept. of Electrical Engineering, Faculty of Engineering, Mansoura University, Egypt saadawi1@gmail.com-

More information

SIMULTANEOUS COMPUTATION OF MODEL ORDER AND PARAMETER ESTIMATION FOR ARX MODEL BASED ON MULTI- SWARM PARTICLE SWARM OPTIMIZATION

SIMULTANEOUS COMPUTATION OF MODEL ORDER AND PARAMETER ESTIMATION FOR ARX MODEL BASED ON MULTI- SWARM PARTICLE SWARM OPTIMIZATION SIMULTANEOUS COMPUTATION OF MODEL ORDER AND PARAMETER ESTIMATION FOR ARX MODEL BASED ON MULTI- SWARM PARTICLE SWARM OPTIMIZATION Kamil Zakwan Mohd Azmi, Zuwairie Ibrahim and Dwi Pebrianti Faculty of Electrical

More information

ATI Material Do Not Duplicate ATI Material. www. ATIcourses.com. www. ATIcourses.com

ATI Material Do Not Duplicate ATI Material. www. ATIcourses.com. www. ATIcourses.com ATI Material Material Do Not Duplicate ATI Material Boost Your Skills with On-Site Courses Tailored to Your Needs www.aticourses.com The Applied Technology Institute specializes in training programs for

More information

Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization

Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization Richa Agnihotri #1, Dr. Shikha Agrawal #1, Dr. Rajeev Pandey #1 # Department of Computer Science Engineering, UIT,

More information

The Design of Pole Placement With Integral Controllers for Gryphon Robot Using Three Evolutionary Algorithms

The Design of Pole Placement With Integral Controllers for Gryphon Robot Using Three Evolutionary Algorithms The Design of Pole Placement With Integral Controllers for Gryphon Robot Using Three Evolutionary Algorithms Somayyeh Nalan-Ahmadabad and Sehraneh Ghaemi Abstract In this paper, pole placement with integral

More information

Equation to LaTeX. Abhinav Rastogi, Sevy Harris. I. Introduction. Segmentation.

Equation to LaTeX. Abhinav Rastogi, Sevy Harris. I. Introduction. Segmentation. Equation to LaTeX Abhinav Rastogi, Sevy Harris {arastogi,sharris5}@stanford.edu I. Introduction Copying equations from a pdf file to a LaTeX document can be time consuming because there is no easy way

More information

K-Means Clustering Using Localized Histogram Analysis

K-Means Clustering Using Localized Histogram Analysis K-Means Clustering Using Localized Histogram Analysis Michael Bryson University of South Carolina, Department of Computer Science Columbia, SC brysonm@cse.sc.edu Abstract. The first step required for many

More information

arxiv: v1 [math.oc] 19 Dec 2013

arxiv: v1 [math.oc] 19 Dec 2013 arxiv:1312.5667v1 [math.oc] 19 Dec 2013 A Framework for Self-Tuning Optimization Algorithm Xin-She Yang 1, Suash Deb, Martin Loomes 1, Mehmet Karamanoglu 1 1) School of Science and Technology, Middlesex

More information

Object Recognition using Particle Swarm Optimization on Fourier Descriptors

Object Recognition using Particle Swarm Optimization on Fourier Descriptors Object Recognition using Particle Swarm Optimization on Fourier Descriptors Muhammad Sarfraz 1 and Ali Taleb Ali Al-Awami 2 1 Department of Information and Computer Science, King Fahd University of Petroleum

More information

Implementation and Comparison between PSO and BAT Algorithms for Path Planning with Unknown Environment

Implementation and Comparison between PSO and BAT Algorithms for Path Planning with Unknown Environment Implementation and Comparison between PSO and BAT Algorithms for Path Planning with Unknown Environment Dr. Mukesh Nandanwar 1, Anuj Nandanwar 2 1 Assistance Professor (CSE), Chhattisgarh Engineering College,

