A temporal fusion algorithm for multi-sensor tracking in wide areas

Size: px
Start display at page:

Download "A temporal fusion algorithm for multi-sensor tracking in wide areas"

Transcription

1 A emporal fuson algorhm for mulsensor rackng n wde areas O. Wallar, C. Moamed, M. Benjelloun Unversé du Loral Côe d'opale Laboraore ASL : 95 av P.L.Kng Calas, FRANCE Olver.Wallar@laslgw.unvloral.fr Absrac Ths arcle presens a dsrbued vson sysem for rackng of moble objecs over wde areas. A emporal daa fuson s used n order o mprove he decson makng a he daa assocaon sage. The emporal fuson s performed wh a possblsc M.H.T. (Mulple Hypohess Trackng). We have decded o use he Possbly Theory whch handle effcenly unceranes. The orgnaly of hs M.H.T. s he conrol of s developmen accordng o he onlne qualy esmaon of he daa assocaon based on a Necessy measuremen. Keywords: Dsrbued sensors, Temporal daa fuson, Mulple Hypohess Trackng, Possbly heory Inroducon The am of our research concerns he survellance of wde areas by a mulsensor approach. Works are focused on he rackng of moble objecs wh a Dsrbued Vson Sysem (DVS). Applcaons are mulple and varous, as monorng of sgnfcan ses (nuclear hermal power), conrol and esmaon of flows (arpor, por, moorway), n our case concerns ndvdual car rackng on hghways or n urban area. Our DVS s based on a cooperaon of geographcally dsrbued fxed sensors whch observe he scene. Due o economc and compuaonal consrans, a lmed number of sensors s used and make mpossble o fully cover he scene. So sensors are separaed by blnd areas.e., are as whch we do no allow any observaon. The man problem for he nerpreaon sysem s o mach objecs from a sensor wh anoher one. Moreover several dffcules may occur. Lack of nformaon resulng from blnd areas We are parcularly concerned by movng objecs for whch we canno defne an exac behavoral model as human acves. A hrd dffculy les n he fac ha n he ransporaon applcaons area, here are numerous cases where objecs are vsually close o each oher hus creang mporan ambgues a he daa assocaon sage. In hs suaon, he global moon nerpreaon becomes naurally a dffcul ask because has o suppor he ncompleeness and uncerany of he avalable nformaon Our approach frs res o generae coarse emporal consrans from conexual knowledge, reducng he complexy of he observed dynamc scene. Then a emporal mulsensor fuson approach s used o help he decson makng a he daa assocaon sage. A MHT algorhm s proposed. Is orgnaly s he conrol of s developmen. For us he defnon of daa fuson consss n he combnaon of mulple complemenary and redundan ses of nformaon n order o mprove he decson makng. For a dsrbued mulsensor archecure wh no common feld of vew, he daa fuson has o work manly wh he complemenary of observaons over me and over feld of sensors. Is objecve s o solve ambgues a he daa assocaon sage. From a daa fuson pon of vew we have decded o use he Possbly Theory. Ths formal approach nroduced by Zadeh [] s based on he noon of fuzzy ses. Inuvely, n hs formalsm when a fac s enrely possble here s no surprse lnk o s occurrence. There s a heursc lnk beween probably and he possbly, snce an nformaon s mpossble, s lkely o be mprobable, however a hgh degree of possbly does no mply a hgh degree of probably. The possblsc approach s parcularly convenen when he sysem has o deal wh coarse model and ncomplee nformaon because can effcenly represen s gnorance. In hs heory each fac s assocaed wh a degree of possbly and a degree of necessy. The necessy expresses he cerany of a fac wh akng no accoun all evenuales. A hgh degree of possbly means ha he fac s fully possble. A low degree of necessy wh a hgh degree of possbly, means ha he sysem knows nohng abou he fac [2][3]. In addon n framework of he possbly heory, combnaon operaors are very large and sar from an opmsc o pessmsc behavor [4]. The user can effcenly choose s operaor a each decson level.

2 In hs paper we presen frs our emporal fuson approach, he DVS. And hen, we explan how he emporal fuson s conrolled. Fnally, we presen some resuls over a real world sequence. 2 Temporal fuson The emporal fuson consss o combne nformaon acqured a dfferen nsans and hen o decde. I mples ha he sysem mus be able o predc objecs sae a each nsan (see Fgure ). sensor sensor 2 I I2 I : nformaon I3 I4 In our mulsensor sysem, when a sensor perceves an observaon, res o make a decson. For hs, focuses on objecs whch have he possbly o appear a hs momen. Ths sage s based on he moon predcons of moble objecs. If he decson qualy s suffcen hen he fuson s sopped and he decson s made (see fgure2). The qualy s esmaed by a necessy measuremen. In our case, he machng decson s realzed when he necessy measuremen of an hypohess s hgh. Ths measure akes no accoun he possbly measuremen of all he hypoheses. The necessy of an hypohess s defned by : N H ) = P( H ) () ( Wh Ω = all hypoheses H = Ω { H P ( H ) max P ( H ) = j } H j H percepon decson predcon decson Fgure : An example of emporal fuson Among works based on emporal fuson approaches, we can quoe hose of M. Rombau [5] whch are assocaed whn he framework of he european projec Promeheus. Two sensors are placed on a moble vehcle, he frs sensor s locaed n fron of he vehcle and he oher one n he back. Blnd zones exs a lef and rgh sdes. The sysem res o esmae he close envronmen. In parcular, o fnd objecs deeced by he frs sensor on he level of he second one. A sgnfcan aspec of hs fuson concerns he mechansms of predcon, makng possble o brng ogeher nformaon ssued from he sensors. One can also quoe works of A.Nfle [6] whch res o classfy mssles by analyzng her dynamc behavor. The recognon s based on he observaon of known daed evens. Our emporal fuson breaks up no 2 levels. A he frs level each sensor res o decde wh s own nformaon. A he second level he decson s obaned by combnng nformaon acqured by several sensors. The crossng of he s level o he 2nd one s based on he qualy of he s level decsons. The esmaon of hs qualy akes no accoun he whole of nformaon whch has been combned. 2. Frs level predcons (DOPs) sensor "" 2.2 Second level Fgure 2 : s level of fuson If ambgues occur,.e. ha he qualy of he curren decsons s consdered o be nsuffcen, hen he sensor whch has deeced he ambgues, collec complemenary nformaon from close sensors. Ths fuson s carred ou n me as long as he qualy of he decsons s low and as long as he sysem s lkely o oban useful nformaon. The conrol n me of hs fuson wll be presened more n deal n secon 4. We now descrbe our DVS as well as he coarse emporal consrans from conexual knowledge whch are essenal o reduce he complexy of he rackng process. 3 Presenaon of he DVS 3. The archecure Use of mulple sensors gves rse o one problem: how o connec all he sensors ogeher (organzaon and archecure).

