An improved support vector machine based on particle swarm optimization in laser ultrasonic defect detection

Size: px
Start display at page:

Download "An improved support vector machine based on particle swarm optimization in laser ultrasonic defect detection"

Transcription

1 A improved support vector machie based o particle swarm optimizatio i laser ultrasoic defect detectio School of Sciece, North Uiversity of Chia, aiyua, Shaxi 35, Chia xut98@63.com,hhp9@63.com,xywag@6.com,43497@qq.com Correspodece should be addressed to Yapig Bai; baiyp666@63.com Abstract: Laser ultrasoic defect detectio ad classificatio has bee widely used i egieerig ad material defect detectio, so detectig ad classifyig the defect targets accurately is sigificat. I order to obtai the higher classificatio accuracy, a improved support vector machie (SVM) based o particle swarm optimizatio algorithm is used as classifier i this paper. o search the optimal parameters of SVM, a ew aget Decreasig Iertia Weight strategy particle swarm optimizatio () algorithm is proposed to determie the optimal parameters for SVM. I additio, to further improve the classificatio accuracy, sparse represetatio is used to extract the target features from the real target echo waveform i experimet. Experimetal results show that the proposed -SVM ca achieve higher classificatio accuracy compared to the commoly -SVM, classical SVM ad BP eural etwork (BPNN) i the laser ultrasoic defect sigals classificatio. Key-words: Laser ultrasoic defect detectio; Classificatio method; Support vector machie; Particle swarm optimizatio; Sparse represetatio. Itroductio Nowadays, with the developmet of moder sciece ad techology, ultrasoic odestructive testig techology has become a multi subect cross egieerig techology. So far, there have bee some researches o the defect detectio ad defect classificatio based o machie learig [-6]. Caoical support vector machie (SVM) ad particle swarm optimizatio optimized support vector machie (-SVM) are applied as classifiers i those studies. Support vector machie proposed by Vapik i 9 [7], is a statistical classificatio method, which based o the structural risk miimizatio approach. It solves the classificatio problem by maximizig the margi of the separatio beside the optimal hyper plae. he kerel fuctio plays a very importat role i the performace of SVM. he iput data is mapped to a higher dimesioal feature space by the kerel fuctio, so that the classificatio problem ca be liearly separable. here are may kids of kerel fuctios, such as liear kerel, polyomial kerel, Gaussia kerel, ad so o. he Gaussia kerel is frequetly used i SVM, due to its excellet oliear classificatio ability. Support vector machies show may uique advatages i solvig small samples, oliear ad high dimesioal patter recogitio, ad to a certai extet, it overcomes the "dimesio disaster" ad "over learig" ad other traditioal difficulties. I additio, today, support vector machies have bee widely cocered, ad have made great progress, due to its solid theoretical foudatio ad a simple ad clear mathematical model. At preset, it has bee successfully applied to solvig patter recogitio, classificatio, approximatio of fuctios, ad time series predictio problems, such as speech recogitio [8], text recogitio [9], target detectio ad recogitio [, ], fault diagosis [], fiacial time series forecast [3], ad so o. However, i those applicatios, the performace of SVM depeds upo the selectio of SVM parameters. I other words, the parameters selectio has great ifluece o the learig ad geeralizatio ability of SVM. herefore, the selectio of the optimal parameters is importat to obtai a excellet performace of SVM. he parameters optimizatio of SVM has gaied great attetios i the past several years. Such as, Shi ad Zhou use grid search method ad geetic algorithm to optimize SVM parameters, ad the study the reliability of blastig vibratio predictio of ope pit miig usig the optimized SVM model, ad achieved good results [4]; Li, Xia, et al, with the combiatio of clusterig method, at coloy algorithm ad support vector machie (SVM) to costruct a efficiet ad reliable classifier, ad the use it to udge whether the etwork access is ormal or ot [5]; E. Avci choose the best subset of features i digital modulatio classificatio with the support vector machie optimized by geetic algorithm [6]. But, grid algorithm method has disadvatages such as E-ISSN: Volume 5, 6

2 computatioally itesive, time cosumig ad low learig accuracy; At coloy algorithm method is iitial pheromoe scarcity, log-time searchig ad local best solutio; Geetic algorithm is operatio complex ad differet issues eed to desig differet crossover or mutatio. So we eed to fid a simpler ad more efficiet optimizatio algorithm to optimize the model parameters. Particle swarm optimizatio () is a ituitive ad easy to implemet algorithm from the swarm itelligece commuity, ad it has bee applied to select the proper parameters of SVM [7-9], but the method is easy to trap ito local optimum ad has a low covergece rate. Focusig o these shortages, the curret focus of improvemet is maily focused o: the update formula of velocity ad particle positio, parameter improvemet, hybrid algorithm ad so o. Differet iertia weight strategies imply differet icremetal chages i velocity per time step which meas exploratio of ew search areas i pursuit of a better solutio. I this paper, a improved iertia weight is proposed. We propose a aget Decreasig Iertia Weight (DIW-) algorithm to get the optimal parameters of SVM. O the other had, feature extractio is a crucial step i laser ultrasoic defect classificatio. I this work, we use sparse represetatio (SR) theory to aalyzig the compoets of received target echo sigals [-], ad describe the characteristics of the target with the sparse coefficiets obtaied. I this paper, the goal is to develop a method of -SVM combied with sparse represetatio-based feature extractio to effective detectio of differet defect categories, i which is used to optimize the parameters of the support vector machie. he structure of the paper is orgaized as follows: Sectio itroduces the basic idea of SVM. I Sectio 3 the basic algorithm ad the improved algorithm based o a taget decreasig iertia weight strategy are itroduced. he optimizatio procedure to the SVM is preseted i Sectio 4. he classificatio model for laser ultrasoic defect is preseted ad the experimetal results are reported i Sectio 5. Fially, the coclusio is preseted i Sectio 6. he classificatio theory of SVM he support vector machie (SVM) is a machie learig method based o statistical learig theory. his ovel learig techique origiated as the priciple of structure risk miimizatio, which performs better tha empirical risk miimizatio utilized by traditioal eural etworks. SVM has bee applicatio as a ew techique for solvig classificatio, approximatio of fuctios, ad time series predictio problems. he aim of SVM is to obtai a optimal hyper plae that ca separate the two class samples as well as maximize the margi of the separatio beside the optimal hyper plae. I order to describe the priciple of SVM, we give a traiig set X x, x, x, Y y, y, y, where x i is the iput of SVM, yi is the output of SVM,the we deote the class A with x i A, yi ad class B with, xi A xi B i.. e yi, xi B I order to get the optimal hyper plae, we ca cosider a equivalet represetatio xi b, if yi xi b, if yi Where is the weight vector adb is the costat, the distace betwee two class is, so maximize is equivalet to miimize Euclidea orm of the weight vector, i.e.to miimize. I additio the decisio fuctio ca be defies as f ( x) sg( w xi b), () his decisio fuctio is used to solve the liear classificatio problem. For the oliear classificatio problem, we ca fid a map that makes the samples x i from low dimesioal space R ito a high dimesioal feature space F, by oliear mappig xi ( xi), the oliear classificatio problem i the low dimesioal space ca be solved as a liear classificatio problem i the high dimesioal feature space. I this method, the decisio fuctio ca be defies as f ( x) sg( ( xi ) b), () Based o the priciple of structural risk miimizatio, the learig process of SVM ca be trasformed ito a covex optimizatio problem. mi( C i ), (3) s. t. ( xi) b i Ad this costraied optimizatio problem ca be trasformed ito a dual problem of as follows max Q( ) i i yi y ( xi ) ( xi ), (4) s. t. y ad C( i,, ) i i i Where i is slack variables, C is a Pealty parameter ad usually it is positive parameter, ad is the i E-ISSN: Volume 5, 6

