An Improved Neural Network Algorithm for Classifying the Transmission Line Faults

Size: px
Start display at page:

Download "An Improved Neural Network Algorithm for Classifying the Transmission Line Faults"

Transcription

1 1 An Improved Neural Network Algorthm for Classfyng the Transmsson Lne Faults S. Vaslc, Student Member, IEEE, M. Kezunovc, Fellow, IEEE Abstract--Ths study ntroduces a new concept of artfcal ntellgence based algorthm for classfyng the faults n power system networks. Ths classfcaton dentfes the exact type and zone of the fault. The algorthm s based on unque type of neural network specally developed to deal wth large set of hghly dmensonal nput data. An mprovement of the algorthm s proposed by mplementng varous steps of nput sgnal preprocessng, through the selecton of parameters for analog flterng, and values for the data wndow and samplng frequency. In addton, an advanced technque for classfcaton of the test patterns s dscussed and the man advantages comparng to prevously used nearest neghbor classfer are shown. Index Terms--clusterng methods, electromagnetc transents, neural networks, pattern classfcaton, power system faults, protectve relayng, testng, tranng. I. INTRODUCTION HIS paper deals wth neural network based technque for Tprotectve relayng n power system networks. The problem of detectng and classfyng the transmsson lne faults has been known for a long tme. Tradtonal relayng prncples are based on predetermned settng and take nto account only common and easly antcpated fault condtons. Several varyng parameters: type of fault, fault locaton, fault mpedance, and fault ncdent tme, as well as many other condtons mposed by actual network confguraton, voltage levels, and varsty of other events, determne the correspondng transent current and voltage waveforms detected by the relays at lne ends. All these effects nfluence the relays tuned only to perform well durng antcpated fault condtons. The new classfcaton approach has to relably conclude, n a very short tme (around one cycle), whether, where and whch type of fault occurs under a varety of tmevaryng operatng condtons [1]. Neural networks can be used to solve power system protecton problems, partcularly those where tradtonal approaches have dffculty n achevng the desred speed, accuracy and selectvty. Neural networks are convenent for Ths study was supported by an Army/EPRI contract # WO between EPRI and Carnege Mellon Unversty, and has been carred out by Texas A&M Unversty under the subcontract # ttled "Self- Evolvng Agents for Montorng, Control and Protecton of Large, Complex Dynamc Systems". S. Vaslc and M. Kezunovc are wth the Department of Electrcal Engneerng, Texas A&M Unversty, College Staton, TX USA (e-mals respectvely: svaslc@ee.tamu.edu, kezunov@ee.tamu.edu). these tasks because they use ndvdual examples to capture general, always complex and nonlnear, relatonshps among the data. They learn from the envronment and adapt ther recognton capabltes. Varous applcatons of neural networks were used n the past to mprove recognton of the mpedance used n dstance relayng of transmsson lnes [2]. These applcatons are manly based on wdely used multlayer feed-forward networks. Whenever nput patterns wth large dmensonalty are present (as n ths partcular case), tranng of these networks s very slow, needs much larger tranng sets, and very easly converges on local mnma, usually very far from the global mnmum. On-lne tranng of ths type of network for any new tranng data requres presentng the entre set of tranng patterns agan. Instead of usng multlayer neural network, a unque type of neural network may be used for fault classfcaton [3]-[6]. The network s appled drectly to the samples of voltages and currents, and produce the fault type and zone classfcaton n real tme. Ths network s based on ISODATA clusterng algorthm [7] and belongs to a group of specal neural networks named Self-Organzng Maps [8]. The adaptve behavor of the neural network s descrbed by Adaptve Resonance Theory [9]. The man am of ths work s to provde some nsghts nto possble approaches to sgnfcantly mprove exstng neural network algorthm, through approprate condtonng of nput patterns as well as optmzng the classfcaton of new patterns that have not been presented n the tranng process. Electromagnetc transent program ATP [10] s used for smulatng a specally developed power network model. Smulaton outputs of a large number of scenaros, ncludng varous fault and normal operatng states, are used as a sgnal generator for the neural network algorthm desgn, mplemented n MATLAB [11]-[13]. The paper s organzed as follows. Secton II presents a bref descrpton of the neural network classfcaton algorthm. Secton III, dvded nto subsectons, demonstrates generaton and preprocessng of neural network nput sgnals, and defnes classfcaton task. An advanced classfcaton technque s ntroduced n secton IV. The concluson s gven at the end. II. NEURAL NETWORK CLASSIFICATION ALGORITHM Neural networks try to produce a concse representaton of system's behavor through dentfyng natural groupngs of data

