DETECTION OF ELECTRICAL FAULTS IN INDUCTION MOTOR FED BY INVERTER USING SUPPORT VECTOR MACHINE AND RECEIVER OPERATING CHARACTERISTIC

Size: px
Start display at page:

Download "DETECTION OF ELECTRICAL FAULTS IN INDUCTION MOTOR FED BY INVERTER USING SUPPORT VECTOR MACHINE AND RECEIVER OPERATING CHARACTERISTIC"

Transcription

1 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: DETECTION OF ELECTRICAL FAULTS IN INDUCTION MOTOR FED BY INVERTER USING SUPPORT VECTOR MACHINE AND RECEIVER OPERATING CHARACTERISTIC, DIAN R. SAWITRI, I. KETUT E. PURNAMA, MOCHAMAD ASHARI Electrcal Engneerng Department, Insttut Teknolog Sepuluh Nopember, Surabaya, Indonesa Electrcal Engneerng Department, Dan Nuswantoro Unversty, Semarang, Indonesa E-mal: ; ; ABSTRACT Fault n nducton motor s crucal problem n ndustral processes. Ths paper presents the system for electrcal fault detecton n nducton motor fed by nverter. Current spectrum wth dfferent frequency s used to fault montorng. Faults observed ncludes varaton of frequency, unbalance voltage, and nter turn short crcuts. Through an experment, the fault was fred and the current spectrum recorded at steady state condton. Preprocessng s performed before the dentfcaton process. It ncludes nose reducton usng wavelet analyss and feature extracton wth Prncpal Component Analyss (PCA). Both processes are ntended to elmnate the nose, reducng the dmenson of feature, and retreve components of the optmal features for classfcaton. Strength of dentfcaton capablty usng Support Vector Machne (SVM) s 83.5%. Based on the ROC (Recever Operatng Characterstc) analyss, the SVM classfer has a good enough performance. Ths s ndcated by the senstvty s 74.3%, specfcty s 47.30% and G-Mean s.08. Keywords: Electrcal Fault Detecton, Inducton Motor, Prncpal Component Analyss (PCA), Recever Operatng Characterstc (ROC), Support Vector Machne (SVM).. INTRODUCTION Inducton motor s wdely used n ndustral applcaton. Durng operaton, the motor may be experence wth faults. If the fault s not treated then the motor may have faled that causes producton actvtes must be stopped. So, the producton process dsrupts and causes energy waste. Fault n nducton motor can be ether mechancal or electrcal fault. In the prevous study, the detected fault n nducton motor are broken rotor bar []-[4] and bearng fault [],[4]- [6]. Electrcal fault s usually nfluenced by power qualty that suppled by ac grd, such as varatons of frequency and unbalanced voltage. Another fault s ntern short crcuts n stator wndng [],[7]-[0]. Some study has appled artfcal transform as denosng algorthm []. For multdmensonal data, wavelet transform wll be combnng wth Prncpal Component Analyss (PCA) as denosng algorthm []. Observed sgnal usually contan many parameter feature. To ntellgent method to detect the both of fault. The methods are fuzzy logc [6][0], Fast Fourer Transform (FFT) [3],[5],[6],[8], Artfcal Neural Network (ANN) [7], Support Vector Machne (SVM) [4], and Kalman Flter []. Motor Current Spectrum Analyss (MCSA) s used as fault parameter n the motor. In ndustral applcatons, nducton machnes are suppled and controlled by nverters. The effect of the nverters causes the hgh harmoncs n the currents that were recorded due to the swtchng operaton. Thus, t becomes more dffcult and demandng to detect faults by usng MCSA n ths drves [3]. To get a more accurate dentfcaton, need to be done several processes to reduce the noses and elmnate features that are not desrable. Current spectrum wth hgh noses can reduced wth wavelet ncrease the dentfcaton, feature extracton need to be done on orgnal sgnal. Feature extracton s performed to elmnate the feature parameter that are not approprate and reduce the dmensonal of data [3]. Feature extracton methods have been 5