More information

LEARNING WEIGHTS OF FUZZY RULES BY USING GRAVITATIONAL SEARCH ALGORITHM

LEARNING WEIGHTS OF FUZZY RULES BY USING GRAVITATIONAL SEARCH ALGORITHM International Journal of Innovative Computing, Information and Control ICIC International c 2013 ISSN 1349-4198 Volume 9, Number 4, April 2013 pp. 1593 1601 LEARNING WEIGHTS OF FUZZY RULES BY USING GRAVITATIONAL

More information

Direct Search Firefly Algorithm for Solving Global Optimization Problems

Direct Search Firefly Algorithm for Solving Global Optimization Problems Appl. Math. Inf. Sci. 10, No. 3, 841-860 (2016) 841 Applied Mathematics & Information Sciences An International Journal http://dx.doi.org/10.18576/amis/100304 Direct Search Firefly Algorithm for Solving

More information

Mobile Robot Path Planning in Static Environments using Particle Swarm Optimization

Mobile Robot Path Planning in Static Environments using Particle Swarm Optimization Mobile Robot Path Planning in Static Environments using Particle Swarm Optimization M. Shahab Alam, M. Usman Rafique, and M. Umer Khan Abstract Motion planning is a key element of robotics since it empowers

More information

Face Recognition Using Vector Quantization Histogram and Support Vector Machine Classifier Rong-sheng LI, Fei-fei LEE *, Yan YAN and Qiu CHEN

Face Recognition Using Vector Quantization Histogram and Support Vector Machine Classifier Rong-sheng LI, Fei-fei LEE *, Yan YAN and Qiu CHEN 2016 International Conference on Artificial Intelligence: Techniques and Applications (AITA 2016) ISBN: 978-1-60595-389-2 Face Recognition Using Vector Quantization Histogram and Support Vector Machine

More information

FUZZY FA: A MODIFIED FIREFLY ALGORITHM

FUZZY FA: A MODIFIED FIREFLY ALGORITHM Applied Artificial Intelligence, 28:47 65, 2014 Copyright 2014 Taylor & Francis Group, LLC ISSN: 0883-9514 print/1087-6545 online DOI: 10.1080/08839514.2014.862773 FUZZY FA: A MODIFIED FIREFLY ALGORITHM

More information

Using the Kolmogorov-Smirnov Test for Image Segmentation

Using the Kolmogorov-Smirnov Test for Image Segmentation Using the Kolmogorov-Smirnov Test for Image Segmentation Yong Jae Lee CS395T Computational Statistics Final Project Report May 6th, 2009 I. INTRODUCTION Image segmentation is a fundamental task in computer

More information

Image Segmentation Based on. Modified Tsallis Entropy

Image Segmentation Based on. Modified Tsallis Entropy Contemporary Engineering Sciences, Vol. 7, 2014, no. 11, 523-529 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ces.2014.4439 Image Segmentation Based on Modified Tsallis Entropy V. Vaithiyanathan

More information

A NEW APPROACH TO SOLVE ECONOMIC LOAD DISPATCH USING PARTICLE SWARM OPTIMIZATION

A NEW APPROACH TO SOLVE ECONOMIC LOAD DISPATCH USING PARTICLE SWARM OPTIMIZATION A NEW APPROACH TO SOLVE ECONOMIC LOAD DISPATCH USING PARTICLE SWARM OPTIMIZATION Manjeet Singh 1, Divesh Thareja 2 1 Department of Electrical and Electronics Engineering, Assistant Professor, HCTM Technical

More information

FITTING PIECEWISE LINEAR FUNCTIONS USING PARTICLE SWARM OPTIMIZATION

FITTING PIECEWISE LINEAR FUNCTIONS USING PARTICLE SWARM OPTIMIZATION Suranaree J. Sci. Technol. Vol. 19 No. 4; October - December 2012 259 FITTING PIECEWISE LINEAR FUNCTIONS USING PARTICLE SWARM OPTIMIZATION Pavee Siriruk * Received: February 28, 2013; Revised: March 12,