2 2 from large data sets. The am of ths procedure, called clusterng, s to partton a gven set of nput data (patterns) nto several groups or clusters, so that each pattern s assgned to a unque cluster. Patterns that belong to the same cluster should be as smlar as possble, whle patterns that belong to dfferent clusters should be as dfferent as possble. Class label s assgned to each cluster, where class symbolzes a group of patterns wth a common characterstc. Self-organzng maps are specal type of neural networks that map nput patterns wth smlar features nto adjacent clusters after enough nput patterns have been presented [8]. The smlarty between patterns s usually measured by calculatng the Eucldean dstance between two n-dmensonal nput vectors. After tranng, self-organzed clusters represent prototypes of classes of nput patterns. ISODATA clusterng algorthm dscovers the most representatve postons of prototypes n the pattern space [7]. Adaptve Resonance Theory s characterzed by ts ablty to form clusters ncrementally, whenever a pattern, suffcently dfferent from all prevously presented patterns, appears [9]. Incremental clusterng capablty can handle an nfnte set of nput data, because ther cluster prototype unts contan mplct representaton of all the nputs prevously encountered. Usng ths technque, the on-lne tranng due to non-statonary nputs may be easly mplemented. Ths neural network s wthout hdden layers and ts selforganzed structure depends only on the presented nput data set. The neural network tranng conssts of unsupervsed and supervsed learnng phases (Fg. 1) [4]. Durng unsupervsed learnng, patterns are presented wthout ther class labels. Ths procedure tres to dentfy characterstc patterns or prototypes that can serve as cluster centers. The outcome of unsupervsed learnng s a stable famly of clusters, defned as hyperspheres n an n-dmensonal space, where n denotes the number of features,.e. the length of nput vector. Unsupervsed learnng forms stable famly of both "clean" (homogenous, havng patterns wth the same class label) and "mxed" (non-homogenous, havng patterns wth two or more class labels) clusters. It does not requre ether the ntal guess of the number of cluster, or the ntal cluster center coordnates. It conssts of two steps: ntalzaton and stablzaton. The Intalzaton establshes ntal cluster structure based on smlarty between the patterns, and by presentng each pattern only ones. Durng the stablzaton all patterns are presented agan untl a stable cluster structure occurs and there are no patterns changng ther cluster membershp durng the teratons. Stablzaton phase s repeated many tmes untl no pattern changes ts cluster membershp. In the supervsed learnng the class label s assocated wth each data pont. Supervsed learnng separates "mxed" clusters from the "clean" ones. It assgns class labels to the "clean" clusters. The tunng parameter ρ, called vglance parameter, controls the number and sze of generated clusters and s beng consecutvely decreased durng teratons. UNSUPERVISED LEARNING SUPERVISED LEARNING Start Patterns zaton Intalzaton Patterns Clusterng Stablzaton Patterns Clusterng Clean and Mxed Clusters Identfcaton Clean Clusters Class Membershp Assgnement Clean Clusters Extracton Input Set Reducton Decreasng Vglance Parameter ρ Number of Mxed Clusters s Zero or ρ < ε? Yes End Fg. 1. Neural Network Clusterng Algorthm New Iteraton Unsupervsed and supervsed learnng procedures are repeated unless only "clean" clusters exst, or current value of vglance parameter s less then specfed value. Durng the testng procedure, dstances between each test pattern and establshed clusters are calculated, and the nearest neghbors classfer [14] s effcently mplemented snce the number of optmzed prototypes s sgnfcantly smaller then the number of tranng patterns. The outcome of the testng are class labels assgned to test patterns accordng to the most common value among the K nearest prototypes. III. ALGORITHM IMPLEMENTATION FOR CLASSIFYING TRANSMISSION LINE FAULTS A. Generaton of Input Sgnals Power system used for algorthm evaluaton and testng s 345 kv power system secton, provded by Relant Energy (RE) HL&P company. The model of the gven power network s mplemented n Alternatve Transent Program (ATP) program [10]. It s used for smulatng varous fault scenaros on one of the transmsson lnes, by varyng fault parameters. The reduced network equvalent was obtaned by usng the load flow and short crcut data, and verfed usng both the steady state and transent state results. Neural network based algorthm takes voltage and current measurements from one end of the lne. It has to be traned to recognze the type and No