3 Lagrage multiplier. Here, the decisio fuctio ca be defies as: f ( x) sg( i yi ( xi ) ( xi ) b), (5) By defiig a kerel matrix K such that: K( x, x ) ( x ) ( x ) ( x ) ( x ), (6) i i i rasform ito the followig form: max Q( ) i i yi y K( xi, x ) he decisio fuctio ca be expressed as:, (7) f ( x) sg( i yik( xi, x ) b). (8) Ad the ier product ca be computed by a kerel fuctio i the low dimesioal space without kowig the oliear mappig explicitly. Ay fuctio that meets Mercer s coditio (Vapik 9) ca be used as the kerel fuctio. here are several commo types of kerel fuctio: Liear kerel: K( x, x ) x x ; i i d Polyomial kerel: K( x, x ) ( x x ), where d is a positive iteger; i i Gaussia kerel: K( x, x ) exp( x x ) i i where is the width of the Gaussia kerel. Research idicates that the Gaussia kerel fuctio shows good performace i oliear classificatio problems. herefore, we selected it as the kerel fuctio for SVM i this study. I additio, the parameters selectio has great ifluece o the learig ad geeralizatio ability of SVM. For SVM with Gaussia kerel fuctio, the parameters iclude kerel fuctio parameter ad Pealty parameter C. he value of is relevat to the rage ad width of the iput data space, C is the compromise betwee the structure risk ad the samples, ad its value is closely related to the tolerated error. As algorithm ot oly has strog global search ability ad but also helps to search for the optimum parameters quickly. herefore, we use the proposed DIW- algorithm to optimize parameter ad C i this study, which we call -SVM. of particles, also kow as a swarm, are flow through the D-dimesioal search space of the optimizatio problem. Each particle alters its ow velocity ad positio based o the experieces of the particle itself ad those of other particles i the swarm. I the searchig process, every particle is coected to ad able to share iformatio with every other particle i the swarm ad the swarm commuicatio topology is kow as a global eighborhood described i [4]. his sharig iformatio mechaism keeps the overall cosistecy to get the global solutio for the overall swarm. he swarm cosists of particles; each particle has a velocity vector vi ( vi, vi,, vid ) ad a positio vector xi ( xi, xi,, xid ), where i,,, ;each particle is represeted as a potetial solutio to a problem i a D-dimesioal search space, respectively ld ad u d are the lower ad upper bouds of the dth dimesio of the search space, x [ l, u ], id d d d,,, D. Durig each geeratio,the particles are accelerated toward the particles previous best positio ad the global best positio. Where, the persoal best positio of the ith particle deotes as p ( p, p,, p ), the global best positio of the i i i id swarm deotes as pg ( pg, pg,, pgd ). he ew velocity value is used to calculate the ext positio of the particle i the search space. his process will keep the iteratio util settig the maximum umber of iteratio or a optimal fitess degree is obtaied. he updatig of velocity ad particle positio ca be obtaied by usig the followig formula: t t t t t t v v c ( p x ) c ( p x ), (9) id id id id gd id x x v. () t t t id id id where c ad c are Learig Factor (also amed as the Acceleratio Coefficiet) ad they re positive costats, ad are radom umbers ragig from to, t is the iteratio couter, is the iertia weight o the iterval keepig the memory of the old velocity vector of the same particle. Whe is a costat [3], it ca lead to a static, ad whe is varyig iteratively, it leads to a dyamic. 3 Particle swarm optimizatio algorithm 3. Caoical particle swarm optimizatio algorithm Particle swarm optimizatio () developed by Keedy ad Eberhart i 9 [3], is a evolutioary computatio algorithm. I algorithm, each potetial solutio to the optimizatio problem is treated as a bird, which is also called a particle. he set 3. Proposed iertia weight variat As we kow, i the operatio of the itelliget optimizatio method, the exploratio ability ad the exploitatio ability of the balaced is very importat. I algorithm, the balace of these two kids of ability is realized by the iertia weight. he larger iertia weight is that the particle has a greater speed i their origial directio which ca further i the origial E-ISSN: Volume 5, 6

4 directio, therefore larger iertia weight is advatageous to fid the global best solutio as soo as possible i the early iterative steps, but may lead to miss the global best solutio easily i later iterative steps. I cotrast, the smaller the iertia weight makes the particles iherit a few of the origial directio, so as to fly closer, with better exploitatio capabilities, so smaller iertia weight meas loger time to provide slower updatig for fie tuig a local exploratio. Appropriate iertia weight is helpful to fid the best solutio with the least umber of iterative steps. herefore, how to adust the iertia weight i order to balace the global exploratio ad local exploitatio better has become a problem. o solve this problem, we usually use the followig approach: i the early iterative steps, larger iertia weight is eeded for coarse global exploratio, but i later iteratios iertia weight should decrease for fie tuig the local exploratio. A larger iertia weight facilitates global exploratio ad a smaller iertia weight teds to facilitates local exploratio to fie tue the curret search area. I order to balace the global exploratio ad local exploratio, we preset a ew aget Decreasig Iertia Weight strategy, hereiafter referred to as DIW strategy. I this strategy, the iertia weight is with the icrease of iterative step t accordig to taget decreasig. he proposed iertia weight is determied based o the followig equatio: t ( t) start ( start ed )*ta * 4. () 4 Optimize the parameters of SVM by I this paper, the Gaussia kerel fuctio is used to costruct SVM classificatio model, the the width of the Gaussia kerel ad pealizatio parameter C eed to be determied. he proposed DIW- algorithm is applied to determie the parameters of SVM. he positio vector x of each particle is eeded to be optimized, which represets the width of the Gaussia kerel ad pealizatio parameter C.he k -fold cross validatio (where k 5) method is used to evaluate fitess i this study, ad average classificatio accuracy (CV) is adopted as the fitess fuctio. he fitess fuctio is deoted by formula: k CV acc(), () k t acc( ) (,,, k). (3) I k -fold cross validatio, the traiig set is roughly divided ito k groups, trai (), trai (), trai() k, acc() is the accuracy whe trai( ) as testig set ad other groups as traiig set. is the sample umber of trai( ), t is the correct classificatio umber of trai( ). he iteratio process of the improved -SVM learig algorithm ca be described clearly as follows. Step:Read sample data, ad we use sparse represetatio theory (SR) for feature extractio, the the feature data is divided ito two subsets: oe subset is traiig set; the other subset is testig set. Step: Iitialize : Iitialize the relative parameters, icludig the size of swarm, the boudary of velocity vmax, v mi ad positio x max, x mi, the acceleratio factors c ad c, ad max umber of iterative max. r ad r are the two radom umbers with the rage from to. Iitialize t ; for each particle, select two D -dimesioal vectors radomly to iitialize the velocity ad positio of this particle. Step3: Calculate the fitess value of each particle accordig to formula ()(3).Set the curret positio of each particle as the persoal best fitess p i. he fid the maximum fitess value as the global best fitess p of the whole swarm. g Step4: Update the iertia weight accordig to formula (). Modify the particle velocity v i ad positio x i accordig to formula (9) (). Step5: Recalculate the fitess values of each particle ad modify p i ad p g.for each particle, if the curret fitess value is better tha the previous local best, the set the curret fitess value to be the local best; or keep the previous local best. For the global swarm, if the best value of all curret local best is better tha the previous global best, the update the value of global best; or keep the previous global best. Step6: Check stop coditio. If t, the stop the iteratio ad pg max is the optimal solutio which represets the best parameters for SVM, ad go to. Step7; Otherwise, let t t go to Step4. Step8: Use the optimal SVM to perform classificatio problem. Apply the above- 7 steps util the obtaied optimal solutio, get the optimized parameters, ad the perform classificatio problem. he flow char is as i Figure. E-ISSN: Volume 5, 6

5 I the part of the feature extractio, we use the method of sparse represetatio to extract features which the defect sigals obtaied from the above experimets. he defect sigal waveform after sparse represetatio was as show i Figure 4. Waveform of trasmissio wave Waveform of reflected wave Fig.. Classificatio model based o -SVM 5 Example of applicatio 5. Laser ultrasoic surface acoustic wave defect detectio experimet Laser ultrasoic defect detectio techology is based o the theory of soud ad light effect. Laser irradiatio o the surface of the sample, the ultrasoic sigal cotaiig the measured surface iformatio is geerated by the thermal elastic effect, ad the the defect iformatio is extracted ad detected by detectig the ultrasoic sigal modulated by the defect. 8mm mm mm (a)waveform of trasmissio wave (b) Waveform of reflectio wave Fig.3. he defect sigal waveform Waveform of trasmissio wave (a)waveform of trasmissio wave (b) Waveform of reflectio wave Fig.4. he defect sigal waveform after sparse represetatio Waveform of reflected wave mm 8mm Laser poit probe mm (a)schematic diagram of reflected wave measuremet m m 8 m m 5 m m L a s e P r o b e p o i 8 m m m (b) Schematic diagram of trasmissio wave measuremet Fig.. Experimetal schematic diagram I the experimet, as show i Figure, test sample is *5*8mm alumium plate, we use M ultrasoic probe i the detectio distace of laser poit mm get reflectio wave, detectio distace of laser poit mm get trasmitted wave, samplig frequecy is MHz, samplig poits is. Five groups of experimets were carried out, each of which was repeated five times. he defect sizes were.*.3mm,.*mm,.*.7mm,.*.9mm ad odestructive. his experimet was doe two times, ad two groups of trasmissio wave (tdata ad data) ad two groups of reflected waves (fdata ad fdata) were obtaied. he obtaied waveforms were as show i Figure3. 5. Feature extractio 5.3 Performace aalysis for the proposed model I this sectio, we evaluate the performace of the previously proposed method for laser ultrasoic surface acoustic wave defect sigal classificatio. he features of the received sigals are extracted based o sparse represetatio. he, the sparse represetatio based features are used as the iput of the classifiers. Fially, the proposed -SVM classifier is used to classify ad forecast the target types. We obtaied five kids of defect sigals by experimet, with 5 i each group. herefore, a dataset icludig 5 samples for all defect types is established. All samples i the dataset are divided ito two sets, i which the testig sets icludig 5 samples (Radomly select sample for each defect types) are used to test the accuracy of the classificatio for each model, ad the trai sets icludig samples are used to trai the classificatio model. o maitai a more realistic ivestigatio of this classificatio techique, the traiig ad testig sets do ot iclude ay overlap i data. I the cotext of optimizatio of the parameters of SVM, i this experimet, we first provide the performace compariso betwee the ad the classical as well as other improvemets (L, N, ad E are i the literature [5-7]). he populatio size N for all is, ad the maximum E-ISSN: Volume 5, 6