3 percepon wh ones, whch are lkely o be perceved, called awaed objec ". predcons (DOPs)! sensor ""... decson percepon predcons (DOPs) sensor "j" Fgure 4. Tracks managemen useful sensors Fgure 3 : 2nd level of fuson We have decded o use a decenralsed archecure. Ths laer has no cenral processng facly, no cenralsed communcaons medum. The srucure of hs archecure s equvalen o a nework of nellgen sensor nodes. Each sensor node s auonomous; has s own processng un and s own communcaons facles. Communcaon can ake place beween any wo conneced sensor nodes. Each node can assmlae and receve nformaon ndependenly of oher nodes. Ths ype of archecure has many advanages [7]. Among he prncpal ones, we can quoe he fac ha s compleely modular, ha ensures he maxmum benef derved from he use of mulple sensors. In parcular, s robus o he loss of sensors. I can use dfferen varees of sensors workng smulaneously. Tracks assocaed wh a moble objec (nalsaon, manenance and ermnaon) are managed by he sensor ha has nalsed hem. When a moble objec s perceved (fgure 4a) by a sensor, emporary racks assocaed wh possble rajecores of he objec n blnd zones are nalsed (fgure 4b). If a close sensor recognses he objec (fgure 4c), hen he sensor whch has nalsed emporary racks s nformed (fgure 4d) n order o valdae he curren rack connecng he wo sensors and o remove he ohers (fgure 4e). Ths managemen mode has been movaed n order o acheve good racks ermnaon. To realse he rack managemen, s essenal ha sensors are able frsly, o predc moon of moble objecs n blnd zones and secondly, o mach perceved objecs 3.2 Compably measuremens When a sensor perceves an observaon, measures he compably of hs observaon agans awaed objecs by a se of measuremens. The machng decson wll be based on hese measuremens. The frs operaon carred ou by a sensor percevng an observaon s o exrac s prmves. The choce of prmves s mporan because objec machng s manly based on hem. These prmves mus be me nvaran. A prmve exraced by a sensor mus be logcally found by anoher one. Afer he exracon of he prmves, compably measuremens are compued for each awaed objec. They are he resuls of he combnaon of her vsual prmves and emporal compables. Each degree akes a value rankng beween 0 and ; value mean a full compably. We have esed several possblsc operaors. Our choce was dreced owards a ype of operaor supporng compably measuremens favourng some exreme suaons,.e. very srong compables or ncompables (" X². Y² "). 3.3 Temporal knowledge We have decded o use fuzzy emporal curves of evens descrbed by Dubos and Prade [8][9] (DOP: Doman Occurrence Possbly) n order o predc objec moon n blnd zones. Ths choce s movaed by he fac ha we work n wde oudoor scenes ncludng blnd zones. Under such condons, we mus be able o manage unceranes relaed o he sysem. I s hus necessary o buld rough models olerang ranges of varaon for he numercal parameers.

4 A DOP(O,Sj,Sk) s he predcon generaed by he sensor Sj, explanng he emporal appearance possbly of a recognsed objec O n he feld of a specfc sensor Sk (fgure 5). possbly measuremen _mn Fgure 5 : A DOP As soon as an awaed objec s deeced by a sensor, hs laer predcs all he fuure moon of he moble objec by generang some DOPs. The number of predcons depends of he number of neghbour sensors lkely o perceve he moble objec. The generaon done, he DOPs are ransmed o hese sensors (fgure 6). We have developed a mehod allowng auomac generaon of DOP accordng o hs knowledge. The creaon of a DOP depends on he conex and dynamc characerscs of he moble objec. For our dsrbued nerpreaon sysem, he conex s based upon four knowledge whch are spaal and workng confguraon of he scene, class nformaon for movng objecs, mage acquson nformaon and dynamc envronmen [9]. C0 DOP(,0,) C DOP(,0,2) Fgure 6 : DOPs generaon 4 The conrol of he fuson 4. Inroducon In order o realze daa assocaon handlng he ambgues, we have developed a emporal mulsensor fuson algorhm based on a mulple hypohess approach : a possblsc MHT algorhm. The orgnaly of hs MHT s he conrol of s developmen accordng o he onlne qualy esmaon of he daa assocaon based on a Necessy measuremen. C2 The underlyng prncple s ha he daa assocaon decson s delayed as long as he confdence level n a hypohess s no sgnfcan enough comparavely o he oher ones. Ths confdence s an ndcaor of he qualy of he daa assocaon. Ths adapave approach has been chosen because one of he man problems regardng MHT s o know when should be sopped n order o realse an effcen daa assocaon. In [0], Cox presens a MHT algorhm o rack vsual prmves. He assumes ha hree observaons are suffcen n s applcaon. Accordng o our conex, a fxed number seems no o be convenen. Indeed, n some suaons, many moble objecs may appear smulaneously n fron of a sensor. Ths ype of suaon nduces many ambgues and s necessary o acqure addonal nformaon n order o be able o make he rgh machng decson. Ths could be obaned by a hgher developmen of he MHT. Thus, we have developed a necessy measuremen expressng he qualy of he daa assocaon a each new observaon. Wh hs measuremen, we can conrol he naon (low necessy) and he ermnaon (hgh necessy) of he MHT. Whn he conex of mulple hypoheses, he wo man algorhms are he JPDA (Jon Probablsc Daa Assocaon) fler [] and he MHT (Mulple Hypohess Trackng) [2]. These wo las ouperform sngle hypohess echnques bu when he problem sze ncreases, he requred compuaon ncreases exponenally. The JPDA, n rackng of closely spaced arges, has poor performance because of he perssen nerference from neghbourng arges. The MHT mehod s a mulple scan mehod. All nformaon concernng he assocaons beween moble objecs and he observaons s sored n order o solve ambguous suaons. The assocaons are defned as hypoheses. Ths mehod manans n parallel a se of hypoheses ha represens compeve worlds. In order o reduce he compuaonal complexy of he MHT, several opmsaon mehods have been developed. Among he prncpal ones we can menon hose whch reduce he number of hypoheses by mananng only he kbes ones. Cox has developed a MHT deermnng he kbes assgnmens n a polynomal me due o he use of he Mury algorhm. [0]. The choce of a Mulple Hypohess echnque was movaed by he fac ha we are n our case confroned wh many ambgues 4.2 Ambgues These ambgues are classfed no hree caegores, as lsed below. * Ambgues of assocaon Our sysem mus be able o solve ambgues relaed o daa assocaon. They occur when several moble objecs are emporally and vsually compable wh an