3 3 zone of the fault. The classfcaton s based only on drect use of samples wthout mpedance computng. The example of the smulaton output data, ncludng three-phase lne currents and voltages, for one specfc case, phases A to B to ground fault, s shown n Fg. 2. Current and voltage samples obtaned through smulatons are used for formng tranng and test patterns for protectve algorthm learnng and evaluaton. Fg. 3. Voltage and current samples and movng data wndow used for formng the patterns. Fg. 2. Typcal voltage and current measurements for ABG fault. Tranng patterns are generated by specfyng several values for each fault parameter, and combnng these values to cover possblty space of fault cases [14]. Prototypes formed durng tranng represent the most typcal patterns obtaned through smulatons. Test patterns for algorthm evaluaton n heurstc, prevously unseen, stuatons are generated by random settng of all fault parameters. The test patterns mght be very heterogeneous and qute dfferent from the tranng patterns snce there are many operatng states and possble events n the power network. They are classfed accordng to ther smlarty to prototypes adopted durng tranng. B. Preprocessng of Input Sgnals Neural network tranng can be made more effcent f certan preprocessng steps are performed on the network nputs. Pattern extracton from obtaned measurements depends on several algorthm tunng parameters. Those parameters sgnfcantly determne qualty of algorthm tranng. Input nto the neural network s n the form of the movng data wndow contanng samples of phase currents and voltages (Fg. 3). Selecton of sampled data for tranng n desred data wndow may nclude: three phase currents, three phase voltages, or both the three phase currents and voltages. Comparson of algorthm performances n all three cases s a challengng task. Examples of extracted patterns for dfferent values of algorthm parameters are shown n Fgs. 4 to 6 for the choce of three-phase currents. Phase current measurements are fltered by an analog flter and sampled wth desred samplng frequency. Patterns are extracted from these measurements durng the desred data wndow (after the fault occurs), normalzed, and placed together n one row to form feature vector components. Fg. 4. Example of pattern feature vectors for dfferent values of II order Butterworth analog flter crossover frequences.

4 4 used for tranng, and gves better nformaton about the sgnals. At the same tme t means slower tranng and testng, and may cause defcences n classfyng the fault n a real tme. Samplng frequency has smlar effect on formng the patterns as the data wndow does. Pattern feature vectors for samplng frequences 1 khz, 2 khz, and 5 khz are shown n Fg. 6. Increased samplng frequency offers mproved sgnal detecton but also may cause sgnfcant computatonal burden. In each partcular applcaton, extensve set of smulatons have to be performed to optmze all mentoned parameters for algorthm tranng. When ths condton s satsfed, algorthm tranng may be started. Fg. 5. Example of pattern feature vectors for dfferent lengths of data wndow. Fg. 6. Example of pattern feature vectors for dfferent values of samplng frequency. The appled analog flter s the second order Butterworth flter, and the effect of varous selected crossover frequences, 0.5 khz, 1 khz, and 2 khz, s shown n Fg. 4. The flter crossover frequency has to be carefully adopted, such that the nose and other hgh frequency components are fltered, whle the characterstc sgnal waveform s retaned. The effect of the selected data wndow s llustrated n Fg. 5. for the wndow lengths of 0.5, 1 and 1.5 cycles (1 cycle = ms). Longer wndow ncreases the number of features C. Classfcaton Task Type of classfcaton mght be based on detectng the fault type (, AG, BG, CG, AB/ABG, BC/BCG, CA/CAG, ABC/ABCG), fault zone (, Zone I, Zone II), fault resstance (, Low, Hgh), or any combnaton among them. Classfcaton of testng patterns s performed by usng the cluster structure establshed durng tranng and applyng the K-nearest neghbor rule. Clusters are formed as spheres n the n-dmensonal space. Smplfed example n two dmensons s gven n Fg. 7. The cluster centers are dentfed as prototypes or characterstc patterns. Input parameter for algorthm testng s only the number K of the nearest neghbors for the rule. Test patterns are extracted and normalzed from generated patterns usng the same procedure as for tranng patterns and have equal number of features. For each test pattern, Eucldean dstances to all clusters retreved from reference set are computed and sorted n an ncreasng order. A-C -C A-C A-G -C A-C-G C-G B-G I B-C -C-G B-C I B-C-G I Fg. 7. Example of cluster structure establshed durng tranng. B-G I A-G B-G I The most frequent class label of K nearest clusters s computed and assgned to actual pattern. If the nput pattern belongs to any of the normal-state clusters then the nput data wndow s "moved" for one sample and the comparson s performed agan. If the nput pattern does not belong to the normal-state clusters then the fault s detected and executon of the fault classfcaton logc s ntated. The parameter used