6 evolutio geeratio is set to. Figure 5 shows the average fitess curve of ad other for fidig the optimal parameters of SVM. From the fitess performaces that are show i Figure 5, it is clearly see that he average fitess curve of is better tha that of other, It idicates that is superior to other i SVM parameters optimizatio. I the optimized SVM, pealty parameter C ad the width of RBF kerel fuctio are optimized by the proposed ad classical ad other improvemets, the adusted parameters with maximal classificatio accuracy rate are selected as the most appropriate parameters. he, the optimal parameters are utilized to trai SVM model. I the caoical SVM model, the parameter C ad are radomly selected to costruct the classificatio model. he, all the samples gaied from experimetal measuremets are adopted to verify the superior classificatio performace of the proposed method compared with that of -optimized support vector machie model (-SVM), caoical SVM model, the back propagatio eural etworks (BPNN) model. Each experimet is carried out times. he experimetal results are showed i ables ad 9 Average Fitess L N E Number of geeratio Average Fitess L N E Number of geeratio E Number of geeratio (a) tdata (b) tdata (c) fdata (d) fdata Fig.5. Fitess performace for the proposed ad others able: Classificatio results compariso of four sets data data Average classificatio accuracy rate (%) -SVM -SVM L-SVM N-SVM E-SVM SVM BPNN tdata tdata fdata fdata able: Classificatio average computig time compariso of four sets data data Average computig time(s) -SVM -SVM L-SVM N-SVM E-SVM SVM BPNN tdata tdata fdata fdata Average Fitess 8 7 L N Average Fitess 9 8 L N E Number of geeratio From ables ad, we ca see that, the -optimized support vector machie model has higher average classificatio accuracy rate tha caoical SVM ad BPNN classificatio algorithmic to laser ultrasoic surface acoustic wave defect sigal classificatio. he average classificatio accuracy rate of the improvemet SVM by is more tha 98.9%. Amog the several -SVM models metioed i this paper, we proposed that -SVM model ot oly has higher classificatio accuracy, but also the computatioal complexity is ot icreased.herefore, the proposed method has more excellet classificatio performace tha these traditioal classificatio algorithmic to laser ultrasoic surface acoustic wave defect sigal classificatio. 6 Coclusio I this paper, a improved support vector machie (SVM) based o particle swarm optimizatio algorithm is used as classifier for laser ultrasoic surface acoustic wave defect sigal classificatio. Ad the classificatio performace of this model is demostrated o four group defect sigals. I the preseted method, a ew aget Decreasig Iertia Weight strategy particle swarm optimizatio () algorithm is proposed to determie the optimal parameters of SVM, ad sparse represetatio is used to extract the target features from the real target echo waveform i experimet. he experimetal data sets are used to evaluate its feasibility ad performace i the classificatio of defect. Experimetal results show that the proposed -SVM has more excellet classificatio compared to the commoly -SVM, E-ISSN: Volume 5, 6

7 classical SVM ad BP eural etwork (BPNN) i the laser ultrasoic defect sigals classificatio. I future research, the proposed method will be made toward a higher efficiet model beyod curret level, ad more attetio should paid to ivestigate the classificatio performace of models o large size of traiig sets oce eough samples are obtaied. Coflict of Iterests he authors declare that there is o coflict of iterests regardig the publicatio of this paper. Ackowledgmet his work is fully supported by the Natioal Natural Sciece Foudatio Grat o.6 of Chia. Refereces [] Y. Wag, Applicatio of Support Vector Machie i Weld Defect Detectio ad Recogitio of X-ray Images, Computer Aided Draftig Desig & Maufacturig, Vol.3,No.3,4,pp.-6. [] D. A. Karras, Improved defect detectio usig support vector machies ad wavelet feature extractio based o vector quatizatio ad SVD techiques, i: Proceedigs of IEEE Iteratioal Coferece o Neural Networks, Vol.3, No.3, 3,pp [3] K. Kurokawa, M. Morimoto, K. Fuii, Heterogeeous Defect Detectio of Metal Surface usig SVM, Ieice echical Report Sigal Processig, Vol.9,No.9,9,pp [4] A. Widodo, B. S. Yag, Applicatio of oliear feature extractio ad support vector machies for fault diagosis of iductio motors, Expert Systems with Applicatios, Vol.33,No.,7,pp.4-5. [5] X.W. Zhag, Y. Q. Dig, Y. Y. Lv, et al. A visio ispectio system for the surface defects of strogly reflected metal based o multi-class SVM, Expert Systems with Applicatios,Vol.38,No.5,, pp [6] F.Y. Zhag, R. J. Liu, Study o the Parts Surface Defect Detectio Method Based o Modifed SVM Algorithm, Applied Mechaics & Materials, Vol.54-54,4, pp [7] V. N. Vapik, he Nature of Statistical Learig heory, Spriger -Verlag, New York, 9. [8] V. Wa, W.M.Campbell, Support vector machies for speaker verificatio ad idetificatio, Neural Networks for Sigal Processig X, IEEE Sigal Processig Society Workshop, Vol.,No.,, pp [9]. Joachims, Learig to Classify ext Usig Support Vector Machies, Methods, heory ad Algorithms, Spriger Berli, Vol.9,No.4,,pp [] P. Lei, J. Wag, P. Guo, et al., Automatic classificatio of radar targets with micro-motios usig etropy segmetatio ad time-frequecy features, AEU - Iteratioal Joural of Electroics ad Commuicatios, Vol.65,No.,, pp [] A. Eryildirim, I. Oara, Pulse Doppler Radar arget Recogitio usig a wo-stage SVM Procedure, IEEE rasactios o Aerospace & Electroics Systems, Vol.47,No.,,pp [] Q. Qi, Z. N. Jiag, K.Feg, et al., A ovel scheme for fault detectio of reciprocatig compressor valves based o basis pursuit, wave matchig ad support vector machie, Measuremet, Vol.45,No.5,,pp [3]C. Huag, L. L. Huag,.. Ha. Fiacial time series forecastig based o wavelet kerel support vector machie, IEEE, Vol.,,pp [4] X. Shi, J. Zhou, Predictio Residetial House's Damage Effect Near Opepit Agaist Blastig Vibratio Based o SVM with Grid Searchig Method/Geetic Algorithm, Joural of Computatioal & heoretical Naosciece, Vol.,No.,, pp [5] Y. Li, J. Xia, S. Zhag, et al., A efficiet itrusio detectio system based o support vector machies ad gradually feature removal method, Expert Systems with Applicatios, Vol.39,No.,,pp [6] E. Avci, Selectig of the optimal feature subset ad kerel parameters i digital modulatio classificatio by usig hybrid geetic algorithm support vector machies: HGASVM, Expert Systems with Applicatios, Vol.36,No.,9, pp [7] X. Wag, J. We J, Y. Zhag, et al., Real estate price forecastig based o SVM optimized by, Optik - Iteratioal Joural for Light ad Electro Optics, Vol.5,No.5,4,pp [8] W.C. Hog, Chaotic particle swarm optimizatio algorithm i a support vector regressio electric load forecastig model, Eergy Coversio ad Maagemet, Vol.5,No.,9,pp.5-7. [9] M. H. Abdulameer, S. N. H. Sheikh Abdullah, Z. A. Othma, Support Vector Machie Based o Adaptive Acceleratio Particle Swarm Optimizatio, Scietific World Joural, Vol.4,No.,4,pp.-4. [] H. Liu, C. Liu, Y. Huag, Adaptive feature extractio usig sparse codig for machiery fault diagosis, Mechaical Systems & Sigal Processig, Vol.5,No.,,pp [] N. Yu,. Qiu, F. Bi, et al., Image Features Extractio ad Fusio Based o Joit Sparse Represetatio, Selected opics i Sigal Processig IEEE Joural of, Vol.5,No.5,,pp []. Guha, Learig Sparse Represetatios for Huma Actio Recogitio, IEEE rasactios o Patter Aalysis & Machie Itelligece, Vol.34,No.8,, pp [3] R. C. Eberhart, J. Keedy, A ew optimizer usig particle swarm theory, i Proceedigs of the 6th iteratioal symposium o micro machie ad huma sciece, Vol.,No.9,9,pp [4] R. M.Che, C. M. Wag, Proect schedulig heuristics-based stadard for task-resource assigmet i heterogeeous grid, Abstract ad Applied Aalysis, Hidawi Publishig Corporatio, Vol.,. [5] Y. Shi, R. Eberhart, A modified particle swarm optimizer, i: Proceedigs of the 998 IEEE Iteratioal Coferece o Evolutioary Computatio(ICEC 98,998,pp [6] A. Chatteree, P. Siarry, Noliear iertia weight variatio for dyamic adaptatio i particle swarm optimizatio, Computers & Operatios Research, Vol.33,No.3,6,pp [7] J. Lu, H. Hu, Y. Bai, Radial Basis Fuctio Neural Network Based o a Improved Expoetial Decreasig Iertia Weight-Particle Swarm Optimizatio Algorithm for AQI Predictio, Abstract ad Applied Aalysis, 4. E-ISSN: Volume 5, 6