5 observaon. Such ambgues occur naurally n cluered envronmens. * Moonrelaed ambgues Moonrelaed ambgues are manly due o he lack of nformaon assocaed wh blnd zones and he absence of accurae behavoural model of movng objecs. Indeed, as no observaons are perceved whn he blnd zones, s very dffcul o defne precsely whch sensor wll perceve a moble objec, whch jus has been deeced. * Ambguous suaons due o new objecs The appearance of a new objec vsually close o an awaed objec, may generae an ambguy. 4.3 The MHT descrpon We remnd ha when an awaed objec s deeced by a sensor, hs laer predcs all he fuure dsplacemens of he moble objec by generang some DOPs. A he begnnng of each daa assocaon ambguy, a ree descrbng all he feasble hypoheses resulng from he observaons s generaed, wh he consran ha each observaon ares from a sngle (exsng or new) arge Each MHT ree examnes all he plausble combnaons, wh he consran ha each observaon arses from eher a sngle (exsng or new) arge. In he ree, each leaf node represens a feasble hypohess. For each branch, a possbly measuremen s assgned. Ths laer s obaned by a combnaon of he compably measuremens of he hypoheses ncluded n he branch by a mean operaor. The Tnorms and Tconorms combnaon operaors are no suable because he combnaon has no o be oo ndulgen or cauous. The machng decson s performed, as n he frs level fuson sep (secon2.), when he necessy measuremen (c.f. formula ) of a branch s hgh. Ths measure akes no accoun he possbly measuremen of all he branches. Such necessy measuremen s movaed by he fac ha an hypohess wll be seleced f s possbly measuremen s hgh and f hs value s dscrmnang wh respec o oher hypoheses. If an awaed objec does no appear n fron of a sensor feld afer a gven perod, hen hanks o he DOP assocaed wh hs objec, whch has s own lfespan, he rack wll be ermnaed. In hs suaon, curren MHTs remove naurally hese hypoheses. 4.4 Machng compably of new objecs The esmaon of machng compably (C j ) of a new objec (H ) wh an observaon j akes no accoun he compably measuremens of hs observaon wh he possble moble objecs (2). If several objecs are emporally and vsually compable wh an observaon, seems naural o hnk ha he observaon could be assocaed o one of hese objecs. Consequenly, he machng compably (MC) for a new objec s low, oherwse we ncrease he compably of new objec. O Φ, C ( H ) = max C j ( O ) j O Φ Φ:moble objecs lkely o be assocaed (2) wh he observaon j 4.5 Implemenaon In our envsaged applcaons workng wh dsan sensors, he majory of objecs assocaons are realsed before hey appear n fron of nex sensors. Oherwse, an ambguy persss, he developmen of he ree s sopped. In hs case, he algorhm valdae he mos plausble hypohess even he necessy measuremen s low. In complemen of our MHT, we use a hgh level exper. Is am s o correc some errors ha he PMHT s unable o manage. The reason s due o he fac ha he PMHT consder ha he predcons and he feaures exracon are rgh. The wo ypes of errors are he false alarms and he predcon errors. The false alarms may occur n nosed envronmens. The predcon errors are generaed when an unexpeced even slow down or speed up an objec moon. The hgh level exper, focus s aenon o new objecs ha he MHT has decded, f possble o ry o assocae o objecs for whch her racks have been jus ermnaed. 4.6 An example An example on fgure 7 shows hree objecs (O,,) and hree observaons (Obs,Obs2,Obs3). The purpose of hs example s frsly o show ha he problem of daa assocaon s no necessarly easy, and secondly o show how nformaon fuson can help o ease he decsonmakng process. The DVS has o mach on sensors C and C2 he hree objecs, whch have been deeced by he sensor C0. In hs example, as he hree objecs are vsually close o each oher, he machng decson s manly based on he emporal compably measuremens. In he realy, he compably measuremens are based on a combnaon of emporal and vsual compably measuremens. On fgure 8, The DOPs assocaed o he moble objecs and he appearance daes of he observaons are represened. The resulng MHT s shown on fgure 9. The hypohess corresponds o a poenal new objec permng a rack naon. The frs level of he ree represens all he feasble assocaons of he observaon Obs. There are hree branches because wo awaed objecs (O and ) are compable and a new objec s also possble. As he necessy measuremen s low because of wo hypoheses are possble, he MHT was he appearance of a new observaon n order o oban more nformaon o mprove a beer machngdecson. In hs example, he decson s realsed afer he

6 appearance of Obs3. Twenywo hypoheses have been generaed. The frs branch s seleced because s necessy measuremen s hgh (). C C2 {O,,} C0 5 Applcaon {Obs,Obs3} C {Obs2} C2 Fgure 7 : he sensors nework O O Obs Obs2 Obs3 Fgure 8 : Observaons appearances Ths s an applcaon of wde area survellance usng dsrbued sensors n a hghway envronmen ( fgure 0). The specfcy of hs applcaon s ha he spaal and workng confguraons of he observed scene are relavely well defned. The confguraon of each sensor s uned n order o make he objec recognon ask easer. Insde a closed segmen of he hghway, an objec seen by a camera has o appear n he feld of he nex sensor. The cameras are 3km dsan. Ths choce was aken accordng o he conex. A hgher dsance (nvolvng more naccuracy n he predcon) would enlarge he suppor of he DOPs and also decrease he qualy of he oal nformaon. In our applcaon, we used four classes of objecs, whch are rucks, vans, cars and moor cycles. Each local vson un has o exrac cnemac and nvaran feaures from each objec deeced. The deecon s based on a dfference beween he curren mage and an adapave reference mage whou movng objecs as Vannoorenberghe [3]. The nvaran feaures chosen for hs applcaon are normalsed color hsograms and calbraed 2D surface of objecs. The parameers used o classfy he objecs are surface and he cnemac behavor of objecs. Color compably beween objecs s based on he χ² dsance of her hsograms [3]. Obs Obs2 Obs3 0.7 O 0.8 O O 0.8 possbly measuremen Fgure 9 : resulng PMHT necessy measuremen We have esed our mehod n hs real envronmen wh wo onehour sequences (able ). The performance of he possblsc MHT s compared wh a sngle hypohess ( neares neghbour : NN) approach and evaluaed n erm of daa assocaon errors esmaed by a human operaor (able2). The NN algorhm assocaes he observaon wh he objec havng he sronges possbly

7 6 Concluson C C2 no sequence sequence Fgure 0 : Hgh way mulsensor rackng Observaons perceved by he sensor C O O4 O5 Dops : Temporal predcons of objecs O Fgure : DOPs generaed by sensor C number of vehcles 52 sequence number of ambguous vehcles 3 75 number of predcon errors 0 Table : sequences and 2 ( daa observed by a human operaor) algorhm N.N. M.H.T. M.H.T. wh hgh level exper Observaons perceved by he sensor C2 sequence 9 (2%) 7 (%) 0 (6%) Table 2 : machng errors 3 sequence 2 70 (22%) 45 (4%) 35 (%) Expermenal resuls show ha he NN approach and he MHT have an equvalen number of good assocaons n an uncluered envronmen (sequence ). The hgh level exper perms o mprove daa assocaon by solvng a par of predcon errors. In hs arcle, we have proposed a dsrbued vson sysem for rackng over wde area ncludng blnd zones. Our approach s based on he uncerany managemen of he global percepon sysem. A a frs level, concerns fuzzy emporal consran generaon expressng he naccuracy mprecson of he avalable behavoral model. Then a possblsc mehodology perms o model he sysem gnorance by akng no accoun all possble evenuales. And fnally a daa fuson approach based on a MHT s proposed o help machng decson. I works wh he complemenary of observaons over me and over feld of sensors. We have also proposed a soluon o conrol hs fuson algorhm over me. References [] L.A. Zadeh, Fuzzy ses as a bass for a heory of possbly, Fuzzy ses and sysems, (978), 328. [2] D. Dubos, H. Prade. Possbly heory, Applcaons o nformaon represenaon n nformac (n french). Masson, Pars, (988). [3] O. Wallar, C. Moamed, M. Benjelloun. A possblsc approach of hgh level rackng n a wde area, Proc Second The 2nd nernaonal conference on nformaon fuson fuson 99, Sunny Vale, USA, (999), [4] I.Bloch, Informaon combnaon operaors for daa fuson : a comparave revew wh classfcaon, IEEE ransacons on sysems, man and cybernecs 2 () (996) [5] M. Rombau & D. Mezel. Dynamc daa emporal mulsensor fuson n he Promeheus Prolab2 demonsraor. IEEE Inernaonal Conference on Robocs and Auomaon. San Dego,.May 83, (994) [6] A. Nfle, R. Reynaud. Behavor classfcaon based on evens occurrence n possbly heory (n french). Traemen du sgnal vol spécal 4, n 5 (997), [7] B.S.Y. Rao, H.F. DurranWhye & J.A. Sheen, A fully decenralzed mulsensor sysem for rackng and survellance, he nernaonal journal of robocs research,vol 2, n, (993), [8] D. Dubos And H. Prade,, Processng fuzzy emporal knowledge, IEEE ransacons on sysems man and cybernecs 9, (4), (989), [9] O. Wallar, C. Moamed & M. Benjelloun, Temporal knowledge for Cooperave Dsrbued Vson, sevenh IEE Conference on Image Processng and s applcaons, Mancheser,July (999) [0] I.J. Cox and S.L. Hngoran, An effcen mplemenaon of Red s mulple hypohess rackng algorhm and s evaluaon for he purpose of vsual rackng, IEEE ransacons on paern analyss and machne nellgence 8 (2), february 996,3850.