5 5 to force the neural network to make the fnal decson s the tme. After decson tme has expred, f pattern stll does not come back to the normal state, the neural network wll classfy the fault event accordng to the fault type detected n that nstance. IV. ADVANCED K-NEAREST NEIGHBORS TECHNIQUE Standard verson of the K-nearest neghbors rule may be mproved to acheve better classfcaton of the test set of patterns that corresponds to a new set of smulated events, prevously unseen durng tranng. Between establshed clusters there s an unlabeled space makng the pattern recognton more dffcult, because n real stuaton many new upcomng patterns appear n that space. Gven a set of classfed clusters, the standard K-nearest neghbors rule [13] determnes the classfcaton of the nput pattern x based only on the class labels of the K closest clusters n the cluster structure establshed durng tranng µ x ) = f [ K, µ ( v )] (1) j ( j l where: µ j ( v l ) s membershp value whch determnes the degree of belongng of cluster l to class j; µ j ( x ) s membershp value of pattern belongng to class j; = 1,, P; where P s number of patterns; j = 1, C; where C s number of classes; l = 1, K; where K s number of neghbors; v 1, v2,, vk denotes the centers of K nearest neghbors of pattern x. µ j ( v l ) has only crsp values 0 or 1, dependng on whether or not a cluster v l belongs to class j: 1 f cluster l belongs to class j µ j ( v l ) = (2) 0 otherwse In ths rule all K nearest clusters have the equal mportance, wthout takng nto account ther rad, and dstances to the pattern that has to be classfed. The advanced K-nearest neghbors technque s a fuzzy classfcaton technque that generalzes the K-nearest neghbors rule. New patterns are classfed based on the weghted dstances ( d l ) to K nearest clusters, as well as on relatve sze ( r l ) and class labels ( c l ) of these clusters (Fg. 8). The advanced K-nearest neghbors technque calculates a vector of membershp values ( µ 1( x), µ 2 ( x), µ C ( x)) of nput pattern x n the exstng classes. The class membershp values are calculated based on the followng formula: µ x ) = f [ K, µ ( v ), d ( x )] (3) j ( j l l where now µ v ) may take any value between 0 and 1, j ( l representng the relatve sze of the actual cluster l. Each cluster belongs to one of the exstng classes, wth membershp value defned by the followng adopted relaton: rl rmax f cluster l belongs to class j µ j ( vl ) = (4) 0 otherwse The membershp degree of cluster v l belongng to class j s equal to the rato between radus ( r l ) of actual cluster l and radus ( r max ) of the largest cluster n the cluster structure. The outcome s that the larger clusters have more nfluences then the smaller ones, and the clusters wth longest radus have µ j ( v l ) = 1. r 9 r 3 c 3 r 6 c 6 c 9 d 9 c r 1 1 d 6 d 3 d 7 d 1 test pattern d 8 d 2 d 5 c r 1 2 c 5 d 4 r 5 Fg. 8. K nearest clusters (K=9 n ths example) to test pattern, wth ther class labels c j, rad r j, and dstances d j. Another mprovement toward realstc classfcaton s takng nto account dstances between pattern x and K nearest clusters. The dstance d l ( x ) may be generally selected to be a weghted Eucldean dstance between pattern x and cluster l l l c 4 m r 4 c 7 d ( x ) = x v (5) where the parameter m determnes how heavly the dstance s weghted when calculatng the class membershp. If two or more of K nearest clusters have the same class label, then they gve cumulatve membershp value to that class. When values µ j ( x ) for all K neghbors have been calculated, pattern x s classfed to the class wth the hghest membershp degree Class( x ) = { j µ ( x ) µ ( x ), j, m = 1,, C m j (6) j m Ths dea of new fuzzyfed classfcaton technque algorthm wll help classfy better a varety of test patterns, comparng to the prevously used K-nearest neghbors algorthm. The optmal values, for number of neghbors K and parameter m whch establshes weghted dstances, have to be determned and appled n each partcular mplementaton. r 7 c 8 r 8