2 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: used n prevous study to mprove the classfcaton accuracy. The methods are Independent Component Analyss (ICA) [4]-[6], Kernel ICA, PCA, and Kernel PCA [6]. Ths study detected the electrcal fault n nducton motor fed by nverter. The faults are caused by varaton of frequency, unbalance voltage, and nter-turn short crcuts. The effect of fault and nverter n ths research caused the hgh nose. To reduce the noses, combnaton wavelet and PCA are used as denosng algorthm. Feature extracton wth PCA and SVM classfcaton wth ones-aganst-ones strategy selected to dentfy the fault condton. The Performance of SVM as classfer wll be determned by Recever Operatng Characterstc (ROC). In most prevous studes, ROC analyss were appled on pattern recognton to dagnose a dsease [4],[5]. The ROC Analyss for power system research has been tested to examne of fault dentfcaton on a radal dstrbuton system wth SVM [6]. In ths paper, we use ROC analyss to determne the performance of SVM classfer for fault dentfcaton system on nducton machne fed by nverter.. METHODS AND EXPERIMENT. The Fault Identfcaton System The block dagram or lab test bench of the proposed fault dentfcaton system s shown n Fg.. The characterstcs of the three-phase nducton motor used n ths experment are lsted n Table I. The Current spectrum s recorded from the motor caused drven by nverter. The faults were rased by varyng the frequency of nverter, unbalance voltage, and nter turn short crcuts. TABLE I MOTOR CHARACTERISTIC USED IN EXPERIMENT Descrpton Value Power 0.5 kw Input Voltage 380 V Full Load Current 0.8 A Suply Frequency 50 Hz Number of Poles 4 Full Load Speed (rpm) 30 Fault dentfcaton conssts of 4 steps that are: denosng, feature calculaton, feature extracton, and classfcaton. To determne the valdty of classfer, the step s resumed by analyzng the performance of classfcaton.. Denosng Sgnal Nose from current spectrum wth hgh harmonc must be removed by flterng or denosng. Combnaton of wavelet transform and Prncpal Component Analyss (PCA) s used as de- Fg.. The proposed fault Identfcaton System nosng algorthm []. In general, the recorded sgnal s modeled as follows: x(t) = f(t) + ε(t) t =,,n () Where x(t) s observed sgnal, ε(t) s a centered Gausan whte nose of unknown var`ance and f(t) s a unknown functon to be recovered through the observatons (Fg. ) Fg.. Observed Sgnal, Orgnal Sgnal, and Gaussan Whte Nose Denosng procedure combnng wavelet and PCA s appled to reduce nose n multchannel sgnal recordng []. The scheme s as follows: ) Perform the wavelet transform at level J of each column of x []-[3] In ths step, the orthogonal wavelet s decomposed to Detal sgnal (D J ) and Approxmaton sgnal (A J ). The orthogonal 6

3 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: x(t) wavelet conssts of scalng functon (φ(x)) and wavelet functon (ψ(x)) (). φ ( x ) = ψ ( x ) = J t = 0 J t = 0 a φ ( x t ) k b φ ( x t ) k () Where (a 0 a J- ) s scalng sequence and (b 0 b J- ) s wavelet sequence. Scalng functons are assocated wth low-pass h( n), n z, flters wth coeffcent { } whle wavelet functons are assocated wth hgh-pass flters wth coeffcent g( n), n z, (Fg 3). { } Fg. 3. Two Level Flter Bank Decomposton Wavelet In ths way, the decomposton coeffcent can be descrbed as (3): [ x( t)] [ ca, cd ] [ ca, cd, cd, cd ] 3 3. (3) ) For j J, perform the PCA of the matrx cd j and select an approprate number p j of useful prncpal component or suppress the detal cd j ; 3) Smlarly, perform the PCA from the matrx ca J and select p J+ prncpal component; 4) From the smplfed detal and approxmaton matrces reconstruct a new matrx x ( contanng the man features of the orgnal matrx x, by nvertng the wavelet transform; 5) Fnally perform the PCA of the matrx x (..3 Feature Calculaton and Feature Extracton After denosng, feature parameters are calculated based on tme doman and frequency doman. There are 5 feature parameters of each phase (phase R, S, and T). We have 0 condtons and each one has 0 measurements, so the total obtaned 00 data calculated. The value of features parameter calculated based statstcal value of tme doman and frequency doman, such as mean, RMS, varance, peaks, moment, entropy, crest factor, Total Harmonc Dstorton, etc. We used PCA as feature extracton algorthm [7]- [8]. PCA wll reduce the dmenson of parameter but not elmnate the nformaton about the sgnal. The result s a parameter called Prncpal Component. The procedure nvolves egenvalue and egenvector as follows:. Gven a set of n dmenson nput vector and each of whch s of m dmenson. x x x( t) = xm x x x m x n x n xmn (4). Subtractng the value of each cell x j wth an overall mean µ j m u j = x j m = (5) φj = x j µ j (6) 3. Calculate the matrx covarance C C T = ( x j µ j )( x j µ j (7) 4. Gettng the egenvalue λ and egenvector u of matrx covarance C λ u = Cu (8) Where λ s egenvalue of C, u s the correspondng egenvector. 5. Based on the estmated u, the components of s t are the orthogonal transformatons of x t T st ( ) = u x t (9) s t () are called prncpal components. Dmensonal reducton of parameters s executed n (9). By usng a specfc threshold of the egenvector, u, the dmenson of the fnal data can be determned..4 Classfcaton In ths study, Support Vector Machne (SVM) s used to dentfy fault of nducton motor. All the measured data are classfed nto 0 types of fault. From each type, t s selected 30% measurement as tranng and 70% as testng data. SVM maps the nput vectors x nto a hghdmensonal features space Z through some nonlnear mappng [9]. In ths space, an optmal separatng hyperplane s constructed (Fg. 4). The learnng machnes construct the decson functons that are nonlnear n the nput space, f ( x) = sgn y K( x, x) b supportvectors α (0) 7