8 [] Y. BarShalom And T. Formann, 988, Trackng and daa assocaon, Academc press, London. [2] D.B. Red, An algorhm for rackng mulple arges, IEEE ransacons on auomac conrol AC24 (6), december (979), [3] P. Vannoorenberghe, C. Moamed and J.G. Posare, Updang a reference mage for deecng moon n urban scenes, Revue de raemen du sgnal 5 (2) (998)39 48.

Variants of Pegasos. December 11, 2009

Variants of Pegasos. December 11, 2009 Inroducon Varans of Pegasos SooWoong Ryu bshboy@sanford.edu December, 009 Youngsoo Cho yc344@sanford.edu Developng a new SVM algorhm s ongong research opc. Among many exng SVM algorhms, we wll focus on

More information

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS R&RATA # Vol.) 8, March FURTHER AALYSIS OF COFIDECE ITERVALS FOR LARGE CLIET/SERVER COMPUTER ETWORKS Vyacheslav Abramov School of Mahemacal Scences, Monash Unversy, Buldng 8, Level 4, Clayon Campus, Wellngon

More information

( ) () we define the interaction representation by the unitary transformation () = ()

( ) () we define the interaction representation by the unitary transformation () = () Hgher Order Perurbaon Theory Mchael Fowler 3/7/6 The neracon Represenaon Recall ha n he frs par of hs course sequence, we dscussed he chrödnger and Hesenberg represenaons of quanum mechancs here n he chrödnger

More information

This document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore.

This document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore. Ths documen s downloaded from DR-NTU, Nanyang Technologcal Unversy Lbrary, Sngapore. Tle A smplfed verb machng algorhm for word paron n vsual speech processng( Acceped verson ) Auhor(s) Foo, Say We; Yong,

More information

On One Analytic Method of. Constructing Program Controls

On One Analytic Method of. Constructing Program Controls Appled Mahemacal Scences, Vol. 9, 05, no. 8, 409-407 HIKARI Ld, www.m-hkar.com hp://dx.do.org/0.988/ams.05.54349 On One Analyc Mehod of Consrucng Program Conrols A. N. Kvko, S. V. Chsyakov and Yu. E. Balyna

More information

Robustness Experiments with Two Variance Components

Robustness Experiments with Two Variance Components Naonal Insue of Sandards and Technology (NIST) Informaon Technology Laboraory (ITL) Sascal Engneerng Dvson (SED) Robusness Expermens wh Two Varance Componens by Ana Ivelsse Avlés avles@ns.gov Conference

More information

Filtrage particulaire et suivi multi-pistes Carine Hue Jean-Pierre Le Cadre and Patrick Pérez

Filtrage particulaire et suivi multi-pistes Carine Hue Jean-Pierre Le Cadre and Patrick Pérez Chaînes de Markov cachées e flrage parculare 2-22 anver 2002 Flrage parculare e suv mul-pses Carne Hue Jean-Perre Le Cadre and Parck Pérez Conex Applcaons: Sgnal processng: arge rackng bearngs-onl rackng

More information

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005 Dynamc Team Decson Theory EECS 558 Proec Shruvandana Sharma and Davd Shuman December 0, 005 Oulne Inroducon o Team Decson Theory Decomposon of he Dynamc Team Decson Problem Equvalence of Sac and Dynamc

More information

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!") i+1,q - [(!

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!) i+1,q - [(! ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL The frs hng o es n wo-way ANOVA: Is here neracon? "No neracon" means: The man effecs model would f. Ths n urn means: In he neracon plo (wh A on he horzonal

More information

THE PREDICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS

THE PREDICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS THE PREICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS INTROUCTION The wo dmensonal paral dfferenal equaons of second order can be used for he smulaon of compeve envronmen n busness The arcle presens he

More information

A Novel Object Detection Method Using Gaussian Mixture Codebook Model of RGB-D Information

A Novel Object Detection Method Using Gaussian Mixture Codebook Model of RGB-D Information A Novel Objec Deecon Mehod Usng Gaussan Mxure Codebook Model of RGB-D Informaon Lujang LIU 1, Gaopeng ZHAO *,1, Yumng BO 1 1 School of Auomaon, Nanjng Unversy of Scence and Technology, Nanjng, Jangsu 10094,

More information

Time-interval analysis of β decay. V. Horvat and J. C. Hardy

Time-interval analysis of β decay. V. Horvat and J. C. Hardy Tme-nerval analyss of β decay V. Horva and J. C. Hardy Work on he even analyss of β decay [1] connued and resuled n he developmen of a novel mehod of bea-decay me-nerval analyss ha produces hghly accurae

More information

Abstract This paper considers the problem of tracking objects with sparsely located binary sensors. Tracking with a sensor network is a

Abstract This paper considers the problem of tracking objects with sparsely located binary sensors. Tracking with a sensor network is a Trackng on a Graph Songhwa Oh and Shankar Sasry Deparmen of Elecrcal Engneerng and Compuer Scences Unversy of Calforna, Berkeley, CA 9470 {sho,sasry}@eecs.berkeley.edu Absrac Ths paper consders he problem

More information

Robust and Accurate Cancer Classification with Gene Expression Profiling

Robust and Accurate Cancer Classification with Gene Expression Profiling Robus and Accurae Cancer Classfcaon wh Gene Expresson Proflng (Compuaonal ysems Bology, 2005) Auhor: Hafeng L, Keshu Zhang, ao Jang Oulne Background LDA (lnear dscrmnan analyss) and small sample sze problem

More information

Linear Response Theory: The connection between QFT and experiments

Linear Response Theory: The connection between QFT and experiments Phys540.nb 39 3 Lnear Response Theory: The connecon beween QFT and expermens 3.1. Basc conceps and deas Q: ow do we measure he conducvy of a meal? A: we frs nroduce a weak elecrc feld E, and hen measure

More information

An introduction to Support Vector Machine

An introduction to Support Vector Machine An nroducon o Suppor Vecor Machne 報告者 : 黃立德 References: Smon Haykn, "Neural Neworks: a comprehensve foundaon, second edon, 999, Chaper 2,6 Nello Chrsann, John Shawe-Tayer, An Inroducon o Suppor Vecor Machnes,

More information

Bayes rule for a classification problem INF Discriminant functions for the normal density. Euclidean distance. Mahalanobis distance

Bayes rule for a classification problem INF Discriminant functions for the normal density. Euclidean distance. Mahalanobis distance INF 43 3.. Repeon Anne Solberg (anne@f.uo.no Bayes rule for a classfcaon problem Suppose we have J, =,...J classes. s he class label for a pxel, and x s he observed feaure vecor. We can use Bayes rule

More information

Cubic Bezier Homotopy Function for Solving Exponential Equations

Cubic Bezier Homotopy Function for Solving Exponential Equations Penerb Journal of Advanced Research n Compung and Applcaons ISSN (onlne: 46-97 Vol. 4, No.. Pages -8, 6 omoopy Funcon for Solvng Eponenal Equaons S. S. Raml *,,. Mohamad Nor,a, N. S. Saharzan,b and M.