6 6 V. CONCLUSION Ths study ntroduces possble drectons to mprove exstng neural network algorthm for classfyng the transmsson lne faults. Ths algorthm was used earler n the efforts amed at replacng dstance relays wth new relays not havng tradtonal settng and based on pattern recognton capabltes. The algorthm s a specfc type of clusterng algorthm. It translates nput patterns nto pattern prototypes, the structure of clusters that represents varous classes of nput data sets. The algorthm s very flexble and easly enables further modfcatons and upgradng. Condtonng of nput sgnals, as well as selecton of values for analog flterng, data wndow for takng the patterns, and samplng frequency, play sgnfcant role n the algorthm behavor durng tranng and performance durng testng. Dfferent aspects of these factors are llustrated through several examples. Furthermore, classfcaton of the test patterns s analyzed through comparson of a smple K-nearest neghbors classfer used so far, and the new fuzzyfed approach of that classfer. The advanced approach offers more realstc classfcaton of the test patterns. Ths s done by takng nto account weghted dstances between a pattern and the K nearest clusters, as well as the relatve sze of those clusters. Proposed tunngs and mprovements of the neural network algorthm may enable better classfcaton of the fault type and zone. Combnes use of neural and fuzzy technques n the same algorthm leads to complex reasonng that mproves the event classfcaton ablty for recognzng a varety of the events that could happen n power networks. VI. REFERENCES [1] Power System Relayng Commttee, Workng Group D5 of the Lne Protecton Subcommttee, Proposed statstcal performance measures for mcroprocessor-based transmsson lne protectve relays, Part I and II, IEEE Trans. Power Delvery, vol. 12, no. 1, pp , Jan [2] M. Kezunovc, "A Survey of Neural Net Applcatons to Protectve Relayng and Fault Analyss", Engneerng Intellgent Systems, vol. 5, no. 4, pp , Dec [3] Y. H. Pao, Adaptve Pattern Recognton and Neural Networks, Readng: Addson Wesley, 1989, p [4] Y. H. Pao and D. J. Sobajc, "Combned Use of Unsupervsed and Supervsed Learnng for Dynamc Securty Assessment", IEEE Trans. Power Systems, vol. 7, no 2, pp , [5] M. Kezunovc M., I. Rkalo, and D. Sobajc, "Hgh-speed Fault Detecton and Classfcaton wth Neural Nets", Electrc Power Systems Research, vol. 34, pp , [6] M. Kezunovc and I. Rkalo, "Detect and Classfy Faults Usng Neural Nets", IEEE Computer Applcatons n Power, vol. 9, no. 4, pp , [7] G. H. Ball and D. J. Hall, "A Clusterng Technque for Summarzng Multvarate Data", Behavoral Scence, vol. 12, pp , [8] T. Kohonen, Self-Organzng Maps, 1st ed., New York: Sprnger- Verlag, 1997, p [9] G. A. Carpenter and S. Grossberg, "ART2: self-organzaton of stable category recognton codes for analog nput patterns", Appled Optcs, vol. 26, no. 23, pp , Dec [10] CanAm EMTP User Group, Alternatve Transent Program (ATP) Rule Book, Portland, [11] The MathWorks, Inc., Usng MATLAB, Natck, Jan [12] M. Kezunovc and S. Vaslc, "Advanced Software Envronment for Evaluatng Protecton Performance Durng Power System Dsturbances Usng Relay Models", CIGRE SC 34 Colloquum, Sbu, Romana, Sep [13] M. Kezunovc and S. Vaslc, "Desgn and Evaluaton of Context- Dependent Protectve Relayng Approach", IEEE Porto Power Tech' Conference, Porto, Portugal, Sep [14] T. M. Cover and P. E. Hart, "Nearest Neghbor Pattern Classfcaton", IEEE Trans. Informaton Theory, vol. IT-13, pp , VII. BIOGRAPHIES Slavko Vaslc (S'00) receved hs B.S. and M.S. degrees n electrcal engneerng from Unversty of Belgrade n 1993 and 1999, respectvely, and currently s a Ph.D. canddate n electrcal engneerng at Texas A&M Unversty. Hs research nterests are neural networks, fuzzy logc, genetc algorthms, multvarable, robust and adaptve systems, and ther mplementaton n process control and pattern recognton, especally n power systems control, protecton and montorng. Mladen Kezunovc (S'77, M'80, SM'85, F'99) receved hs Dpl. Ing. degree from the Unversty of Sarajevo, the M.S. and Ph.D. degrees from the Unversty of Kansas, all n electrcal engneerng, n 1974, 1977 and 1980, respectvely. He has been wth Texas A&M Unversty snce 1987 where he s the Eugene E. Webb Professor and Drector of Electrc Power and Power Electroncs Insttute. Hs man research nterests are dgtal smulators and smulaton methods for equpment evaluaton and testng as well as applcaton of ntellgent methods to control, protecton and power qualty montorng. Dr. Kezunovc s a regstered professonal engneer n Texas, and a Fellow of IEEE.