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem A Improved Shuffled Frog-Leapig Algorithm for Kapsack Problem Zhoufag Li, Ya Zhou, ad Peg Cheg School of Iformatio Sciece ad Egieerig Hea Uiversity of Techology ZhegZhou, Chia lzhf1978@126.com Abstract.

More information

3D Model Retrieval Method Based on Sample Prediction

3D Model Retrieval Method Based on Sample Prediction 20 Iteratioal Coferece o Computer Commuicatio ad Maagemet Proc.of CSIT vol.5 (20) (20) IACSIT Press, Sigapore 3D Model Retrieval Method Based o Sample Predictio Qigche Zhag, Ya Tag* School of Computer

More information

Performance Comparisons of PSO based Clustering

Performance Comparisons of PSO based Clustering Performace Comparisos of PSO based Clusterig Suresh Chadra Satapathy, 2 Guaidhi Pradha, 3 Sabyasachi Pattai, 4 JVR Murthy, 5 PVGD Prasad Reddy Ail Neeruoda Istitute of Techology ad Scieces, Sagivalas,Vishaapatam

More information

Dynamic Programming and Curve Fitting Based Road Boundary Detection

Dynamic Programming and Curve Fitting Based Road Boundary Detection Dyamic Programmig ad Curve Fittig Based Road Boudary Detectio SHYAM PRASAD ADHIKARI, HYONGSUK KIM, Divisio of Electroics ad Iformatio Egieerig Chobuk Natioal Uiversity 664-4 Ga Deokji-Dog Jeoju-City Jeobuk

More information

ANN WHICH COVERS MLP AND RBF

ANN WHICH COVERS MLP AND RBF ANN WHICH COVERS MLP AND RBF Josef Boští, Jaromír Kual Faculty of Nuclear Scieces ad Physical Egieerig, CTU i Prague Departmet of Software Egieerig Abstract Two basic types of artificial eural etwors Multi

More information

A New Morphological 3D Shape Decomposition: Grayscale Interframe Interpolation Method

A New Morphological 3D Shape Decomposition: Grayscale Interframe Interpolation Method A ew Morphological 3D Shape Decompositio: Grayscale Iterframe Iterpolatio Method D.. Vizireau Politehica Uiversity Bucharest, Romaia ae@comm.pub.ro R. M. Udrea Politehica Uiversity Bucharest, Romaia mihea@comm.pub.ro

More information

Harris Corner Detection Algorithm at Sub-pixel Level and Its Application Yuanfeng Han a, Peijiang Chen b * and Tian Meng c

Harris Corner Detection Algorithm at Sub-pixel Level and Its Application Yuanfeng Han a, Peijiang Chen b * and Tian Meng c Iteratioal Coferece o Computatioal Sciece ad Egieerig (ICCSE 015) Harris Corer Detectio Algorithm at Sub-pixel Level ad Its Applicatio Yuafeg Ha a, Peijiag Che b * ad Tia Meg c School of Automobile, Liyi

More information

Evaluation of Support Vector Machine Kernels for Detecting Network Anomalies

Evaluation of Support Vector Machine Kernels for Detecting Network Anomalies Evaluatio of Support Vector Machie Kerels for Detectig Network Aomalies Prera Batta, Maider Sigh, Zhida Li, Qigye Dig, ad Ljiljaa Trajković Commuicatio Networks Laboratory http://www.esc.sfu.ca/~ljilja/cl/

More information

Euclidean Distance Based Feature Selection for Fault Detection Prediction Model in Semiconductor Manufacturing Process

Euclidean Distance Based Feature Selection for Fault Detection Prediction Model in Semiconductor Manufacturing Process Vol.133 (Iformatio Techology ad Computer Sciece 016), pp.85-89 http://dx.doi.org/10.1457/astl.016. Euclidea Distace Based Feature Selectio for Fault Detectio Predictio Model i Semicoductor Maufacturig

More information

Ones Assignment Method for Solving Traveling Salesman Problem

Ones Assignment Method for Solving Traveling Salesman Problem Joural of mathematics ad computer sciece 0 (0), 58-65 Oes Assigmet Method for Solvig Travelig Salesma Problem Hadi Basirzadeh Departmet of Mathematics, Shahid Chamra Uiversity, Ahvaz, Ira Article history:

More information

A Comparative Study on Modeling Methods for Switched Reluctance Machines

A Comparative Study on Modeling Methods for Switched Reluctance Machines Computer ad Iformatio Sciece Vol. 3, No. 2; May 21 A Comparative Study o Modelig Methods for Switched Reluctace Machies Shouju Sog & Weiguo iu School of Automatio, Northwester Polytechical Uiversity PO

More information

ISSN (Print) Research Article. *Corresponding author Nengfa Hu

ISSN (Print) Research Article. *Corresponding author Nengfa Hu Scholars Joural of Egieerig ad Techology (SJET) Sch. J. Eg. Tech., 2016; 4(5):249-253 Scholars Academic ad Scietific Publisher (A Iteratioal Publisher for Academic ad Scietific Resources) www.saspublisher.com

More information

New Fuzzy Color Clustering Algorithm Based on hsl Similarity

New Fuzzy Color Clustering Algorithm Based on hsl Similarity IFSA-EUSFLAT 009 New Fuzzy Color Clusterig Algorithm Based o hsl Similarity Vasile Ptracu Departmet of Iformatics Techology Tarom Compay Bucharest Romaia Email: patrascu.v@gmail.com Abstract I this paper

More information

Probabilistic Fuzzy Time Series Method Based on Artificial Neural Network

Probabilistic Fuzzy Time Series Method Based on Artificial Neural Network America Joural of Itelliget Systems 206, 6(2): 42-47 DOI: 0.5923/j.ajis.2060602.02 Probabilistic Fuzzy Time Series Method Based o Artificial Neural Network Erol Egrioglu,*, Ere Bas, Cagdas Haka Aladag

More information

Research on Identification Model of Financial Fraud of Listed Company Based on Data Mining Technology

Research on Identification Model of Financial Fraud of Listed Company Based on Data Mining Technology 208 2d Iteratioal Coferece o Systems, Computig, ad Applicatios (SYSTCA 208) Research o Idetificatio Model of Fiacial Fraud of Listed Compay Based o Data Miig Techology Jiaqi Hu, Xiao Che School of Busiess,

More information

Neural Networks A Model of Boolean Functions

Neural Networks A Model of Boolean Functions Neural Networks A Model of Boolea Fuctios Berd Steibach, Roma Kohut Freiberg Uiversity of Miig ad Techology Istitute of Computer Sciece D-09596 Freiberg, Germay e-mails: steib@iformatik.tu-freiberg.de

More information

Study on effective detection method for specific data of large database LI Jin-feng

Study on effective detection method for specific data of large database LI Jin-feng Iteratioal Coferece o Automatio, Mechaical Cotrol ad Computatioal Egieerig (AMCCE 205) Study o effective detectio method for specific data of large database LI Ji-feg (Vocatioal College of DogYig, Shadog