More information

Learning Objectives. Self Organization Map. Hamming Distance(1/5) Introduction. Hamming Distance(3/5) Hamming Distance(2/5) 15/04/2015

Learning Objectives. Self Organization Map. Hamming Distance(1/5) Introduction. Hamming Distance(3/5) Hamming Distance(2/5) 15/04/2015 /4/ Learnng Objecves Self Organzaon Map Learnng whou Exaples. Inroducon. MAXNET 3. Cluserng 4. Feaure Map. Self-organzng Feaure Map 6. Concluson 38 Inroducon. Learnng whou exaples. Daa are npu o he syse

More information

Clustering (Bishop ch 9)

Clustering (Bishop ch 9) Cluserng (Bshop ch 9) Reference: Daa Mnng by Margare Dunham (a slde source) 1 Cluserng Cluserng s unsupervsed learnng, here are no class labels Wan o fnd groups of smlar nsances Ofen use a dsance measure

More information

Reactive Methods to Solve the Berth AllocationProblem with Stochastic Arrival and Handling Times

Reactive Methods to Solve the Berth AllocationProblem with Stochastic Arrival and Handling Times Reacve Mehods o Solve he Berh AllocaonProblem wh Sochasc Arrval and Handlng Tmes Nsh Umang* Mchel Berlare* * TRANSP-OR, Ecole Polyechnque Fédérale de Lausanne Frs Workshop on Large Scale Opmzaon November

More information

Li An-Ping. Beijing , P.R.China

Li An-Ping. Beijing , P.R.China A New Type of Cpher: DICING_csb L An-Png Bejng 100085, P.R.Chna apl0001@sna.com Absrac: In hs paper, we wll propose a new ype of cpher named DICING_csb, whch s derved from our prevous sream cpher DICING.

More information

Outline. Probabilistic Model Learning. Probabilistic Model Learning. Probabilistic Model for Time-series Data: Hidden Markov Model

Outline. Probabilistic Model Learning. Probabilistic Model Learning. Probabilistic Model for Time-series Data: Hidden Markov Model Probablsc Model for Tme-seres Daa: Hdden Markov Model Hrosh Mamsuka Bonformacs Cener Kyoo Unversy Oulne Three Problems for probablsc models n machne learnng. Compung lkelhood 2. Learnng 3. Parsng (predcon

More information

Econ107 Applied Econometrics Topic 5: Specification: Choosing Independent Variables (Studenmund, Chapter 6)

Econ107 Applied Econometrics Topic 5: Specification: Choosing Independent Variables (Studenmund, Chapter 6) Econ7 Appled Economercs Topc 5: Specfcaon: Choosng Independen Varables (Sudenmund, Chaper 6 Specfcaon errors ha we wll deal wh: wrong ndependen varable; wrong funconal form. Ths lecure deals wh wrong ndependen

More information

doi: info:doi/ /

doi: info:doi/ / do: nfo:do/0.063/.322393 nernaonal Conference on Power Conrol and Opmzaon, Bal, ndonesa, -3, June 2009 A COLOR FEATURES-BASED METHOD FOR OBJECT TRACKNG EMPLOYNG A PARTCLE FLTER ALGORTHM Bud Sugand, Hyoungseop

More information

Fall 2010 Graduate Course on Dynamic Learning

Fall 2010 Graduate Course on Dynamic Learning Fall 200 Graduae Course on Dynamc Learnng Chaper 4: Parcle Flers Sepember 27, 200 Byoung-Tak Zhang School of Compuer Scence and Engneerng & Cognve Scence and Bran Scence Programs Seoul aonal Unversy hp://b.snu.ac.kr/~bzhang/

More information

2.1 Constitutive Theory

2.1 Constitutive Theory Secon.. Consuve Theory.. Consuve Equaons Governng Equaons The equaons governng he behavour of maerals are (n he spaal form) dρ v & ρ + ρdv v = + ρ = Conservaon of Mass (..a) d x σ j dv dvσ + b = ρ v& +

More information

WiH Wei He

WiH Wei He Sysem Idenfcaon of onlnear Sae-Space Space Baery odels WH We He wehe@calce.umd.edu Advsor: Dr. Chaochao Chen Deparmen of echancal Engneerng Unversy of aryland, College Par 1 Unversy of aryland Bacground

More information

Detection of Waving Hands from Images Using Time Series of Intensity Values

Detection of Waving Hands from Images Using Time Series of Intensity Values Deecon of Wavng Hands from Images Usng Tme eres of Inensy Values Koa IRIE, Kazunor UMEDA Chuo Unversy, Tokyo, Japan re@sensor.mech.chuo-u.ac.jp, umeda@mech.chuo-u.ac.jp Absrac Ths paper proposes a mehod

More information

DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL

DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL Sco Wsdom, John Hershey 2, Jonahan Le Roux 2, and Shnj Waanabe 2 Deparmen o Elecrcal Engneerng, Unversy o Washngon, Seale, WA, USA

More information

CHAPTER 10: LINEAR DISCRIMINATION

CHAPTER 10: LINEAR DISCRIMINATION CHAPER : LINEAR DISCRIMINAION Dscrmnan-based Classfcaon 3 In classfcaon h K classes (C,C,, C k ) We defned dscrmnan funcon g j (), j=,,,k hen gven an es eample, e chose (predced) s class label as C f g

More information

Introduction ( Week 1-2) Course introduction A brief introduction to molecular biology A brief introduction to sequence comparison Part I: Algorithms

Introduction ( Week 1-2) Course introduction A brief introduction to molecular biology A brief introduction to sequence comparison Part I: Algorithms Course organzaon Inroducon Wee -2) Course nroducon A bref nroducon o molecular bology A bref nroducon o sequence comparson Par I: Algorhms for Sequence Analyss Wee 3-8) Chaper -3, Models and heores» Probably

More information

Computing Relevance, Similarity: The Vector Space Model

Computing Relevance, Similarity: The Vector Space Model Compung Relevance, Smlary: The Vecor Space Model Based on Larson and Hears s sldes a UC-Bereley hp://.sms.bereley.edu/courses/s0/f00/ aabase Managemen Sysems, R. Ramarshnan ocumen Vecors v ocumens are

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 9 Hamlonan Equaons of Moon (Chaper 8) Wha We Dd Las Tme Consruced Hamlonan formalsm H ( q, p, ) = q p L( q, q, ) H p = q H q = p H = L Equvalen o Lagrangan formalsm Smpler, bu

More information

Particle Filter Based Robot Self-localization Using RGBD Cues and Wheel Odometry Measurements Enyang Gao1, a*, Zhaohua Chen1 and Qizhuhui Gao1

Particle Filter Based Robot Self-localization Using RGBD Cues and Wheel Odometry Measurements Enyang Gao1, a*, Zhaohua Chen1 and Qizhuhui Gao1 6h Inernaonal Conference on Elecronc, Mechancal, Informaon and Managemen (EMIM 206) Parcle Fler Based Robo Self-localzaon Usng RGBD Cues and Wheel Odomery Measuremens Enyang Gao, a*, Zhaohua Chen and Qzhuhu