4 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: To fnd the coeffcents α n the separable case (analogously n the non-separable case) t s suffcent to fnd the maxmum of the functonal. n n W ( α ) = α αα j y y K ( x, x j ) () =, j j Optm al hyperplane n the feature space Feature Spaces (Z) Input Space (x) Fg. 4 The SV Machne maps the nput space nto a hgh-dmensonal feature and then constructs an Optmal hyperplane n the feature space Subject to the constrants n α = = n = y 0 α 0,,,, () K(x,x) s the Kernel functon n nput space that equvalent wth nner product n feature space. K s a symmetrc postve defnte functon whch satsfes Mercer s condton. K( x, ) = x akψ k( x ) ψk( x) ak 0 (3) k= It s necessary and suffcent that the condton K( x, x) g( x ) g( x) dx dx 0 Be vald for all g 0 for whch g ( x ) dx Examples of such kernels are gven n Table II. TABLE II EXAMPLE OF KERNEL FUNCTIONS No Kernel Type Functons Polynomal K ( x, x ) = x, x d Gausan Radal Bass Functon 3 Exponental Radal Bass Functon 4 Fourer Seres K ( x, ) = exp x K ( x, x) x x σ = x x exp σ sn N + K ( x =, x) sn ( x x) ( x x) The optmal hyperplane separated wthout error and the dstance between the closest vector to the hyperplane s maxmal (Fg. 5). Suppose the tranng data n ( x, y ),, ( x, y ), x R, y { +, } l l Can be separated by a hyperplane : ( w, x) + b = 0 (4) where, w and b shall be derved n such a way that unseen data can be classfed correctly. To descrbe the separatng hyperplane, the followng canoncal form can be use: ( w. x ) + b f y = +, ( w. x ) + b f y =. In the followng we use a compact notaton for these nequaltes: [(. ) b, = n] y w x +,, (5) It s easy to check that the optmal hyperplane s the one that satsfes the condton (5) and mnmzes (Fg. 5). Φ ( w ) = w (6) (The mnmzaton s taken wth respect to both vector w and scalar b) Fg. 5. The Optmal separatng hyperplane s the one that separates the data wth maxmal margn..5 Analyss the Performance of Classfcaton Classfcaton performance can be analyzed by graph Recever Operatng Characterstc (ROC). ROC graph s a technque to vsualze, organze, and choose the type of classfcaton based on ts performance [0]. 8

5 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: RESULT AND ANALYSIS FP fp rate = N TP Precson = TP + FP TP + TN Accuracy = P + N TP tp rate = P TP Recall = P F measure = / Precson + / Recall Fg. 6. Confuson matrx and common performance matrx calculated from t Classfcaton performance s determned by confuson matrx. In bnary classfcaton, the outcomes are labeled ether as postve (p) or negatve (n) class. There are four possble outcomes from a bnary classfer. They are true postve, false negatve, true negatve, and false postve (Fg 6). For example, to determne weather a sgnal has a certan fault. A true postve n ths case occurs when the sgnal test s postve (fault) and actually t s postve. A false postve, on the other hand, occurs when the test s postve and actually t s negatve, smlarly for true negatve and false negatve. Fg 6 shows a confuson matrx and equatons of several common that can be calculated from t. The numbers along the major dagonal represent the correct decsons made, whle the numbers n ths dagonal represent the errors the confuson between the varous classes. Performance of a classfer can be caused by the mbalance data set. Evaluaton the mbalanced classfcaton based on confuson matrx n Fg. 6. There are two knds of metrcs to deal wth class mbalanced []. The frst metrcs to obtan an optmal balanced classfcaton ablty are senstvty (7), specfcty (8) and G-Mean (9). There are usually adopted to separately montor the classfcaton performance on two classes. G-Mean s the geometrc mean of senstvty and specfcty (9). True Postve (7) senstvty = ( True Postve + Fals Negatve) (8) True Negatve specfcty = ( True Negatve + Fals Postve) G Mean = senstvty + specfcty (9) The second metrcs are precson, recall, and F-Measure (Fg. 6). Notce that recall s the same as senstvty. F-Measure s used to ntegrate precson and recall nto a sngle metrc. Lab experments have been done and result n recorded sgnal of 0 fault condtons wth 0 measurements every condton. To dentfy the fault, t s conducted steps as mentoned n prevous secton that are denosng process, feature calculaton, feature extracton, and classfcaton. 3. Denosng Result Current sgnal are recorded n ths study and have a very hgh nose. Ths nose must be removed to avod bas. Fg. 7 shows example 4 types of three phase current sgnals that have been fltered usng combnaton of wavelet and PCA. The current sgnals nclude normal (no-fault, nter-turn short crcut, and 5% and 0% unbalanced voltage condton. As seen n Fg.7, combnaton of wavelet and PCA elmnate the noses. To dstngush the current sgnal due to unbalance voltage 5% and 0% s dffcult, because the spectrum seems very smlar. It needs a tool to quckly and accurately recognze faults usng the step as descrbe n prevous secton. Fg. 7. Denosng Current Sgnal wth Combnaton of Wavelet and PCA. 3. Feature Calculaton and Feature Extracton Result Fgure 8 shows the egenvalue of 5 feature parameters. Only 4 of 5 parameters are selected as prncpal components to be dentfed. Selecton s based on the hgh enough egenvalue. The others are dscarded, due to too small. 9