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

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

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

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

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

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

Unsupervised Learning

Unsupervised Learning Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and

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

Backpropagation: In Search of Performance Parameters

Backpropagation: In Search of Performance Parameters Bacpropagaton: In Search of Performance Parameters ANIL KUMAR ENUMULAPALLY, LINGGUO BU, and KHOSROW KAIKHAH, Ph.D. Computer Scence Department Texas State Unversty-San Marcos San Marcos, TX-78666 USA ae049@txstate.edu,

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

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

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

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

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

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

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

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

KOHONEN'S SELF ORGANIZING NETWORKS WITH "CONSCIENCE"

KOHONEN'S SELF ORGANIZING NETWORKS WITH CONSCIENCE Kohonen's Self Organzng Maps and ther use n Interpretaton, Dr. M. Turhan (Tury) Taner, Rock Sold Images Page: 1 KOHONEN'S SELF ORGANIZING NETWORKS WITH "CONSCIENCE" By: Dr. M. Turhan (Tury) Taner, Rock

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like: Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A

More information

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

Performance Evaluation of an ANFIS Based Power System Stabilizer Applied in Multi-Machine Power Systems

Performance Evaluation of an ANFIS Based Power System Stabilizer Applied in Multi-Machine Power Systems Performance Evaluaton of an ANFIS Based Power System Stablzer Appled n Mult-Machne Power Systems A. A GHARAVEISI 1,2 A.DARABI 3 M. MONADI 4 A. KHAJEH-ZADEH 5 M. RASHIDI-NEJAD 1,2,5 1. Shahd Bahonar Unversty

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

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

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively Regularized Nearest Points for Set Based Recognition Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,

More information

Concurrent Apriori Data Mining Algorithms

Concurrent Apriori Data Mining Algorithms Concurrent Apror Data Mnng Algorthms Vassl Halatchev Department of Electrcal Engneerng and Computer Scence York Unversty, Toronto October 8, 2015 Outlne Why t s mportant Introducton to Assocaton Rule Mnng

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

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

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

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

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

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

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT 3. - 5. 5., Brno, Czech Republc, EU APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT Abstract Josef TOŠENOVSKÝ ) Lenka MONSPORTOVÁ ) Flp TOŠENOVSKÝ

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

DYNAMIC NETWORK OF CONCEPTS FROM WEB-PUBLICATIONS

DYNAMIC NETWORK OF CONCEPTS FROM WEB-PUBLICATIONS DYNAMIC NETWORK OF CONCEPTS FROM WEB-PUBLICATIONS Lande D.V. (dwl@vst.net), IC «ELVISTI», NTUU «KPI» Snarsk A.A. (asnarsk@gmal.com), NTUU «KPI» The network, the nodes of whch are concepts (people's names,

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

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

Module Management Tool in Software Development Organizations

Module Management Tool in Software Development Organizations Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,

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

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

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

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 15

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 15 CS434a/541a: Pattern Recognton Prof. Olga Veksler Lecture 15 Today New Topc: Unsupervsed Learnng Supervsed vs. unsupervsed learnng Unsupervsed learnng Net Tme: parametrc unsupervsed learnng Today: nonparametrc