More information

Department of Economics University of Toronto

Department of Economics University of Toronto Deparmen of Economcs Unversy of Torono ECO408F M.A. Economercs Lecure Noes on Heeroskedascy Heeroskedascy o Ths lecure nvolves lookng a modfcaons we need o make o deal wh he regresson model when some of

More information

Machine Learning 2nd Edition

Machine Learning 2nd Edition INTRODUCTION TO Lecure Sldes for Machne Learnng nd Edon ETHEM ALPAYDIN, modfed by Leonardo Bobadlla and some pars from hp://www.cs.au.ac.l/~aparzn/machnelearnng/ The MIT Press, 00 alpaydn@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/mle

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 9 Hamlonan Equaons of Moon (Chaper 8) Wha We Dd Las Tme Consruced Hamlonan formalsm Hqp (,,) = qp Lqq (,,) H p = q H q = p H L = Equvalen o Lagrangan formalsm Smpler, bu wce as

More information

CS286.2 Lecture 14: Quantum de Finetti Theorems II

CS286.2 Lecture 14: Quantum de Finetti Theorems II CS286.2 Lecure 14: Quanum de Fne Theorems II Scrbe: Mara Okounkova 1 Saemen of he heorem Recall he las saemen of he quanum de Fne heorem from he prevous lecure. Theorem 1 Quanum de Fne). Le ρ Dens C 2

More information

Solution in semi infinite diffusion couples (error function analysis)

Solution in semi infinite diffusion couples (error function analysis) Soluon n sem nfne dffuson couples (error funcon analyss) Le us consder now he sem nfne dffuson couple of wo blocks wh concenraon of and I means ha, n a A- bnary sysem, s bondng beween wo blocks made of

More information

Lecture 2 L n i e n a e r a M od o e d l e s

Lecture 2 L n i e n a e r a M od o e d l e s Lecure Lnear Models Las lecure You have learned abou ha s machne learnng Supervsed learnng Unsupervsed learnng Renforcemen learnng You have seen an eample learnng problem and he general process ha one

More information

ECE 366 Honors Section Fall 2009 Project Description

ECE 366 Honors Section Fall 2009 Project Description ECE 366 Honors Secon Fall 2009 Projec Descrpon Inroducon: Muscal genres are caegorcal labels creaed by humans o characerze dfferen ypes of musc. A muscal genre s characerzed by he common characerscs shared

More information

Genetic Algorithm in Parameter Estimation of Nonlinear Dynamic Systems

Genetic Algorithm in Parameter Estimation of Nonlinear Dynamic Systems Genec Algorhm n Parameer Esmaon of Nonlnear Dynamc Sysems E. Paeraks manos@egnaa.ee.auh.gr V. Perds perds@vergna.eng.auh.gr Ah. ehagas kehagas@egnaa.ee.auh.gr hp://skron.conrol.ee.auh.gr/kehagas/ndex.hm

More information

TSS = SST + SSE An orthogonal partition of the total SS

TSS = SST + SSE An orthogonal partition of the total SS ANOVA: Topc 4. Orhogonal conrass [ST&D p. 183] H 0 : µ 1 = µ =... = µ H 1 : The mean of a leas one reamen group s dfferen To es hs hypohess, a basc ANOVA allocaes he varaon among reamen means (SST) equally

More information

Anomaly Detection. Lecture Notes for Chapter 9. Introduction to Data Mining, 2 nd Edition by Tan, Steinbach, Karpatne, Kumar

Anomaly Detection. Lecture Notes for Chapter 9. Introduction to Data Mining, 2 nd Edition by Tan, Steinbach, Karpatne, Kumar Anomaly eecon Lecure Noes for Chaper 9 Inroducon o aa Mnng, 2 nd Edon by Tan, Senbach, Karpane, Kumar 2/14/18 Inroducon o aa Mnng, 2nd Edon 1 Anomaly/Ouler eecon Wha are anomales/oulers? The se of daa

More information

Chapter 6: AC Circuits

Chapter 6: AC Circuits Chaper 6: AC Crcus Chaper 6: Oulne Phasors and he AC Seady Sae AC Crcus A sable, lnear crcu operang n he seady sae wh snusodal excaon (.e., snusodal seady sae. Complee response forced response naural response.

More information

UNIVERSITAT AUTÒNOMA DE BARCELONA MARCH 2017 EXAMINATION

UNIVERSITAT AUTÒNOMA DE BARCELONA MARCH 2017 EXAMINATION INTERNATIONAL TRADE T. J. KEHOE UNIVERSITAT AUTÒNOMA DE BARCELONA MARCH 27 EXAMINATION Please answer wo of he hree quesons. You can consul class noes, workng papers, and arcles whle you are workng on he

More information

Algorithm Research on Moving Object Detection of Surveillance Video Sequence *

Algorithm Research on Moving Object Detection of Surveillance Video Sequence * Opcs and Phooncs Journal 03 3 308-3 do:0.436/opj.03.3b07 Publshed Onlne June 03 (hp://www.scrp.org/journal/opj) Algorhm Research on Movng Objec Deecon of Survellance Vdeo Sequence * Kuhe Yang Zhmng Ca

More information

Using Fuzzy Pattern Recognition to Detect Unknown Malicious Executables Code

Using Fuzzy Pattern Recognition to Detect Unknown Malicious Executables Code Usng Fuzzy Paern Recognon o Deec Unknown Malcous Execuables Code Boyun Zhang,, Janpng Yn, and Jngbo Hao School of Compuer Scence, Naonal Unversy of Defense Technology, Changsha 40073, Chna hnxzby@yahoo.com.cn

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4 CS434a/54a: Paern Recognon Prof. Olga Veksler Lecure 4 Oulne Normal Random Varable Properes Dscrmnan funcons Why Normal Random Varables? Analycally racable Works well when observaon comes form a corruped

More information

A Bayesian algorithm for tracking multiple moving objects in outdoor surveillance video

A Bayesian algorithm for tracking multiple moving objects in outdoor surveillance video A Bayesan algorhm for racng mulple movng obecs n oudoor survellance vdeo Manunah Narayana Unversy of Kansas Lawrence, Kansas manu@u.edu Absrac Relable racng of mulple movng obecs n vdes an neresng challenge,

More information

e-journal Reliability: Theory& Applications No 2 (Vol.2) Vyacheslav Abramov

e-journal Reliability: Theory& Applications No 2 (Vol.2) Vyacheslav Abramov June 7 e-ournal Relably: Theory& Applcaons No (Vol. CONFIDENCE INTERVALS ASSOCIATED WITH PERFORMANCE ANALYSIS OF SYMMETRIC LARGE CLOSED CLIENT/SERVER COMPUTER NETWORKS Absrac Vyacheslav Abramov School

More information

EEL 6266 Power System Operation and Control. Chapter 5 Unit Commitment

EEL 6266 Power System Operation and Control. Chapter 5 Unit Commitment EEL 6266 Power Sysem Operaon and Conrol Chaper 5 Un Commmen Dynamc programmng chef advanage over enumeraon schemes s he reducon n he dmensonaly of he problem n a src prory order scheme, here are only N

More information

FTCS Solution to the Heat Equation

FTCS Solution to the Heat Equation FTCS Soluon o he Hea Equaon ME 448/548 Noes Gerald Reckenwald Porland Sae Unversy Deparmen of Mechancal Engneerng gerry@pdxedu ME 448/548: FTCS Soluon o he Hea Equaon Overvew Use he forward fne d erence

More information

Performance Analysis for a Network having Standby Redundant Unit with Waiting in Repair