6 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: Fg 8. Egenvalue of Covarance matrx for Feature Selecton In order to show the scatter dagram of faults belongs to each prncpal component, the 3 bggest egenvalues of prncpal components s selected. Fg. 9 shows the scatter dagram of extracted 3 prncpal component usng PCA. As can be seen, the data tends to dstrbute n a group based on the fault types. 3.3 Identfcaton Result and It s Performance Fault dentfcaton of 4 prncpal components usng SVM method s presented n Table III. The normal condton of 50 Hz frequency can be dentfed perfectly, presentng 00% of the strength of dentfcaton ndex. The average n 83.5%. TABLE III IDENTIFICATION RESULT USING SVM Fault Condton Frequency Strength of Identfcaton (Hz) (%) No Fault No Fault Inter-Turn Short Crcuts Inter-Turn Short Crcuts Average 5% Unbalance Voltage Average 0% Unbalance Voltage Average 83.5 The performance of SVM classfer s based on ROC analyss, determned n Fg. 0 and Table IV. As seen n Fg. 0, there are 56 Postve sample (P) and 389 Negatve samples (N). There are 60 samples of 56 demonstratng a correct dentfcaton (True Postve), whle 40 samples are False Negatve (FN). FN means that the system has no faults, but t was recognzed havng fault. On the other hand, FP s a resent of SVM dentfcaton that system has faults, but t was recognzed normal condton. Fg. 0 Confuson Matrx of Fault Identfcaton Table IV shows Senstvty or TP rate as Ths ndcates the system ablty to recognze true sgnal as 74.3%. However, the ablty of system to recognze a wrong sgnal as specfcty s 47.30%. It s calculated from the rato of TN and total negatve samples. Other parameter s found as precson, whch s 84.98% calculatng from Fg. 6. Fg. 9 Scatter Dagram for 3 Largest Prncpal Component TABLE IV PERFORMANCE OF SVM CLASSIFIER BY ROC No Varable Prob. Value Senstvty = TP rate Specfcty Precson G-mean.08 The Geometrc mean (G-Mean) s presented as.08. It represents the balanced result between 0

7 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: senstvty and specfcty. G-mean s evenly balanced when the value s.0. A better G-Mean s obtaned f the value s greater than CONCLUSION Electrcty faults n nducton motor n ths study are dentfed by SVM based on current sgnal. The current sgnals are very nosy snce they are generated from a small rate of equpment. Before the process of dentfcaton, It s necessary to pre process the recorded sgnals for accurate mprovement. Ths ncludes, denosng sgnal wth combnaton of wavelet and PCA, feature calculaton based on tme and frequency doman, and feature extracton usng PCA. The study present was 4 prncpal components of 5 feature parameters based on ts egenvalue. The faults of nducton motor are dentfed by SVM wth strength of dentfcaton ndex average n 83.5%. Based on ROC analyss, the ablty of system recognzes the true sgnal s 74.3% (senstvty) and the wrong sgnal s 47.30% (specfcty). G- Mean s balanced because the value s greater than.0. In the future, current sgnals from bgger rate of equpment wll be observed wth more types of fault. To mprove the performance of SVM consders mprovng the feature extracton and feature selecton algorthm. REFERENCES: [] P.J. Travner, Revew of condton montorng of rotatng electrcal machnes, IET Electrcal Power Applcaton, 008, Vol., No. 4, pp [] F. Karam, J. Poshtan, and M. Poshtan, Detecton of broken rotor bars n nducton motors usng nonlnear Kalman Flters, ISA Transacton, 49, (00), pp [3] I. P. Georgakopoulus, E.D. Mtroknas, and A.N. Safacas, Detecton of nducton motor faults n nverter drves usng nventer nput current analyss, IEEE Transacton on Industral Electronc, Vol. 55, No. 9, September 0.pp [4] A. Wdodo, Bo-Suk Yang, Dong-Sk Gu, and Byeong-Keun Cho, 009, Intellgent fault dagnoss system of nducton motor based on transent current sgnal, Mechatroncs, 9, (009), pp [5] Zhaoxa Wang, C.S. Chang, and Yfan Zhang, A feature based frequency doman analyss algorthm for fault detecton of nducton motors, 0, 6 th IEEE conference on Industral Electronc and Applcaton [6] T. W. Chua, W.W. Tan, Z-X. Wang, and C.S. Chang, Hybrd tme frequency doman for nverter-fed nducton motor fault detecton, IEEE Internatonal Symposum on Industral Electronc (ISIE), 4-7 July, pp [7] V. N. Ghate and S. V. Dudul, Optmal MLP neural network classfer for fault detecton of three phase nducton motor, Expert System wth Applcatons, 37, (00), pp [8] R. Sharf, and M. Ebrahm, Detecton of stator wndng faults n nducton motors usng three-phase current montorng, ISA Transactons, 50, 0, pp [9] A. Cakr, H. Cals, E.U. Kucukslle, Data mnng approach for supply unbalance detecton n nducton motor, Expert System wth Applcatons, 36 (009), pp [0] V. J. Rodrguez, and A. Arkko, Detecton of stator wndng fault n nducton motor usng fuzzy logc, Appled Soft Computng, 8, (008), pp. -0. [] D.L. Donoho, De-Nosng by soft thresholdng, IEEE Transacton on Informaton Theory, Vol. 4, No. 3, May 995, pp [] M. Amnghafar, N. Cheze, and J.M. Pogg, Multvarate denosng usng wavelet and prncpal component analyss, Computatonal Statstc & Data Analyss 50 (006), pp [3] C. Gargour, M. Cabrea, V. Ramachandran, and J.M. Lna, A Short Introducton to Wavelet and Ther Applcatons, IEEE Crcuts and System Magazne, nd Quarter, 009. [4] Xn He, Xyun Song, Erc C. Frey, Applcaton of Three-Class ROC Analyss to Task-Based Image Qualty Assessment of Smultaneous Dual-Isotope Myocardal Perfuson SPECT(MPS), IEEE Transacton on Medcal Imagng, Vol. 7, No.. Nopember, 008. [5] Xn He, Charles E. M., Benjamn M. W. T., Jonathan M. L., and Erc C. Frey, Three Class ROC Analyss A Decson Theoretc Approach Under the Ideal Observer Framework, IEEE Transacton on Medcal Imagng, Vol. 5, No. 5. May, 006.