More information

Neuro Fuzzy Model for Human Face Expression Recognition

Neuro Fuzzy Model for Human Face Expression Recognition IOSR Joural of Computer Egieerig (IOSRJCE) ISSN : 2278-0661 Volume 1, Issue 2 (May-Jue 2012), PP 01-06 Neuro Fuzzy Model for Huma Face Expressio Recogitio Mr. Mayur S. Burage 1, Prof. S. V. Dhopte 2 1

More information

Cubic Polynomial Curves with a Shape Parameter

Cubic Polynomial Curves with a Shape Parameter roceedigs of the th WSEAS Iteratioal Coferece o Robotics Cotrol ad Maufacturig Techology Hagzhou Chia April -8 00 (pp5-70) Cubic olyomial Curves with a Shape arameter MO GUOLIANG ZHAO YANAN Iformatio ad

More information

Pattern Recognition Systems Lab 1 Least Mean Squares

Pattern Recognition Systems Lab 1 Least Mean Squares Patter Recogitio Systems Lab 1 Least Mea Squares 1. Objectives This laboratory work itroduces the OpeCV-based framework used throughout the course. I this assigmet a lie is fitted to a set of poits usig

More information

New HSL Distance Based Colour Clustering Algorithm

New HSL Distance Based Colour Clustering Algorithm The 4th Midwest Artificial Itelligece ad Cogitive Scieces Coferece (MAICS 03 pp 85-9 New Albay Idiaa USA April 3-4 03 New HSL Distace Based Colour Clusterig Algorithm Vasile Patrascu Departemet of Iformatics

More information

Introduction. Nature-Inspired Computing. Terminology. Problem Types. Constraint Satisfaction Problems - CSP. Free Optimization Problem - FOP

Introduction. Nature-Inspired Computing. Terminology. Problem Types. Constraint Satisfaction Problems - CSP. Free Optimization Problem - FOP Nature-Ispired Computig Hadlig Costraits Dr. Şima Uyar September 2006 Itroductio may practical problems are costraied ot all combiatios of variable values represet valid solutios feasible solutios ifeasible

More information

A Novel Feature Extraction Algorithm for Haar Local Binary Pattern Texture Based on Human Vision System

A Novel Feature Extraction Algorithm for Haar Local Binary Pattern Texture Based on Human Vision System A Novel Feature Extractio Algorithm for Haar Local Biary Patter Texture Based o Huma Visio System Liu Tao 1,* 1 Departmet of Electroic Egieerig Shaaxi Eergy Istitute Xiayag, Shaaxi, Chia Abstract The locality

More information

x x 2 x Iput layer = quatity of classificatio mode X T = traspositio matrix The core of such coditioal probability estimatig method is calculatig the

x x 2 x Iput layer = quatity of classificatio mode X T = traspositio matrix The core of such coditioal probability estimatig method is calculatig the COMPARATIVE RESEARCHES ON PROBABILISTIC NEURAL NETWORKS AND MULTI-LAYER PERCEPTRON NETWORKS FOR REMOTE SENSING IMAGE SEGMENTATION Liu Gag a, b, * a School of Electroic Iformatio, Wuha Uiversity, 430079,

More information

Designing a learning system

Designing a learning system CS 75 Machie Learig Lecture Desigig a learig system Milos Hauskrecht milos@cs.pitt.edu 539 Seott Square, x-5 people.cs.pitt.edu/~milos/courses/cs75/ Admiistrivia No homework assigmet this week Please try

More information

Optimization for framework design of new product introduction management system Ma Ying, Wu Hongcui

Optimization for framework design of new product introduction management system Ma Ying, Wu Hongcui 2d Iteratioal Coferece o Electrical, Computer Egieerig ad Electroics (ICECEE 2015) Optimizatio for framework desig of ew product itroductio maagemet system Ma Yig, Wu Hogcui Tiaji Electroic Iformatio Vocatioal

More information

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation Improvemet of the Orthogoal Code Covolutio Capabilities Usig FPGA Implemetatio Naima Kaabouch, Member, IEEE, Apara Dhirde, Member, IEEE, Saleh Faruque, Member, IEEE Departmet of Electrical Egieerig, Uiversity

More information

Accuracy Improvement in Camera Calibration

Accuracy Improvement in Camera Calibration Accuracy Improvemet i Camera Calibratio FaJie L Qi Zag ad Reihard Klette CITR, Computer Sciece Departmet The Uiversity of Aucklad Tamaki Campus, Aucklad, New Zealad fli006, qza001@ec.aucklad.ac.z r.klette@aucklad.ac.z

More information

Solving Fuzzy Assignment Problem Using Fourier Elimination Method

Solving Fuzzy Assignment Problem Using Fourier Elimination Method Global Joural of Pure ad Applied Mathematics. ISSN 0973-768 Volume 3, Number 2 (207), pp. 453-462 Research Idia Publicatios http://www.ripublicatio.com Solvig Fuzzy Assigmet Problem Usig Fourier Elimiatio

More information

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON Roberto Lopez ad Eugeio Oñate Iteratioal Ceter for Numerical Methods i Egieerig (CIMNE) Edificio C1, Gra Capitá s/, 08034 Barceloa, Spai ABSTRACT I this work

More information

RESEARCH ON AUTOMATIC INSPECTION TECHNIQUE OF REAL-TIME RADIOGRAPHY FOR TURBINE-BLADE

RESEARCH ON AUTOMATIC INSPECTION TECHNIQUE OF REAL-TIME RADIOGRAPHY FOR TURBINE-BLADE RESEARCH ON AUTOMATIC INSPECTION TECHNIQUE OF REAL-TIME RADIOGRAPHY FOR TURBINE-BLADE Z.G. Zhou, S. Zhao, ad Z.G. A School of Mechaical Egieerig ad Automatio, Beijig Uiversity of Aeroautics ad Astroautics,

More information

Image Segmentation EEE 508

Image Segmentation EEE 508 Image Segmetatio Objective: to determie (etract) object boudaries. It is a process of partitioig a image ito distict regios by groupig together eighborig piels based o some predefied similarity criterio.

More information

Algorithms for Disk Covering Problems with the Most Points

Algorithms for Disk Covering Problems with the Most Points Algorithms for Disk Coverig Problems with the Most Poits Bi Xiao Departmet of Computig Hog Kog Polytechic Uiversity Hug Hom, Kowloo, Hog Kog csbxiao@comp.polyu.edu.hk Qigfeg Zhuge, Yi He, Zili Shao, Edwi

More information

Web Text Feature Extraction with Particle Swarm Optimization

Web Text Feature Extraction with Particle Swarm Optimization 32 IJCSNS Iteratioal Joural of Computer Sciece ad Network Security, VOL.7 No.6, Jue 2007 Web Text Feature Extractio with Particle Swarm Optimizatio Sog Liagtu,, Zhag Xiaomig Istitute of Itelliget Machies,

More information

Redundancy Allocation for Series Parallel Systems with Multiple Constraints and Sensitivity Analysis

Redundancy Allocation for Series Parallel Systems with Multiple Constraints and Sensitivity Analysis IOSR Joural of Egieerig Redudacy Allocatio for Series Parallel Systems with Multiple Costraits ad Sesitivity Aalysis S. V. Suresh Babu, D.Maheswar 2, G. Ragaath 3 Y.Viaya Kumar d G.Sakaraiah e (Mechaical

More information

Journal of Chemical and Pharmaceutical Research, 2013, 5(12): Research Article

Journal of Chemical and Pharmaceutical Research, 2013, 5(12): Research Article Available olie www.jocpr.com Joural of Chemical ad Pharmaceutical Research, 2013, 5(12):745-749 Research Article ISSN : 0975-7384 CODEN(USA) : JCPRC5 K-meas algorithm i the optimal iitial cetroids based

More information

A METHOD OF GENERATING RULES FOR A KERNEL FUZZY CLASSIFIER

A METHOD OF GENERATING RULES FOR A KERNEL FUZZY CLASSIFIER Proceedigs of the Sixth Iteratioal Coferece o Machie Learig ad Cyberetics, Hog Kog, 9- August 7 A METHOD OF GENERATING RULES FOR A KERNEL FUZZY CLASSIFIER AI-MIN YANG,, XIN-GUANG LI, LING-MIN JIANG,YONG-MEI

More information

The measurement of overhead conductor s sag with DLT method

The measurement of overhead conductor s sag with DLT method Advaces i Egieerig Research (AER), volume 7 2d Aual Iteratioal Coferece o Electroics, Electrical Egieerig ad Iformatio Sciece (EEEIS 206) he measuremet of overhead coductor s sag with DL method Fag Ye,

More information

Stone Images Retrieval Based on Color Histogram

Stone Images Retrieval Based on Color Histogram Stoe Images Retrieval Based o Color Histogram Qiag Zhao, Jie Yag, Jigyi Yag, Hogxig Liu School of Iformatio Egieerig, Wuha Uiversity of Techology Wuha, Chia Abstract Stoe images color features are chose