Performance Analysis for a Network having Standby Redundant Unit with Waiting in Repair TECHNI Inernaonal Journal of Compung Scence Communcaon Technologes VOL.5 NO. July 22 (ISSN 974-3375 erformance nalyss for a Nework havng Sby edundan Un wh ang n epar Jendra Sngh 2 abns orwal 2 Deparmen

More information

Math 128b Project. Jude Yuen

Math 128b Project. Jude Yuen Mah 8b Proec Jude Yuen . Inroducon Le { Z } be a sequence of observed ndependen vecor varables. If he elemens of Z have a on normal dsrbuon hen { Z } has a mean vecor Z and a varancecovarance marx z. Geomercally

More information

Laser-Based Pedestrian Tracking in Outdoor Environments by Multiple Mobile Robots

Laser-Based Pedestrian Tracking in Outdoor Environments by Multiple Mobile Robots Sensors 0,, 4489-4507; do:0.3390/s4489 Arcle OPEN ACCESS sensors ISSN 44-80 www.mdp.com/ournal/sensors Laser-Based Pedesran rackng n Oudoor Envronmens by Mulple Moble Robos Masaaka Ozak, Ke Kakmuma, Masafum

More information

Lecture 11 SVM cont

Lecture 11 SVM cont Lecure SVM con. 0 008 Wha we have done so far We have esalshed ha we wan o fnd a lnear decson oundary whose margn s he larges We know how o measure he margn of a lnear decson oundary Tha s: he mnmum geomerc

More information

GENERATING CERTAIN QUINTIC IRREDUCIBLE POLYNOMIALS OVER FINITE FIELDS. Youngwoo Ahn and Kitae Kim

GENERATING CERTAIN QUINTIC IRREDUCIBLE POLYNOMIALS OVER FINITE FIELDS. Youngwoo Ahn and Kitae Kim Korean J. Mah. 19 (2011), No. 3, pp. 263 272 GENERATING CERTAIN QUINTIC IRREDUCIBLE POLYNOMIALS OVER FINITE FIELDS Youngwoo Ahn and Kae Km Absrac. In he paper [1], an explc correspondence beween ceran

More information

Density Matrix Description of NMR BCMB/CHEM 8190

Density Matrix Description of NMR BCMB/CHEM 8190 Densy Marx Descrpon of NMR BCMBCHEM 89 Operaors n Marx Noaon Alernae approach o second order specra: ask abou x magnezaon nsead of energes and ranson probables. If we say wh one bass se, properes vary

More information

Lecture 18: The Laplace Transform (See Sections and 14.7 in Boas)

Lecture 18: The Laplace Transform (See Sections and 14.7 in Boas) Lecure 8: The Lalace Transform (See Secons 88- and 47 n Boas) Recall ha our bg-cure goal s he analyss of he dfferenal equaon, ax bx cx F, where we emloy varous exansons for he drvng funcon F deendng on

More information

A Novel Efficient Stopping Criterion for BICM-ID System

A Novel Efficient Stopping Criterion for BICM-ID System A Novel Effcen Soppng Creron for BICM-ID Sysem Xao Yng, L Janpng Communcaon Unversy of Chna Absrac Ths paper devses a novel effcen soppng creron for b-nerleaved coded modulaon wh erave decodng (BICM-ID)

More information

A Deterministic Algorithm for Summarizing Asynchronous Streams over a Sliding Window

A Deterministic Algorithm for Summarizing Asynchronous Streams over a Sliding Window A Deermnsc Algorhm for Summarzng Asynchronous Sreams over a Sldng ndow Cosas Busch Rensselaer Polyechnc Insue Srkana Trhapura Iowa Sae Unversy Oulne of Talk Inroducon Algorhm Analyss Tme C Daa sream: 3

More information

Comb Filters. Comb Filters

Comb Filters. Comb Filters The smple flers dscussed so far are characered eher by a sngle passband and/or a sngle sopband There are applcaons where flers wh mulple passbands and sopbands are requred Thecomb fler s an example of

More information

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC CH.3. COMPATIBILITY EQUATIONS Connuum Mechancs Course (MMC) - ETSECCPB - UPC Overvew Compably Condons Compably Equaons of a Poenal Vecor Feld Compably Condons for Infnesmal Srans Inegraon of he Infnesmal

More information

An Effective TCM-KNN Scheme for High-Speed Network Anomaly Detection

An Effective TCM-KNN Scheme for High-Speed Network Anomaly Detection Vol. 24, November,, 200 An Effecve TCM-KNN Scheme for Hgh-Speed Nework Anomaly eecon Yang L Chnese Academy of Scences, Bejng Chna, 00080 lyang@sofware.c.ac.cn Absrac. Nework anomaly deecon has been a ho

More information

Modélisation de la détérioration basée sur les données de surveillance conditionnelle et estimation de la durée de vie résiduelle

Modélisation de la détérioration basée sur les données de surveillance conditionnelle et estimation de la durée de vie résiduelle Modélsaon de la dééroraon basée sur les données de survellance condonnelle e esmaon de la durée de ve résduelle T. T. Le, C. Bérenguer, F. Chaelan Unv. Grenoble Alpes, GIPSA-lab, F-38000 Grenoble, France

More information

The Analysis of the Thickness-predictive Model Based on the SVM Xiu-ming Zhao1,a,Yan Wang2,band Zhimin Bi3,c

The Analysis of the Thickness-predictive Model Based on the SVM Xiu-ming Zhao1,a,Yan Wang2,band Zhimin Bi3,c h Naonal Conference on Elecrcal, Elecroncs and Compuer Engneerng (NCEECE The Analyss of he Thcknesspredcve Model Based on he SVM Xumng Zhao,a,Yan Wang,band Zhmn B,c School of Conrol Scence and Engneerng,

More information

Object Tracking Based on Visual Attention Model and Particle Filter

Object Tracking Based on Visual Attention Model and Particle Filter Inernaonal Journal of Informaon Technology Vol. No. 9 25 Objec Trackng Based on Vsual Aenon Model and Parcle Fler Long-Fe Zhang, Yuan-Da Cao 2, Mng-Je Zhang 3, Y-Zhuo Wang 4 School of Compuer Scence and

More information

( t) Outline of program: BGC1: Survival and event history analysis Oslo, March-May Recapitulation. The additive regression model

( t) Outline of program: BGC1: Survival and event history analysis Oslo, March-May Recapitulation. The additive regression model BGC1: Survval and even hsory analyss Oslo, March-May 212 Monday May 7h and Tuesday May 8h The addve regresson model Ørnulf Borgan Deparmen of Mahemacs Unversy of Oslo Oulne of program: Recapulaon Counng

More information

PARTICLE FILTER BASED VEHICLE TRACKING APPROACH WITH IMPROVED RESAMPLING STAGE

PARTICLE FILTER BASED VEHICLE TRACKING APPROACH WITH IMPROVED RESAMPLING STAGE ISS: 0976-910(OLIE) ICTACT JOURAL O IMAGE AD VIDEO PROCESSIG, FEBRUARY 014, VOLUME: 04, ISSUE: 03 PARTICLE FILTER BASED VEHICLE TRACKIG APPROACH WITH IMPROVED RESAMPLIG STAGE We Leong Khong 1, We Yeang

More information

Attribute Reduction Algorithm Based on Discernibility Matrix with Algebraic Method GAO Jing1,a, Ma Hui1, Han Zhidong2,b