8 Journal of Theoretcal and Appled Informaton Technology JATIT & LLS. All rghts reserved. ISSN: E-ISSN: [6] D. R. Sawtr, A. Muntasa, K. E. Purnama, M. Ashar, and M. H. Purnomo, Data Mnng Based Recever Operatng Characterstc (ROC) for Fault detecton and Dagnoss n Radal Dstrbuton System, The Journal for Technolog and Scence IPTEK, Vol. 0, No. 4, Nopember, 009. [7] L. I. Smth, A Tutoral Prncpal Component Analyss, Unversty of Otago, New Zealand, 00, student_tutorals/prncpal_components.pdf. [8] I.J. Jollfe, Prncpal Component Analyss, Sprnger, New York, 986. [9] V. N. Vapnk, The Nature of Stasttcal Learnng Theory, Sprnger Verlag New York, Inc., 995. [0] T. Fawcett, An Introducton to ROC analyss, Pattern Recognton Letters, 7, 006, pp [] Yuchun Tang, Yan-Qng Zhang, N. V. Chawla, and S. Krasser, SVMs Modellng for Hghly Imbalanced Classfcaton, IEEE Transacton On System, Man, and Cybernetcs Part B: Cybernetcs, Vol. 39, No., February, 009

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

Detection of an Object by using Principal Component Analysis

Detection of an Object by using Principal Component Analysis Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,

More information

Feature Reduction and Selection

Feature Reduction and Selection Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components

More information

CLASSIFICATION OF ULTRASONIC SIGNALS

CLASSIFICATION OF ULTRASONIC SIGNALS The 8 th Internatonal Conference of the Slovenan Socety for Non-Destructve Testng»Applcaton of Contemporary Non-Destructve Testng n Engneerng«September -3, 5, Portorož, Slovena, pp. 7-33 CLASSIFICATION

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face Recognition University at Buffalo CSE666 Lecture Slides Resources: Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1 4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

Lecture 4: Principal components

Lecture 4: Principal components /3/6 Lecture 4: Prncpal components 3..6 Multvarate lnear regresson MLR s optmal for the estmaton data...but poor for handlng collnear data Covarance matrx s not nvertble (large condton number) Robustness

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

Using Neural Networks and Support Vector Machines in Data Mining

Using Neural Networks and Support Vector Machines in Data Mining Usng eural etworks and Support Vector Machnes n Data Mnng RICHARD A. WASIOWSKI Computer Scence Department Calforna State Unversty Domnguez Hlls Carson, CA 90747 USA Abstract: - Multvarate data analyss

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Announcements. Supervised Learning

Announcements. Supervised Learning Announcements See Chapter 5 of Duda, Hart, and Stork. Tutoral by Burge lnked to on web page. Supervsed Learnng Classfcaton wth labeled eamples. Images vectors n hgh-d space. Supervsed Learnng Labeled eamples

More information

Discrimination of Faulted Transmission Lines Using Multi Class Support Vector Machines

Discrimination of Faulted Transmission Lines Using Multi Class Support Vector Machines 16th NAIONAL POWER SYSEMS CONFERENCE, 15th-17th DECEMBER, 2010 497 Dscrmnaton of Faulted ransmsson Lnes Usng Mult Class Support Vector Machnes D.hukaram, Senor Member IEEE, and Rmjhm Agrawal Abstract hs

More information

Data Mining: Model Evaluation

Data Mining: Model Evaluation Data Mnng: Model Evaluaton Aprl 16, 2013 1 Issues: Evaluatng Classfcaton Methods Accurac classfer accurac: predctng class label predctor accurac: guessng value of predcted attrbutes Speed tme to construct

More information

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines A Modfed Medan Flter for the Removal of Impulse Nose Based on the Support Vector Machnes H. GOMEZ-MORENO, S. MALDONADO-BASCON, F. LOPEZ-FERRERAS, M. UTRILLA- MANSO AND P. GIL-JIMENEZ Departamento de Teoría

More information

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law)

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law) Machne Learnng Support Vector Machnes (contans materal adapted from talks by Constantn F. Alfers & Ioanns Tsamardnos, and Martn Law) Bryan Pardo, Machne Learnng: EECS 349 Fall 2014 Support Vector Machnes

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents

More information

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z. TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

Journal of Process Control

Journal of Process Control Journal of Process Control (0) 738 750 Contents lsts avalable at ScVerse ScenceDrect Journal of Process Control j ourna l ho me pag e: wwwelsevercom/locate/jprocont Decentralzed fault detecton and dagnoss