More information

Robust Shape Retrieval Using Maximum Likelihood Theory

Robust Shape Retrieval Using Maximum Likelihood Theory Robust Shape Retrieval Using Maximum Likelihood Theory Naif Alajlan 1, Paul Fieguth 2, and Mohamed Kamel 1 1 PAMI Lab, E & CE Dept., UW, Waterloo, ON, N2L 3G1, Canada. naif, mkamel@pami.uwaterloo.ca 2

More information

Modified Particle Swarm Optimization with Novel Modulated Inertia for Velocity Update

Modified Particle Swarm Optimization with Novel Modulated Inertia for Velocity Update Modified Particle Swarm Optimization with Novel Modulated Inertia for Velocity Update Abdul Hadi Hamdan #1, Fazida Hanim Hashim #2, Abdullah Zawawi Mohamed *3, W. M. Diyana W. Zaki #4, Aini Hussain #5

More information

Robust Descriptive Statistics Based PSO Algorithm for Image Segmentation

Robust Descriptive Statistics Based PSO Algorithm for Image Segmentation Robust Descriptive Statistics Based PSO Algorithm for Image Segmentation Ripandeep Kaur 1, Manpreet Kaur 2 1, 2 Punjab Technical University, Chandigarh Engineering College, Landran, Punjab, India Abstract:

More information

CHAPTER 6 HYBRID AI BASED IMAGE CLASSIFICATION TECHNIQUES

CHAPTER 6 HYBRID AI BASED IMAGE CLASSIFICATION TECHNIQUES CHAPTER 6 HYBRID AI BASED IMAGE CLASSIFICATION TECHNIQUES 6.1 INTRODUCTION The exploration of applications of ANN for image classification has yielded satisfactory results. But, the scope for improving

More information

Analysis of Directional Beam Patterns from Firefly Optimization

Analysis of Directional Beam Patterns from Firefly Optimization Analysis of Directional Beam Patterns from Firefly Optimization Nicholas Misiunas, Charles Thompson and Kavitha Chandra Center for Advanced Computation and Telecommunications Department of Electrical and

More information

MATH 209, Lab 5. Richard M. Slevinsky

MATH 209, Lab 5. Richard M. Slevinsky MATH 209, Lab 5 Richard M. Slevinsky Problems 1. Say the temperature T at any point (x, y, z) in space is given by T = 4 x y z 2. Find the hottest point on the sphere F = x 2 + y 2 + z 2 100 = 0; We equate

More information

Sketchable Histograms of Oriented Gradients for Object Detection

Sketchable Histograms of Oriented Gradients for Object Detection Sketchable Histograms of Oriented Gradients for Object Detection No Author Given No Institute Given Abstract. In this paper we investigate a new representation approach for visual object recognition. The

More information

AN EFFICIENT HYBRID APPROACH FOR DATA CLUSTERING USING DYNAMIC K-MEANS ALGORITHM AND FIREFLY ALGORITHM

AN EFFICIENT HYBRID APPROACH FOR DATA CLUSTERING USING DYNAMIC K-MEANS ALGORITHM AND FIREFLY ALGORITHM AN EFFICIENT HYBRID APPROACH FOR DATA CLUSTERING USING DYNAMIC K-MEANS ALGORITHM AND FIREFLY ALGORITHM Sundararajan S. 1 and Karthikeyan S. 2 1 Department of MCA, SNS College of Technology, Coimbatore,

More information

Tracking Changing Extrema with Particle Swarm Optimizer

Tracking Changing Extrema with Particle Swarm Optimizer Tracking Changing Extrema with Particle Swarm Optimizer Anthony Carlisle Department of Mathematical and Computer Sciences, Huntingdon College antho@huntingdon.edu Abstract The modification of the Particle