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

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

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

Study of Data Stream Clustering Based on Bio-inspired Model

Study of Data Stream Clustering Based on Bio-inspired Model , pp.412-418 http://dx.do.org/10.14257/astl.2014.53.86 Study of Data Stream lusterng Based on Bo-nspred Model Yngme L, Mn L, Jngbo Shao, Gaoyang Wang ollege of omputer Scence and Informaton Engneerng,

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

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

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

More information

Local Quaternary Patterns and Feature Local Quaternary Patterns

Local Quaternary Patterns and Feature Local Quaternary Patterns Local Quaternary Patterns and Feature Local Quaternary Patterns Jayu Gu and Chengjun Lu The Department of Computer Scence, New Jersey Insttute of Technology, Newark, NJ 0102, USA Abstract - Ths paper presents

More information

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

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

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

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

MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS XUNYU PAN

MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS XUNYU PAN MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS by XUNYU PAN (Under the Drecton of Suchendra M. Bhandarkar) ABSTRACT In modern tmes, more and more

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

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

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

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

More information

Review of approximation techniques

Review of approximation techniques CHAPTER 2 Revew of appromaton technques 2. Introducton Optmzaton problems n engneerng desgn are characterzed by the followng assocated features: the objectve functon and constrants are mplct functons evaluated

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

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

Fuzzy Modeling of the Complexity vs. Accuracy Trade-off in a Sequential Two-Stage Multi-Classifier System

Fuzzy Modeling of the Complexity vs. Accuracy Trade-off in a Sequential Two-Stage Multi-Classifier System Fuzzy Modelng of the Complexty vs. Accuracy Trade-off n a Sequental Two-Stage Mult-Classfer System MARK LAST 1 Department of Informaton Systems Engneerng Ben-Guron Unversty of the Negev Beer-Sheva 84105

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

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

VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) , Fax: (370-5) ,

VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) , Fax: (370-5) , VRT012 User s gude V0.1 Thank you for purchasng our product. We hope ths user-frendly devce wll be helpful n realsng your deas and brngng comfort to your lfe. Please take few mnutes to read ths manual

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

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

A Clustering Algorithm for Chinese Adjectives and Nouns 1

A Clustering Algorithm for Chinese Adjectives and Nouns 1 Clusterng lgorthm for Chnese dectves and ouns Yang Wen, Chunfa Yuan, Changnng Huang 2 State Key aboratory of Intellgent Technology and System Deptartment of Computer Scence & Technology, Tsnghua Unversty,

More information

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,

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

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks Seventh Internatonal Conference on Intellgent Systems Desgn and Applcatons GA-Based Learnng Algorthms to Identfy Fuzzy Rules for Fuzzy Neural Networks K Almejall, K Dahal, Member IEEE, and A Hossan, Member

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

AUTOMATIC ROAD EXTRACTION FROM HIGH RESOLUTION SATELLITE IMAGES USING NEURAL NETWORKS, TEXTURE ANALYSIS, FUZZY CLUSTERING AND GENETIC ALGORITHMS

AUTOMATIC ROAD EXTRACTION FROM HIGH RESOLUTION SATELLITE IMAGES USING NEURAL NETWORKS, TEXTURE ANALYSIS, FUZZY CLUSTERING AND GENETIC ALGORITHMS AUTOMATIC ROAD EXTRACTION FROM HIGH RESOLUTION SATELLITE IMAGES USING NEURAL NETWORKS, TEXTURE ANALYSIS, FUZZY CLUSTERING AND GENETIC ALGORITHMS M Mokhtarzade a, *, M J Valadan Zoej b, H Ebad b a Dept

More information

A Hybrid Genetic Algorithm for Routing Optimization in IP Networks Utilizing Bandwidth and Delay Metrics

A Hybrid Genetic Algorithm for Routing Optimization in IP Networks Utilizing Bandwidth and Delay Metrics A Hybrd Genetc Algorthm for Routng Optmzaton n IP Networks Utlzng Bandwdth and Delay Metrcs Anton Redl Insttute of Communcaton Networks, Munch Unversty of Technology, Arcsstr. 21, 80290 Munch, Germany

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

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

Unsupervised Learning and Clustering

Unsupervised Learning and Clustering Unsupervsed Learnng and Clusterng Supervsed vs. Unsupervsed Learnng Up to now we consdered supervsed learnng scenaro, where we are gven 1. samples 1,, n 2. class labels for all samples 1,, n Ths s also