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

RECOGNIZING GENDER THROUGH FACIAL IMAGE USING SUPPORT VECTOR MACHINE

RECOGNIZING GENDER THROUGH FACIAL IMAGE USING SUPPORT VECTOR MACHINE Journal of Theoretcal and Appled Informaton Technology 30 th June 06. Vol.88. No.3 005-06 JATIT & LLS. All rghts reserved. ISSN: 99-8645 www.jatt.org E-ISSN: 87-395 RECOGNIZING GENDER THROUGH FACIAL IMAGE

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and

More information

Classifying Acoustic Transient Signals Using Artificial Intelligence

Classifying Acoustic Transient Signals Using Artificial Intelligence Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)

More information

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET 1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School

More information

Modular PCA Face Recognition Based on Weighted Average

Modular PCA Face Recognition Based on Weighted Average odern Appled Scence odular PCA Face Recognton Based on Weghted Average Chengmao Han (Correspondng author) Department of athematcs, Lny Normal Unversty Lny 76005, Chna E-mal: hanchengmao@163.com Abstract

More information

Recognizing Faces. Outline

Recognizing Faces. Outline Recognzng Faces Drk Colbry Outlne Introducton and Motvaton Defnng a feature vector Prncpal Component Analyss Lnear Dscrmnate Analyss !"" #$""% http://www.nfotech.oulu.f/annual/2004 + &'()*) '+)* 2 ! &

More information

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches Proceedngs of the Internatonal Conference on Cognton and Recognton Fuzzy Flterng Algorthms for Image Processng: Performance Evaluaton of Varous Approaches Rajoo Pandey and Umesh Ghanekar Department of

More information

Classification / Regression Support Vector Machines

Classification / Regression Support Vector Machines Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM

More information

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION 1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute

More information

Face Recognition Based on SVM and 2DPCA

Face Recognition Based on SVM and 2DPCA Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, Unversty

More information

Quadratic Program Optimization using Support Vector Machine for CT Brain Image Classification

Quadratic Program Optimization using Support Vector Machine for CT Brain Image Classification IJCSI Internatonal Journal of Computer Scence Issues, Vol. 9, Issue 4, o, July ISS (Onlne): 694-84 www.ijcsi.org 35 Quadratc Program Optmzaton usng Support Vector Machne for CT Bran Image Classfcaton J

More information

Impact of a New Attribute Extraction Algorithm on Web Page Classification

Impact of a New Attribute Extraction Algorithm on Web Page Classification Impact of a New Attrbute Extracton Algorthm on Web Page Classfcaton Gösel Brc, Banu Dr, Yldz Techncal Unversty, Computer Engneerng Department Abstract Ths paper ntroduces a new algorthm for dmensonalty

More information

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline mage Vsualzaton mage Vsualzaton mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and Analyss outlne mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and

More information

Non-Negative Matrix Factorization and Support Vector Data Description Based One Class Classification

Non-Negative Matrix Factorization and Support Vector Data Description Based One Class Classification IJCSI Internatonal Journal of Computer Scence Issues, Vol. 9, Issue 5, No, September 01 ISSN (Onlne): 1694-0814 www.ijcsi.org 36 Non-Negatve Matrx Factorzaton and Support Vector Data Descrpton Based One

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School

More information

Feature Selection as an Improving Step for Decision Tree Construction

Feature Selection as an Improving Step for Decision Tree Construction 2009 Internatonal Conference on Machne Learnng and Computng IPCSIT vol.3 (2011) (2011) IACSIT Press, Sngapore Feature Selecton as an Improvng Step for Decson Tree Constructon Mahd Esmael 1, Fazekas Gabor

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based

More information

Support Vector Machines. CS534 - Machine Learning

Support Vector Machines. CS534 - Machine Learning Support Vector Machnes CS534 - Machne Learnng Perceptron Revsted: Lnear Separators Bnar classfcaton can be veed as the task of separatng classes n feature space: b > 0 b 0 b < 0 f() sgn( b) Lnear Separators

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Novel Fuzzy logic Based Edge Detection Technique

Novel Fuzzy logic Based Edge Detection Technique Novel Fuzzy logc Based Edge Detecton Technque Aborsade, D.O Department of Electroncs Engneerng, adoke Akntola Unversty of Tech., Ogbomoso. Oyo-state. doaborsade@yahoo.com Abstract Ths paper s based on

More information

Hybrid Non-Blind Color Image Watermarking

Hybrid Non-Blind Color Image Watermarking Hybrd Non-Blnd Color Image Watermarkng Ms C.N.Sujatha 1, Dr. P. Satyanarayana 2 1 Assocate Professor, Dept. of ECE, SNIST, Yamnampet, Ghatkesar Hyderabad-501301, Telangana 2 Professor, Dept. of ECE, AITS,

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

MULTI-SENSOR-BASED FAULT DETECTION AND CLASSIFICATION METHOD FOR RADIAL POWER DISTRIBUTION SYSTEMS. A Thesis by. Nan Wang