More information

Mobile terminal 3D image reconstruction program development based on Android Lin Qinhua

Mobile terminal 3D image reconstruction program development based on Android Lin Qinhua Iteratioal Coferece o Automatio, Mechaical Cotrol ad Computatioal Egieerig (AMCCE 05) Mobile termial 3D image recostructio program developmet based o Adroid Li Qihua Sichua Iformatio Techology College

More information

An Estimation of Distribution Algorithm for solving the Knapsack problem

An Estimation of Distribution Algorithm for solving the Knapsack problem Vol.4,No.5, 214 Published olie: May 25, 214 DOI: 1.7321/jscse.v4.5.1 A Estimatio of Distributio Algorithm for solvig the Kapsack problem 1 Ricardo Pérez, 2 S. Jös, 3 Arturo Herádez, 4 Carlos A. Ochoa *1,

More information

An Efficient Algorithm for Graph Bisection of Triangularizations

An Efficient Algorithm for Graph Bisection of Triangularizations A Efficiet Algorithm for Graph Bisectio of Triagularizatios Gerold Jäger Departmet of Computer Sciece Washigto Uiversity Campus Box 1045 Oe Brookigs Drive St. Louis, Missouri 63130-4899, USA jaegerg@cse.wustl.edu

More information

A DIGITAL WATERMARKING SCHEME BASED ON ICA DETECTION

A DIGITAL WATERMARKING SCHEME BASED ON ICA DETECTION A DIGITAL WATERMARKING SCHEME BASED ON ICA DETECTION Ju Liu,, Xigag Zhag, Jiade Su ad Miguel Agel Laguas, 3 School of Iformatio Sciece ad Egieerig, Shadog Uiversity, Jia 5, Chia Telecommuicatios Techological

More information

Image based Cats and Possums Identification for Intelligent Trapping Systems

Image based Cats and Possums Identification for Intelligent Trapping Systems Volume 159 No, February 017 Image based Cats ad Possums Idetificatio for Itelliget Trappig Systems T. A. S. Achala Perera School of Egieerig Aucklad Uiversity of Techology New Zealad Joh Collis School

More information

Sectio 4, a prototype project of settig field weight with AHP method is developed ad the experimetal results are aalyzed. Fially, we coclude our work

Sectio 4, a prototype project of settig field weight with AHP method is developed ad the experimetal results are aalyzed. Fially, we coclude our work 200 2d Iteratioal Coferece o Iformatio ad Multimedia Techology (ICIMT 200) IPCSIT vol. 42 (202) (202) IACSIT Press, Sigapore DOI: 0.7763/IPCSIT.202.V42.0 Idex Weight Decisio Based o AHP for Iformatio Retrieval

More information

Analysis of Server Resource Consumption of Meteorological Satellite Application System Based on Contour Curve

Analysis of Server Resource Consumption of Meteorological Satellite Application System Based on Contour Curve Advaces i Computer, Sigals ad Systems (2018) 2: 19-25 Clausius Scietific Press, Caada Aalysis of Server Resource Cosumptio of Meteorological Satellite Applicatio System Based o Cotour Curve Xiagag Zhao

More information

Heuristic Approaches for Solving the Multidimensional Knapsack Problem (MKP)

Heuristic Approaches for Solving the Multidimensional Knapsack Problem (MKP) Heuristic Approaches for Solvig the Multidimesioal Kapsack Problem (MKP) R. PARRA-HERNANDEZ N. DIMOPOULOS Departmet of Electrical ad Computer Eg. Uiversity of Victoria Victoria, B.C. CANADA Abstract: -

More information

Chapter 3 Classification of FFT Processor Algorithms

Chapter 3 Classification of FFT Processor Algorithms Chapter Classificatio of FFT Processor Algorithms The computatioal complexity of the Discrete Fourier trasform (DFT) is very high. It requires () 2 complex multiplicatios ad () complex additios [5]. As

More information

BASED ON ITERATIVE ERROR-CORRECTION

BASED ON ITERATIVE ERROR-CORRECTION A COHPARISO OF CRYPTAALYTIC PRICIPLES BASED O ITERATIVE ERROR-CORRECTIO Miodrag J. MihaljeviC ad Jova Dj. GoliC Istitute of Applied Mathematics ad Electroics. Belgrade School of Electrical Egieerig. Uiversity

More information

Adaptive Resource Allocation for Electric Environmental Pollution through the Control Network

Adaptive Resource Allocation for Electric Environmental Pollution through the Control Network Available olie at www.sciecedirect.com Eergy Procedia 6 (202) 60 64 202 Iteratioal Coferece o Future Eergy, Eviromet, ad Materials Adaptive Resource Allocatio for Electric Evirometal Pollutio through the

More information

Improving Template Based Spike Detection

Improving Template Based Spike Detection Improvig Template Based Spike Detectio Kirk Smith, Member - IEEE Portlad State Uiversity petra@ee.pdx.edu Abstract Template matchig algorithms like SSE, Covolutio ad Maximum Likelihood are well kow for

More information

Investigating methods for improving Bagged k-nn classifiers

Investigating methods for improving Bagged k-nn classifiers Ivestigatig methods for improvig Bagged k-nn classifiers Fuad M. Alkoot Telecommuicatio & Navigatio Istitute, P.A.A.E.T. P.O.Box 4575, Alsalmia, 22046 Kuwait Abstract- We experimet with baggig knn classifiers

More information

Digital Microfluidic Biochips Online Fault Detection Route Optimization Scheme Chuan-pei XU and Tong-zhou ZHAO *

Digital Microfluidic Biochips Online Fault Detection Route Optimization Scheme Chuan-pei XU and Tong-zhou ZHAO * 2018 Iteratioal Coferece o Commuicatio, Network ad Artificial Itelligece (CNAI 2018) ISBN: 978-1-60595-065-5 Digital Microfluidic Biochips Olie Fault Detectio Route Optimizatio Scheme Chua-pei XU ad Tog-zhou

More information

An Efficient Algorithm for Graph Bisection of Triangularizations

An Efficient Algorithm for Graph Bisection of Triangularizations Applied Mathematical Scieces, Vol. 1, 2007, o. 25, 1203-1215 A Efficiet Algorithm for Graph Bisectio of Triagularizatios Gerold Jäger Departmet of Computer Sciece Washigto Uiversity Campus Box 1045, Oe

More information

Position and Velocity Estimation by Ultrasonic Sensor

Position and Velocity Estimation by Ultrasonic Sensor Positio ad Velocity Estimatio by Ultrasoic Sesor N Ramarao 1, A R Subramayam 2, J Chara Raj 2, Lalith B V 2, Varu K R 2 1 (Faculty of EEE, BMSIT & M, INDIA) 2 (Studets of EEE, BMSIT & M, INDIA) Abstract:

More information

Fuzzy Rule Selection by Data Mining Criteria and Genetic Algorithms

Fuzzy Rule Selection by Data Mining Criteria and Genetic Algorithms Fuzzy Rule Selectio by Data Miig Criteria ad Geetic Algorithms Hisao Ishibuchi Dept. of Idustrial Egieerig Osaka Prefecture Uiversity 1-1 Gakue-cho, Sakai, Osaka 599-8531, JAPAN E-mail: hisaoi@ie.osakafu-u.ac.jp

More information

Text Feature Selection based on Feature Dispersion Degree and Feature Concentration Degree

Text Feature Selection based on Feature Dispersion Degree and Feature Concentration Degree Available olie at www.ijpe-olie.com vol. 13, o. 7, November 017, pp. 1159-1164 DOI: 10.3940/ijpe.17.07.p19.11591164 Text Feature Selectio based o Feature Dispersio Degree ad Feature Cocetratio Degree Zhifeg

More information

Outline. Research Definition. Motivation. Foundation of Reverse Engineering. Dynamic Analysis and Design Pattern Detection in Java Programs

Outline. Research Definition. Motivation. Foundation of Reverse Engineering. Dynamic Analysis and Design Pattern Detection in Java Programs Dyamic Aalysis ad Desig Patter Detectio i Java Programs Outlie Lei Hu Kamra Sartipi {hul4, sartipi}@mcmasterca Departmet of Computig ad Software McMaster Uiversity Caada Motivatio Research Problem Defiitio

More information

Chapter 1. Introduction to Computers and C++ Programming. Copyright 2015 Pearson Education, Ltd.. All rights reserved.