More information

Fuzzy Ant Clustering by Centroid Positioning

Fuzzy Ant Clustering by Centroid Positioning Fuzzy Ant Clustering by Centroid Positioning Parag M. Kanade and Lawrence O. Hall Computer Science & Engineering Dept University of South Florida, Tampa FL 33620 @csee.usf.edu Abstract We

More information

Multicriteria Image Thresholding Based on Multiobjective Particle Swarm Optimization

Multicriteria Image Thresholding Based on Multiobjective Particle Swarm Optimization Applied Mathematical Sciences, Vol. 8, 2014, no. 3, 131-137 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2014.3138 Multicriteria Image Thresholding Based on Multiobjective Particle Swarm

More information

Color Image Segmentation

Color Image Segmentation Color Image Segmentation Yining Deng, B. S. Manjunath and Hyundoo Shin* Department of Electrical and Computer Engineering University of California, Santa Barbara, CA 93106-9560 *Samsung Electronics Inc.

More information

Image Segmentation for Image Object Extraction

Image Segmentation for Image Object Extraction Image Segmentation for Image Object Extraction Rohit Kamble, Keshav Kaul # Computer Department, Vishwakarma Institute of Information Technology, Pune kamble.rohit@hotmail.com, kaul.keshav@gmail.com ABSTRACT

More information

Implementation of Energy Efficient Clustering Using Firefly Algorithm in Wireless Sensor Networks

Implementation of Energy Efficient Clustering Using Firefly Algorithm in Wireless Sensor Networks 014 1 st International Congress on Computer, Electronics, Electrical, and Communication Engineering (ICCEECE014) IPCSIT vol. 59 (014) (014) IACSIT Press, Singapore DOI: 10.7763/IPCSIT.014.V59.1 Implementation

More information

IMPROVING THE PARTICLE SWARM OPTIMIZATION ALGORITHM USING THE SIMPLEX METHOD AT LATE STAGE

IMPROVING THE PARTICLE SWARM OPTIMIZATION ALGORITHM USING THE SIMPLEX METHOD AT LATE STAGE IMPROVING THE PARTICLE SWARM OPTIMIZATION ALGORITHM USING THE SIMPLEX METHOD AT LATE STAGE Fang Wang, and Yuhui Qiu Intelligent Software and Software Engineering Laboratory, Southwest-China Normal University,

More information

Genetic Algorithm for Seismic Velocity Picking

Genetic Algorithm for Seismic Velocity Picking Proceedings of International Joint Conference on Neural Networks, Dallas, Texas, USA, August 4-9, 2013 Genetic Algorithm for Seismic Velocity Picking Kou-Yuan Huang, Kai-Ju Chen, and Jia-Rong Yang Abstract

More information

A PSO-based Generic Classifier Design and Weka Implementation Study

A PSO-based Generic Classifier Design and Weka Implementation Study International Forum on Mechanical, Control and Automation (IFMCA 16) A PSO-based Generic Classifier Design and Weka Implementation Study Hui HU1, a Xiaodong MAO1, b Qin XI1, c 1 School of Economics and

More information

Local Selection for Heuristic Algorithms as a Factor in Accelerating Optimum Search

Local Selection for Heuristic Algorithms as a Factor in Accelerating Optimum Search Local Selection for Heuristic Algorithms as a Factor in Accelerating Optimum Search Danuta Jama Institute of Mathematics Silesian University of Technology Kaszubska 23, 44-100 Gliwice, Poland Email: Danuta.Jama@polsl.pl

More information

Model Parameter Estimation

Model Parameter Estimation Model Parameter Estimation Shan He School for Computational Science University of Birmingham Module 06-23836: Computational Modelling with MATLAB Outline Outline of Topics Concepts about model parameter

More information

Improving Image Segmentation Quality Via Graph Theory

Improving Image Segmentation Quality Via Graph Theory International Symposium on Computers & Informatics (ISCI 05) Improving Image Segmentation Quality Via Graph Theory Xiangxiang Li, Songhao Zhu School of Automatic, Nanjing University of Post and Telecommunications,