MULTI-SENSOR-BASED FAULT DETECTION AND CLASSIFICATION METHOD FOR RADIAL POWER DISTRIBUTION SYSTEMS. A Thesis by. Nan Wang MULTI-SENSOR-BASED FAULT DETECTION AND CLASSIFICATION METHOD FOR RADIAL POWER DISTRIBUTION SYSTEMS A Thess by Nan Wang Bachelor of Engneerng, Informaton Engneerng Unversty, 2011 Submtted to the Department

More information

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images Internatonal Journal of Informaton and Electroncs Engneerng Vol. 5 No. 6 November 015 Usng Fuzzy Logc to Enhance the Large Sze Remote Sensng Images Trung Nguyen Tu Huy Ngo Hoang and Thoa Vu Van Abstract

More information

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM PERFORMACE EVALUAIO FOR SCEE MACHIG ALGORIHMS BY SVM Zhaohu Yang a, b, *, Yngyng Chen a, Shaomng Zhang a a he Research Center of Remote Sensng and Geomatc, ongj Unversty, Shangha 200092, Chna - yzhac@63.com

More information

Support Vector Machine Based Arrhythmia Classification Using Reduced Features

Support Vector Machine Based Arrhythmia Classification Using Reduced Features Internatonal Journal Support of Control, Vector Automaton, Machne Based and Arrhythma Systems, vol. Classfcaton, no., pp. Usng 57-579, Reduced December Features 5 57 Support Vector Machne Based Arrhythma

More information

Spam Filtering Based on Support Vector Machines with Taguchi Method for Parameter Selection

Spam Filtering Based on Support Vector Machines with Taguchi Method for Parameter Selection E-mal Spam Flterng Based on Support Vector Machnes wth Taguch Method for Parameter Selecton We-Chh Hsu, Tsan-Yng Yu E-mal Spam Flterng Based on Support Vector Machnes wth Taguch Method for Parameter Selecton

More information

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated.

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated. Some Advanced SP Tools 1. umulatve Sum ontrol (usum) hart For the data shown n Table 9-1, the x chart can be generated. However, the shft taken place at sample #21 s not apparent. 92 For ths set samples,

More information

Human Face Recognition Using Generalized. Kernel Fisher Discriminant

Human Face Recognition Using Generalized. Kernel Fisher Discriminant Human Face Recognton Usng Generalzed Kernel Fsher Dscrmnant ng-yu Sun,2 De-Shuang Huang Ln Guo. Insttute of Intellgent Machnes, Chnese Academy of Scences, P.O.ox 30, Hefe, Anhu, Chna. 2. Department of

More information

SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE

SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE Dorna Purcaru Faculty of Automaton, Computers and Electroncs Unersty of Craoa 13 Al. I. Cuza Street, Craoa RO-1100 ROMANIA E-mal: dpurcaru@electroncs.uc.ro

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

More information

Machine Learning 9. week

Machine Learning 9. week Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below

More information

AN APPROPRIATE PROCEDURE FOR DETECTION OF JOURNAL-BEARING FAULT USING POWER SPECTRAL DENSITY, K-NEAREST NEIGHBOR AND SUPPORT VECTOR MACHINE

AN APPROPRIATE PROCEDURE FOR DETECTION OF JOURNAL-BEARING FAULT USING POWER SPECTRAL DENSITY, K-NEAREST NEIGHBOR AND SUPPORT VECTOR MACHINE INTERNATIONAL JOURNAL ON SMART SENSING AND INTELLIGENT SYSTEMS, VOL. 5, NO. 3, SEPTEMBER 0 AN APPROPRIATE PROCEDURE FOR DETECTION OF JOURNAL-BEARING FAULT USING POWER SPECTRAL DENSITY, K-NEAREST NEIGHBOR

More information

CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION

CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION 48 CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION 3.1 INTRODUCTION The raw mcroarray data s bascally an mage wth dfferent colors ndcatng hybrdzaton (Xue

More information

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines (IJCSIS) Internatonal Journal of Computer Scence and Informaton Securty, Herarchcal Web Page Classfcaton Based on a Topc Model and Neghborng Pages Integraton Wongkot Srura Phayung Meesad Choochart Haruechayasak

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

More information

A Saturation Binary Neural Network for Crossbar Switching Problem

A Saturation Binary Neural Network for Crossbar Switching Problem A Saturaton Bnary Neural Network for Crossbar Swtchng Problem Cu Zhang 1, L-Qng Zhao 2, and Rong-Long Wang 2 1 Department of Autocontrol, Laonng Insttute of Scence and Technology, Benx, Chna bxlkyzhangcu@163.com

More information

Face Recognition Method Based on Within-class Clustering SVM

Face Recognition Method Based on Within-class Clustering SVM Face Recognton Method Based on Wthn-class Clusterng SVM Yan Wu, Xao Yao and Yng Xa Department of Computer Scence and Engneerng Tong Unversty Shangha, Chna Abstract - A face recognton method based on Wthn-class

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

Unsupervised Learning and Clustering

Unsupervised Learning and Clustering Unsupervsed Learnng and Clusterng Why consder unlabeled samples?. Collectng and labelng large set of samples s costly Gettng recorded speech s free, labelng s tme consumng 2. Classfer could be desgned

More information

Discriminative classifiers for object classification. Last time

Discriminative classifiers for object classification. Last time Dscrmnatve classfers for object classfcaton Thursday, Nov 12 Krsten Grauman UT Austn Last tme Supervsed classfcaton Loss and rsk, kbayes rule Skn color detecton example Sldng ndo detecton Classfers, boostng

More information

Lecture 13: High-dimensional Images

Lecture 13: High-dimensional Images Lec : Hgh-dmensonal Images Grayscale Images Lecture : Hgh-dmensonal Images Math 90 Prof. Todd Wttman The Ctadel A grayscale mage s an nteger-valued D matrx. An 8-bt mage takes on values between 0 and 55.