Chapter 1. Introduction to Computers and C++ Programming. Copyright 2015 Pearson Education, Ltd.. All rights reserved. Chapter 1 Itroductio to Computers ad C++ Programmig Copyright 2015 Pearso Educatio, Ltd.. All rights reserved. Overview 1.1 Computer Systems 1.2 Programmig ad Problem Solvig 1.3 Itroductio to C++ 1.4 Testig

More information

Application of Decision Tree and Support Vector Machine for Inspecting Bubble Defects on LED Sealing Glue Images

Application of Decision Tree and Support Vector Machine for Inspecting Bubble Defects on LED Sealing Glue Images 66 Applicatio of Decisio Tree ad Support Vector Machie for Ispectig Bubble Defects o LED Sealig Glue Images * Chua-Yu Chag ad Yi-Feg Li Abstract Bubble defect ispectio is a importat step i light-emittig

More information

Lip Contour Extraction Based on Support Vector Machine

Lip Contour Extraction Based on Support Vector Machine Lip Cotour Extractio Based o Support Vector Machie Author Pa, Xiaosheg, Kog, Jiagpig, Liew, Ala Wee-Chug Published 008 Coferece Title CISP 008 : Proceedigs, First Iteratioal Cogress o Image ad Sigal Processig

More information

The identification of key quality characteristics based on FAHP

The identification of key quality characteristics based on FAHP Iteratioal Joural of Research i Egieerig ad Sciece (IJRES ISSN (Olie: 2320-9364, ISSN (Prit: 2320-9356 Volume 3 Issue 6 ǁ Jue 2015 ǁ PP.01-07 The idetificatio of ey quality characteristics based o FAHP

More information

A Note on Least-norm Solution of Global WireWarping

A Note on Least-norm Solution of Global WireWarping A Note o Least-orm Solutio of Global WireWarpig Charlie C. L. Wag Departmet of Mechaical ad Automatio Egieerig The Chiese Uiversity of Hog Kog Shati, N.T., Hog Kog E-mail: cwag@mae.cuhk.edu.hk Abstract

More information

Fuzzy Membership Function Optimization for System Identification Using an Extended Kalman Filter

Fuzzy Membership Function Optimization for System Identification Using an Extended Kalman Filter Fuzzy Membership Fuctio Optimizatio for System Idetificatio Usig a Eteded Kalma Filter Srikira Kosaam ad Da Simo Clevelad State Uiversity NAFIPS Coferece Jue 4, 2006 Embedded Cotrol Systems Research Lab

More information

Designing a learning system

Designing a learning system CS 75 Itro to Machie Learig Lecture Desigig a learig system Milos Hauskrecht milos@pitt.edu 539 Seott Square, -5 people.cs.pitt.edu/~milos/courses/cs75/ Admiistrivia No homework assigmet this week Please

More information

Lecture 13: Validation

Lecture 13: Validation Lecture 3: Validatio Resampli methods Holdout Cross Validatio Radom Subsampli -Fold Cross-Validatio Leave-oe-out The Bootstrap Bias ad variace estimatio Three-way data partitioi Itroductio to Patter Recoitio

More information

An Algorithm to Solve Multi-Objective Assignment. Problem Using Interactive Fuzzy. Goal Programming Approach

An Algorithm to Solve Multi-Objective Assignment. Problem Using Interactive Fuzzy. Goal Programming Approach It. J. Cotemp. Math. Scieces, Vol. 6, 0, o. 34, 65-66 A Algorm to Solve Multi-Objective Assigmet Problem Usig Iteractive Fuzzy Goal Programmig Approach P. K. De ad Bharti Yadav Departmet of Mathematics

More information

An Image Retrieval Method Based on Hu Invariant Moment and Improved Annular Histogram

An Image Retrieval Method Based on Hu Invariant Moment and Improved Annular Histogram http://dx.doi.org/10.5755/j01.eee.0.4.6888 ELEKTROIKA IR ELEKTROTECHIKA ISS 139 115 VOL. 0 O. 4 014 A Image Retrieval Method Based o Hu Ivariat Momet ad Improved Aular Histogram F. Xiag 1 H. Yog 1 S. Dada

More information

Intrusion Detection Method Using Protocol Classification and Rough Set Based Support Vector Machine

Intrusion Detection Method Using Protocol Classification and Rough Set Based Support Vector Machine Computer ad formatio Sciece trusio Detectio Method Usig Protocol Classificatio ad Rough Set Based Support Vector Machie Xuyi Re College of Computer Sciece, Najig Uiversity of Post & Telecommuicatios Najig

More information

An Algorithm of Mobile Robot Node Location Based on Wireless Sensor Network

An Algorithm of Mobile Robot Node Location Based on Wireless Sensor Network A Algorithm of Mobile Robot Node Locatio Based o Wireless Sesor Network https://doi.org/0.399/ijoe.v3i05.7044 Peg A Nigbo Uiversity of Techology, Zhejiag, Chia eirxvrp2269@26.com Abstract I the wireless

More information

An Anomaly Detection Method Based On Deep Learning

An Anomaly Detection Method Based On Deep Learning Sed Orders for Reprits to reprits@bethamsciece.ae 734 The Ope Automatio ad Cotrol Systems Joural, 05, 7, 734-739 A Aomaly Detectio Method Based O Deep Learig Ope Access Hog-li Deg *, Tao yag ad Jiag-i

More information

Criterion in selecting the clustering algorithm in Radial Basis Functional Link Nets

Criterion in selecting the clustering algorithm in Radial Basis Functional Link Nets WSEAS TRANSACTIONS o SYSTEMS Ag Sau Loog, Og Hog Choo, Low Heg Chi Criterio i selectig the clusterig algorithm i Radial Basis Fuctioal Lik Nets ANG SAU LOONG 1, ONG HONG CHOON 2 & LOW HENG CHIN 3 Departmet

More information

Creating Exact Bezier Representations of CST Shapes. David D. Marshall. California Polytechnic State University, San Luis Obispo, CA , USA

Creating Exact Bezier Representations of CST Shapes. David D. Marshall. California Polytechnic State University, San Luis Obispo, CA , USA Creatig Exact Bezier Represetatios of CST Shapes David D. Marshall Califoria Polytechic State Uiversity, Sa Luis Obispo, CA 93407-035, USA The paper presets a method of expressig CST shapes pioeered by

More information

Feature classification for multi-focus image fusion

Feature classification for multi-focus image fusion Iteratioal Joural of the Physical Scieces Vol. 6(0), pp. 4838-4847, 3 September, 0 Available olie at http://www.academicjourals.org/ijps DOI: 0.5897/IJPS.73 ISSN 99-950 0 Academic Jourals Full Legth Research

More information

SOFTWARE usually does not work alone. It must have

SOFTWARE usually does not work alone. It must have Proceedigs of the 203 Federated Coferece o Computer Sciece ad Iformatio Systems pp. 343 348 A method for selectig eviromets for software compatibility testig Łukasz Pobereżik AGH Uiversity of Sciece ad

More information

Pseudocode ( 1.1) Analysis of Algorithms. Primitive Operations. Pseudocode Details. Running Time ( 1.1) Estimating performance

Pseudocode ( 1.1) Analysis of Algorithms. Primitive Operations. Pseudocode Details. Running Time ( 1.1) Estimating performance Aalysis of Algorithms Iput Algorithm Output A algorithm is a step-by-step procedure for solvig a problem i a fiite amout of time. Pseudocode ( 1.1) High-level descriptio of a algorithm More structured

More information

Eigenimages. Digital Image Processing: Bernd Girod, 2013 Stanford University -- Eigenimages 1

Eigenimages. Digital Image Processing: Bernd Girod, 2013 Stanford University -- Eigenimages 1 Eigeimages Uitary trasforms Karhue-Loève trasform ad eigeimages Sirovich ad Kirby method Eigefaces for geder recogitio Fisher liear discrimat aalysis Fisherimages ad varyig illumiatio Fisherfaces vs. eigefaces

More information

arxiv: v2 [cs.ds] 24 Mar 2018

arxiv: v2 [cs.ds] 24 Mar 2018 Similar Elemets ad Metric Labelig o Complete Graphs arxiv:1803.08037v [cs.ds] 4 Mar 018 Pedro F. Felzeszwalb Brow Uiversity Providece, RI, USA pff@brow.edu March 8, 018 We cosider a problem that ivolves

More information

Fuzzy Linear Regression Analysis

Fuzzy Linear Regression Analysis 12th IFAC Coferece o Programmable Devices ad Embedded Systems The Iteratioal Federatio of Automatic Cotrol September 25-27, 2013. Fuzzy Liear Regressio Aalysis Jaa Nowaková Miroslav Pokorý VŠB-Techical

More information

Optimization of Multiple Input Single Output Fuzzy Membership Functions Using Clonal Selection Algorithm

Optimization of Multiple Input Single Output Fuzzy Membership Functions Using Clonal Selection Algorithm Optimizatio of Multiple Iput Sigle Output Fuzzy Membership Fuctios Usig Cloal Selectio Algorithm AYŞE MERVE ACILAR, AHMET ARSLAN Computer Egieerig Departmet Selcuk Uiversity Selcuk Uiversity, Eg.-Arch.