More information

MASSIVE PARALLELISM OF HARMONY MEMORY IN N-DIMENSIONAL SPACE

MASSIVE PARALLELISM OF HARMONY MEMORY IN N-DIMENSIONAL SPACE MASSIVE PARALLELISM OF HARMONY MEMORY IN N-DIMENSIONAL SPACE DR. SHAFAATUNNUR HASAN GPU PRINCIPAL RESEARCHER, UTM BIG DATA CENTRE, IBNU SINA INSTITUTE FOR SCIENTIFIC & INDUSTRIAL RESEARCH UNIVERSITI TEKNOLOGI

More information

ASIAN JOURNAL OF CIVIL ENGINEERING (BUILDING AND HOUSING) VOL. 9, NO. 3 (2008) PAGES

ASIAN JOURNAL OF CIVIL ENGINEERING (BUILDING AND HOUSING) VOL. 9, NO. 3 (2008) PAGES ASIAN JOURNAL OF CIVIL ENGINEERING (BUILDING AND HOUSING) VOL. 9, NO. 3 (008) PAGES 5-8 EFFECT OF BEAM SPACING IN THE HARMONY SEARCH BASED OPTIMUM DESIGN OF GRILLAGES F. Erdal *a and M.P. Saka b a Department

More information

International Conference on Modeling and SimulationCoimbatore, August 2007

International Conference on Modeling and SimulationCoimbatore, August 2007 International Conference on Modeling and SimulationCoimbatore, 27-29 August 2007 OPTIMIZATION OF FLOWSHOP SCHEDULING WITH FUZZY DUE DATES USING A HYBRID EVOLUTIONARY ALGORITHM M.S.N.Kiran Kumara, B.B.Biswalb,

More information

FOREST PLANNING USING PSO WITH A PRIORITY REPRESENTATION

FOREST PLANNING USING PSO WITH A PRIORITY REPRESENTATION FOREST PLANNING USING PSO WITH A PRIORITY REPRESENTATION P.W. Brooks and W.D. Potter Institute for Artificial Intelligence, University of Georgia, USA Overview Background: (NIO Project 1 ) PSO -- GA --

More information

Diffusion Wavelets for Natural Image Analysis

Diffusion Wavelets for Natural Image Analysis Diffusion Wavelets for Natural Image Analysis Tyrus Berry December 16, 2011 Contents 1 Project Description 2 2 Introduction to Diffusion Wavelets 2 2.1 Diffusion Multiresolution............................

More information

Hybrid Optimization Coupling Electromagnetism and Descent Search for Engineering Problems

Hybrid Optimization Coupling Electromagnetism and Descent Search for Engineering Problems Proceedings of the International Conference on Computational and Mathematical Methods in Science and Engineering, CMMSE 2008 13 17 June 2008. Hybrid Optimization Coupling Electromagnetism and Descent Search

More information

Package metaheuristicopt

Package metaheuristicopt Title Metaheuristic for Optimization Version 1.0.0 Package metaheuristicopt October 16, 2017 Author Lala Septem Riza [aut, cre], Iip [aut], Eddy Prasetyo Nugroho [aut] Maintainer Lala Septem Riza

More information

Particle Swarm Optimization

Particle Swarm Optimization Dario Schor, M.Sc., EIT schor@ieee.org Space Systems Department Magellan Aerospace Winnipeg Winnipeg, Manitoba 1 of 34 Optimization Techniques Motivation Optimization: Where, min x F(x), subject to g(x)

More information

I How does the formulation (5) serve the purpose of the composite parameterization

I How does the formulation (5) serve the purpose of the composite parameterization Supplemental Material to Identifying Alzheimer s Disease-Related Brain Regions from Multi-Modality Neuroimaging Data using Sparse Composite Linear Discrimination Analysis I How does the formulation (5)

More information

Identifying Layout Classes for Mathematical Symbols Using Layout Context

Identifying Layout Classes for Mathematical Symbols Using Layout Context Rochester Institute of Technology RIT Scholar Works Articles 2009 Identifying Layout Classes for Mathematical Symbols Using Layout Context Ling Ouyang Rochester Institute of Technology Richard Zanibbi