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

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

Parallel matrix-vector multiplication

Parallel matrix-vector multiplication Appendx A Parallel matrx-vector multplcaton The reduced transton matrx of the three-dmensonal cage model for gel electrophoress, descrbed n secton 3.2, becomes excessvely large for polymer lengths more

More information

IMAGE FUSION TECHNIQUES

IMAGE FUSION TECHNIQUES Int. J. Chem. Sc.: 14(S3), 2016, 812-816 ISSN 0972-768X www.sadgurupublcatons.com IMAGE FUSION TECHNIQUES A Short Note P. SUBRAMANIAN *, M. SOWNDARIYA, S. SWATHI and SAINTA MONICA ECE Department, Aarupada

More information

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING An Improved K-means Algorthm based on Cloud Platform for Data Mnng Bn Xa *, Yan Lu 2. School of nformaton and management scence, Henan Agrcultural Unversty, Zhengzhou, Henan 450002, P.R. Chna 2. College

More information

Object-Based Techniques for Image Retrieval

Object-Based Techniques for Image Retrieval 54 Zhang, Gao, & Luo Chapter VII Object-Based Technques for Image Retreval Y. J. Zhang, Tsnghua Unversty, Chna Y. Y. Gao, Tsnghua Unversty, Chna Y. Luo, Tsnghua Unversty, Chna ABSTRACT To overcome the

More information

The Comparison of Calibration Method of Binocular Stereo Vision System Ke Zhang a *, Zhao Gao b

The Comparison of Calibration Method of Binocular Stereo Vision System Ke Zhang a *, Zhao Gao b 3rd Internatonal Conference on Materal, Mechancal and Manufacturng Engneerng (IC3ME 2015) The Comparson of Calbraton Method of Bnocular Stereo Vson System Ke Zhang a *, Zhao Gao b College of Engneerng,

More information

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT Journal of Theoretcal and Appled Informaton Technology 30 th Aprl 013. Vol. 50 No.3 005-013 JATIT & LLS. All rghts reserved. ISSN: 199-8645 www.jatt.org E-ISSN: 1817-3195 THE PATH PLANNING ALGORITHM AND

More information

Recognition of Handwritten Numerals Using a Combined Classifier with Hybrid Features

Recognition of Handwritten Numerals Using a Combined Classifier with Hybrid Features Recognton of Handwrtten Numerals Usng a Combned Classfer wth Hybrd Features Kyoung Mn Km 1,4, Joong Jo Park 2, Young G Song 3, In Cheol Km 1, and Chng Y. Suen 1 1 Centre for Pattern Recognton and Machne

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

A Lazy Ensemble Learning Method to Classification

A Lazy Ensemble Learning Method to Classification IJCSI Internatonal Journal of Computer Scence Issues, Vol. 7, Issue 5, September 2010 ISSN (Onlne): 1694-0814 344 A Lazy Ensemble Learnng Method to Classfcaton Haleh Homayoun 1, Sattar Hashem 2 and Al

More information

Available online at Available online at Advanced in Control Engineering and Information Science

Available online at   Available online at   Advanced in Control Engineering and Information Science Avalable onlne at wwwscencedrectcom Avalable onlne at wwwscencedrectcom Proceda Proceda Engneerng Engneerng 00 (2011) 15000 000 (2011) 1642 1646 Proceda Engneerng wwwelsevercom/locate/proceda Advanced

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

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

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

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between

More information

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems A Unfed Framework for Semantcs and Feature Based Relevance Feedback n Image Retreval Systems Ye Lu *, Chunhu Hu 2, Xngquan Zhu 3*, HongJang Zhang 2, Qang Yang * School of Computng Scence Smon Fraser Unversty

More information

Research of Neural Network Classifier Based on FCM and PSO for Breast Cancer Classification

Research of Neural Network Classifier Based on FCM and PSO for Breast Cancer Classification Research of Neural Network Classfer Based on FCM and PSO for Breast Cancer Classfcaton Le Zhang 1, Ln Wang 1, Xujewen Wang 2, Keke Lu 2, and Ajth Abraham 3 1 Shandong Provncal Key Laboratory of Network

More information