More information

Pruning and Summarizing the Discovered Time Series Association Rules from Mechanical Sensor Data Qing YANG1,a,*, Shao-Yu WANG1,b, Ting-Ting ZHANG2,c

Pruning and Summarizing the Discovered Time Series Association Rules from Mechanical Sensor Data Qing YANG1,a,*, Shao-Yu WANG1,b, Ting-Ting ZHANG2,c Advaces i Egieerig Research (AER), volume 131 3rd Aual Iteratioal Coferece o Electroics, Electrical Egieerig ad Iformatio Sciece (EEEIS 2017) Pruig ad Summarizig the Discovered Time Series Associatio Rules

More information

6.854J / J Advanced Algorithms Fall 2008

6.854J / J Advanced Algorithms Fall 2008 MIT OpeCourseWare http://ocw.mit.edu 6.854J / 18.415J Advaced Algorithms Fall 2008 For iformatio about citig these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. 18.415/6.854 Advaced Algorithms

More information

CSC 220: Computer Organization Unit 11 Basic Computer Organization and Design

CSC 220: Computer Organization Unit 11 Basic Computer Organization and Design College of Computer ad Iformatio Scieces Departmet of Computer Sciece CSC 220: Computer Orgaizatio Uit 11 Basic Computer Orgaizatio ad Desig 1 For the rest of the semester, we ll focus o computer architecture:

More information

Toward Automated Quality Classification via Statistical Modeling of Grain Images for Rice Processing Monitoring

Toward Automated Quality Classification via Statistical Modeling of Grain Images for Rice Processing Monitoring Iteratioal Joural of Computatioal Itelligece Systems, Vol. 9, No. 1 (2016) 120-132 Toward Automated Quality Classificatio via Statistical Modelig of Grai Images for Rice Processig Moitorig Jipig Liu* 1,

More information

Linearising Calibration Methods for a Generic Embedded Sensor Interface (GESI)

Linearising Calibration Methods for a Generic Embedded Sensor Interface (GESI) 1st Iteratioal Coferece o Sesig Techology November 21-23, 2005 Palmersto North, New Zealad Liearisig Calibratio Methods for a Geeric Embedded Sesor Iterface (GESI) Abstract Amra Pašić Work doe i: PEI Techologies,

More information

Eigenimages. Digital Image Processing: Bernd Girod, Stanford University -- Eigenimages 1

Eigenimages. Digital Image Processing: Bernd Girod, Stanford University -- Eigenimages 1 Eigeimages Uitary trasforms Karhue-Loève trasform ad eigeimages Sirovich ad Kirby method Eigefaces for geder recogitio Fisher liear discrimat aalysis Fisherimages ad varyig illumiatio Fisherfaces vs. eigefaces

More information

A new algorithm to build feed forward neural networks.

A new algorithm to build feed forward neural networks. A ew algorithm to build feed forward eural etworks. Amit Thombre Cetre of Excellece, Software Techologies ad Kowledge Maagemet, Tech Mahidra, Pue, Idia Abstract The paper presets a ew algorithm to build

More information

AN OPTIMIZATION NETWORK FOR MATRIX INVERSION

AN OPTIMIZATION NETWORK FOR MATRIX INVERSION 397 AN OPTIMIZATION NETWORK FOR MATRIX INVERSION Ju-Seog Jag, S~ Youg Lee, ad Sag-Yug Shi Korea Advaced Istitute of Sciece ad Techology, P.O. Box 150, Cheogryag, Seoul, Korea ABSTRACT Iverse matrix calculatio

More information

The Research Based on Fuzzy Sliding Mode Control for Linear Double. Inverted Pendulum

The Research Based on Fuzzy Sliding Mode Control for Linear Double. Inverted Pendulum Advaced Materials Research Olie: 04-05-3 ISSN: 66-8985, Vols. 96-930, pp 463-467 doi:0.408/www.scietific.et/amr.96-930.463 04 Tras Tech Publicatios, Switzerlad The Research Based o Fuzzy Slidig Mode Cotrol

More information

Description of some supervised learning algorithms

Description of some supervised learning algorithms Descriptio of some supervised learig algorithms Patrick Keekayoro patrick.keekayoro@outlook.com Statistical Cybermetrics Research Group Uiversity of Wolverhampto 1. Supervised learig Supervised machie

More information

Memory Convergence and Optimization with Fuzzy PSO and ACS

Memory Convergence and Optimization with Fuzzy PSO and ACS Joural of Computer Sciece 4 (2): 139-147, 2008 ISSN 1549-3636 2008 Sciece Publicatios Memory Covergece ad Optimizatio with Fuzzy PSO ad ACS 1 Subhash Chadra Padey ad 2 Dr. Pueet Misra 1 Birla Istitute

More information

Research Article Combined Kernel-Based BDT-SMO Classification of Hyperspectral Fused Images

Research Article Combined Kernel-Based BDT-SMO Classification of Hyperspectral Fused Images e Scietific World Joural, Article ID 738250, 13 pages http://dx.doi.org/10.1155/2014/738250 Research Article Combied Kerel-Based BDT-SMO Classificatio of Hyperspectral Fused Images Feghua Huag 1,2 ad Lumig

More information

Training Recurrent Neural Networks by a Hybrid PSO- Cuckoo Search Algorithm for Problems Optimization

Training Recurrent Neural Networks by a Hybrid PSO- Cuckoo Search Algorithm for Problems Optimization Traiig Recurret Neural Networks by a Hybrid PSO- Cuckoo Search Algorithm for Problems Optimizatio Ruba Talal Ibrahim Computer Sciece Departmet, Faculty of Computer Sciece & Mathematics, Mosul Uiversity,

More information

World Scientific Research Journal (WSRJ) ISSN: Research on Fresnel Lens Optical Receiving Antenna in Indoor Visible

World Scientific Research Journal (WSRJ) ISSN: Research on Fresnel Lens Optical Receiving Antenna in Indoor Visible World Scietific Research Joural (WSRJ) ISSN: 2472-3703 www.wsr-j.org Research o Fresel Les Optical Receivig Atea i Idoor Visible Light Commuicatio Zhihua Du College of Electroics Egieerig, Chogqig Uiversity

More information

A New Bit Wise Technique for 3-Partitioning Algorithm

A New Bit Wise Technique for 3-Partitioning Algorithm Special Issue of Iteratioal Joural of Computer Applicatios (0975 8887) o Optimizatio ad O-chip Commuicatio, No.1. Feb.2012, ww.ijcaolie.org A New Bit Wise Techique for 3-Partitioig Algorithm Rajumar Jai

More information

Classification by Evolutionary Generalized Radial Basis Functions

Classification by Evolutionary Generalized Radial Basis Functions 2009 Nith teratioal Coferece o telliget Systems Desig ad Applicatios Classificatio by Evolutioary Geeralized Radial Basis Fuctios A. Castaño, C. Hervás-Martíez 2, P. A. Gutierrez 2, F. Ferádez-Navarro

More information

Evolutionary Hybrid Genetic-Firefly Algorithm for Global Optimization

Evolutionary Hybrid Genetic-Firefly Algorithm for Global Optimization Iteratioal Joural of Computatioal Egieerig & Maagemet, Vol. 6 Issue, May ISSN (Olie): -789 www..org Evolutioary Hybrid Geetic-Firefly Algorithm for Global Optimizatio 7 Shaik Farook, P. Sagameswara Raju

More information

Evaluation scheme for Tracking in AMI

Evaluation scheme for Tracking in AMI A M I C o m m u i c a t i o A U G M E N T E D M U L T I - P A R T Y I N T E R A C T I O N http://www.amiproject.org/ Evaluatio scheme for Trackig i AMI S. Schreiber a D. Gatica-Perez b AMI WP4 Trackig:

More information

Appendix D. Controller Implementation

Appendix D. Controller Implementation COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Iterface 5 th Editio Appedix D Cotroller Implemetatio Cotroller Implemetatios Combiatioal logic (sigle-cycle); Fiite state machie (multi-cycle, pipelied);

More information

The golden search method: Question 1

The golden search method: Question 1 1. Golde Sectio Search for the Mode of a Fuctio The golde search method: Questio 1 Suppose the last pair of poits at which we have a fuctio evaluatio is x(), y(). The accordig to the method, If f(x())

More information