More information

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation Optimization Methods: Introduction and Basic concepts 1 Module 1 Lecture Notes 2 Optimization Problem and Model Formulation Introduction In the previous lecture we studied the evolution of optimization

More information

Segmentation of Images

Segmentation of Images Segmentation of Images SEGMENTATION If an image has been preprocessed appropriately to remove noise and artifacts, segmentation is often the key step in interpreting the image. Image segmentation is a

More information

Sci.Int.(Lahore),28(1), ,2016 ISSN ; CODEN: SINTE 8 201

Sci.Int.(Lahore),28(1), ,2016 ISSN ; CODEN: SINTE 8 201 Sci.Int.(Lahore),28(1),201-209,2016 ISSN 1013-5316; CODEN: SINTE 8 201 A NOVEL PLANT PROPAGATION ALGORITHM: MODIFICATIONS AND IMPLEMENTATION Muhammad Sulaiman 1, Abdel Salhi 2, Eric S Fraga 3, Wali Khan

More information

CHAPTER 4 AN OPTIMIZED K-MEANS CLUSTERING TECHNIQUE USING BAT ALGORITHM

CHAPTER 4 AN OPTIMIZED K-MEANS CLUSTERING TECHNIQUE USING BAT ALGORITHM 63 CHAPTER 4 AN OPTIMIZED K-MEANS CLUSTERING TECHNIQUE USING BAT ALGORITHM This chapter introduces the new algorithm K-Means and Bat Algorithm (KMBA), for identifying the initial centroid of each cluster.

More information

Kyrre Glette INF3490 Evolvable Hardware Cartesian Genetic Programming

Kyrre Glette INF3490 Evolvable Hardware Cartesian Genetic Programming Kyrre Glette kyrrehg@ifi INF3490 Evolvable Hardware Cartesian Genetic Programming Overview Introduction to Evolvable Hardware (EHW) Cartesian Genetic Programming Applications of EHW 3 Evolvable Hardware

More information

Hybrid Approach for Energy Optimization in Wireless Sensor Networks

Hybrid Approach for Energy Optimization in Wireless Sensor Networks ISSN (Online) : 2319-8753 ISSN (Print) : 2347-6710 International Journal of Innovative Research in Science, Engineering and Technology Volume 3, Special Issue 3, March 2014 2014 International Conference

More information

A New Method For Forecasting Enrolments Combining Time-Variant Fuzzy Logical Relationship Groups And K-Means Clustering

A New Method For Forecasting Enrolments Combining Time-Variant Fuzzy Logical Relationship Groups And K-Means Clustering A New Method For Forecasting Enrolments Combining Time-Variant Fuzzy Logical Relationship Groups And K-Means Clustering Nghiem Van Tinh 1, Vu Viet Vu 1, Tran Thi Ngoc Linh 1 1 Thai Nguyen University of

More information

LECTURE 16: SWARM INTELLIGENCE 2 / PARTICLE SWARM OPTIMIZATION 2

LECTURE 16: SWARM INTELLIGENCE 2 / PARTICLE SWARM OPTIMIZATION 2 15-382 COLLECTIVE INTELLIGENCE - S18 LECTURE 16: SWARM INTELLIGENCE 2 / PARTICLE SWARM OPTIMIZATION 2 INSTRUCTOR: GIANNI A. DI CARO BACKGROUND: REYNOLDS BOIDS Reynolds created a model of coordinated animal

More information

Seismic regionalization based on an artificial neural network

Seismic regionalization based on an artificial neural network Seismic regionalization based on an artificial neural network *Jaime García-Pérez 1) and René Riaño 2) 1), 2) Instituto de Ingeniería, UNAM, CU, Coyoacán, México D.F., 014510, Mexico 1) jgap@pumas.ii.unam.mx

More information