More information

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

Analysis of Continuous Beams in General

Analysis of Continuous Beams in General Analyss of Contnuous Beams n General Contnuous beams consdered here are prsmatc, rgdly connected to each beam segment and supported at varous ponts along the beam. onts are selected at ponts of support,

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.

More information

Novel Pattern-based Fingerprint Recognition Technique Using 2D Wavelet Decomposition

Novel Pattern-based Fingerprint Recognition Technique Using 2D Wavelet Decomposition Mathematcal Methods for Informaton Scence and Economcs Novel Pattern-based Fngerprnt Recognton Technque Usng D Wavelet Decomposton TUDOR BARBU Insttute of Computer Scence of the Romanan Academy T. Codrescu,,

More information

Network Intrusion Detection Based on PSO-SVM

Network Intrusion Detection Based on PSO-SVM TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*

More information

PCA Based Gait Segmentation

PCA Based Gait Segmentation Honggu L, Cupng Sh & Xngguo L PCA Based Gat Segmentaton PCA Based Gat Segmentaton Honggu L, Cupng Sh, and Xngguo L 2 Electronc Department, Physcs College, Yangzhou Unversty, 225002 Yangzhou, Chna 2 Department

More information

Face Detection with Deep Learning

Face Detection with Deep Learning Face Detecton wth Deep Learnng Yu Shen Yus122@ucsd.edu A13227146 Kuan-We Chen kuc010@ucsd.edu A99045121 Yzhou Hao y3hao@ucsd.edu A98017773 Mn Hsuan Wu mhwu@ucsd.edu A92424998 Abstract The project here

More information

Under-Sampling Approaches for Improving Prediction of the Minority Class in an Imbalanced Dataset

Under-Sampling Approaches for Improving Prediction of the Minority Class in an Imbalanced Dataset Under-Samplng Approaches for Improvng Predcton of the Mnorty Class n an Imbalanced Dataset Show-Jane Yen and Yue-Sh Lee Department of Computer Scence and Informaton Engneerng, Mng Chuan Unversty 5 The-Mng

More information

Feature Extractions for Iris Recognition

Feature Extractions for Iris Recognition Feature Extractons for Irs Recognton Jnwook Go, Jan Jang, Yllbyung Lee, and Chulhee Lee Department of Electrcal and Electronc Engneerng, Yonse Unversty 134 Shnchon-Dong, Seodaemoon-Gu, Seoul, KOREA Emal:

More information

The Study of Remote Sensing Image Classification Based on Support Vector Machine

The Study of Remote Sensing Image Classification Based on Support Vector Machine Sensors & Transducers 03 by IFSA http://www.sensorsportal.com The Study of Remote Sensng Image Classfcaton Based on Support Vector Machne, ZHANG Jan-Hua Key Research Insttute of Yellow Rver Cvlzaton and

More information

A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES

A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES Aram AlSuer, Ahmed Al-An and Amr Atya 2 Faculty of Engneerng and Informaton Technology, Unversty of Technology, Sydney, Australa

More information

Research and Application of Fingerprint Recognition Based on MATLAB

Research and Application of Fingerprint Recognition Based on MATLAB Send Orders for Reprnts to reprnts@benthamscence.ae The Open Automaton and Control Systems Journal, 205, 7, 07-07 Open Access Research and Applcaton of Fngerprnt Recognton Based on MATLAB Nng Lu* Department

More information

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

On Evaluating Open Biometric Identification Systems

On Evaluating Open Biometric Identification Systems Proceedngs of Student/Faculty Research Day, CSIS, Pace Unversty, May 6th, 2005 On Evaluatng Open Bometrc Identfcaton Systems Mchael Gbbons, Sungsoo Yoon, Sung-Hyuk Cha and Charles Tappert mkegbb@us.bm.com,

More information

Optimizing Document Scoring for Query Retrieval

Optimizing Document Scoring for Query Retrieval Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng

More information

UB at GeoCLEF Department of Geography Abstract

UB at GeoCLEF Department of Geography   Abstract UB at GeoCLEF 2006 Mguel E. Ruz (1), Stuart Shapro (2), June Abbas (1), Slva B. Southwck (1) and Davd Mark (3) State Unversty of New York at Buffalo (1) Department of Lbrary and Informaton Studes (2) Department

More information

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

More information

SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING USING A CALCULATOR

SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING USING A CALCULATOR SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING USING A CALCULATOR Judth Aronow Rchard Jarvnen Independent Consultant Dept of Math/Stat 559 Frost Wnona State Unversty Beaumont, TX 7776 Wnona, MN 55987 aronowju@hal.lamar.edu

More information