Attribute Reduction Algorithm Based on Discernibility Matrix with Algebraic Method GAO Jing1,a, Ma Hui1, Han Zhidong2,b Inernaonal Indusral Informacs and Compuer Engneerng Conference (IIICEC 05) Arbue educon Algorhm Based on Dscernbly Marx wh Algebrac Mehod GAO Jng,a, Ma Hu, Han Zhdong,b Informaon School, Capal Unversy

More information

Chalmers Publication Library

Chalmers Publication Library Chalmers Publcaon Lbrary Exended Objec Tracng usng a Radar Resoluon Model Ths documen has been downloaded from Chalmers Publcaon Lbrary CPL. I s he auhor s verson of a wor ha was aeped for publcaon n:

More information

Application of thermal error in machine tools based on Dynamic. Bayesian Network

Application of thermal error in machine tools based on Dynamic. Bayesian Network Inernaonal Journal of Research n Engneerng and Scence (IJRES) ISS (Onlne): 2320-9364, ISS (Prn): 2320-9356 www.res.org Volume 3 Issue 6 ǁ June 2015 ǁ PP.22-27 pplcaon of hermal error n machne ools based

More information

Density Matrix Description of NMR BCMB/CHEM 8190

Density Matrix Description of NMR BCMB/CHEM 8190 Densy Marx Descrpon of NMR BCMBCHEM 89 Operaors n Marx Noaon If we say wh one bass se, properes vary only because of changes n he coeffcens weghng each bass se funcon x = h< Ix > - hs s how we calculae

More information

Hidden Markov Models

Hidden Markov Models 11-755 Machne Learnng for Sgnal Processng Hdden Markov Models Class 15. 12 Oc 2010 1 Admnsrva HW2 due Tuesday Is everyone on he projecs page? Where are your projec proposals? 2 Recap: Wha s an HMM Probablsc

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 0 Canoncal Transformaons (Chaper 9) Wha We Dd Las Tme Hamlon s Prncple n he Hamlonan formalsm Dervaon was smple δi δ Addonal end-pon consrans pq H( q, p, ) d 0 δ q ( ) δq ( ) δ

More information

Analysis And Evaluation of Econometric Time Series Models: Dynamic Transfer Function Approach

Analysis And Evaluation of Econometric Time Series Models: Dynamic Transfer Function Approach 1 Appeared n Proceedng of he 62 h Annual Sesson of he SLAAS (2006) pp 96. Analyss And Evaluaon of Economerc Tme Seres Models: Dynamc Transfer Funcon Approach T.M.J.A.COORAY Deparmen of Mahemacs Unversy

More information

Bandlimited channel. Intersymbol interference (ISI) This non-ideal communication channel is also called dispersive channel

Bandlimited channel. Intersymbol interference (ISI) This non-ideal communication channel is also called dispersive channel Inersymol nererence ISI ISI s a sgnal-dependen orm o nererence ha arses ecause o devaons n he requency response o a channel rom he deal channel. Example: Bandlmed channel Tme Doman Bandlmed channel Frequency

More information

Relative controllability of nonlinear systems with delays in control

Relative controllability of nonlinear systems with delays in control Relave conrollably o nonlnear sysems wh delays n conrol Jerzy Klamka Insue o Conrol Engneerng, Slesan Techncal Unversy, 44- Glwce, Poland. phone/ax : 48 32 37227, {jklamka}@a.polsl.glwce.pl Keywor: Conrollably.

More information

Volatility Interpolation

Volatility Interpolation Volaly Inerpolaon Prelmnary Verson March 00 Jesper Andreasen and Bran Huge Danse Mares, Copenhagen wan.daddy@danseban.com brno@danseban.com Elecronc copy avalable a: hp://ssrn.com/absrac=69497 Inro Local

More information

Gray-dynamic EKF for Mobile Robot SLAM in Indoor Environment

Gray-dynamic EKF for Mobile Robot SLAM in Indoor Environment Gray-dynamc EKF for Moble obo SLAM n Indoor Envronmen Peng Wang, Qbn Zhang, Zongha hen Deparmen of Auomaon, Unversy of Scence and echnology of hna, Hefe, 6, hna grapesonwang@gmalcom, zqb@malusceducn, chenzh@usceducn

More information

Sampling Procedure of the Sum of two Binary Markov Process Realizations

Sampling Procedure of the Sum of two Binary Markov Process Realizations Samplng Procedure of he Sum of wo Bnary Markov Process Realzaons YURY GORITSKIY Dep. of Mahemacal Modelng of Moscow Power Insue (Techncal Unversy), Moscow, RUSSIA, E-mal: gorsky@yandex.ru VLADIMIR KAZAKOV

More information

SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β

SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β SARAJEVO JOURNAL OF MATHEMATICS Vol.3 (15) (2007), 137 143 SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β M. A. K. BAIG AND RAYEES AHMAD DAR Absrac. In hs paper, we propose

More information

A MACHINE LEARNING APPROACH FOR HUMAN POSTURE DETECTION IN DOMOTICS APPLICATIONS

A MACHINE LEARNING APPROACH FOR HUMAN POSTURE DETECTION IN DOMOTICS APPLICATIONS A MACHINE LEARNING APPROACH FOR HUMAN POSTURE DETECTION IN DOMOTICS APPLICATIONS L.Pann*, R.Cucchara*, *D.S.I. Unversy of Modena, va Vgnolese 905-41100 Modena, Ialy emal: luca.pann@lbero. ; ra.cucchara@unmo.

More information

Effect of Resampling Steepness on Particle Filtering Performance in Visual Tracking

Effect of Resampling Steepness on Particle Filtering Performance in Visual Tracking 102 The Inernaonal Arab Journal of Informaon Technology, Vol. 10, No. 1, January 2013 Effec of Resamplng Seepness on Parcle Flerng Performance n Vsual Trackng Zahdul Islam, Ch-Mn Oh, and Chl-Woo Lee School

More information

Advanced Machine Learning & Perception

Advanced Machine Learning & Perception Advanced Machne Learnng & Percepon Insrucor: Tony Jebara SVM Feaure & Kernel Selecon SVM Eensons Feaure Selecon (Flerng and Wrappng) SVM Feaure Selecon SVM Kernel Selecon SVM Eensons Classfcaon Feaure/Kernel

More information

Comparison of Differences between Power Means 1

Comparison of Differences between Power Means 1 In. Journal of Mah. Analyss, Vol. 7, 203, no., 5-55 Comparson of Dfferences beween Power Means Chang-An Tan, Guanghua Sh and Fe Zuo College of Mahemacs and Informaon Scence Henan Normal Unversy, 453007,

More information

Dual Approximate Dynamic Programming for Large Scale Hydro Valleys

Dual Approximate Dynamic Programming for Large Scale Hydro Valleys Dual Approxmae Dynamc Programmng for Large Scale Hydro Valleys Perre Carpener and Jean-Phlppe Chanceler 1 ENSTA ParsTech and ENPC ParsTech CMM Workshop, January 2016 1 Jon work wh J.-C. Alas, suppored

More information

The Dynamic Programming Models for Inventory Control System with Time-varying Demand

The Dynamic Programming Models for Inventory Control System with Time-varying Demand The Dynamc Programmng Models for Invenory Conrol Sysem wh Tme-varyng Demand Truong Hong Trnh (Correspondng auhor) The Unversy of Danang, Unversy of Economcs, Venam Tel: 84-236-352-5459 E-mal: rnh.h@due.edu.vn

More information

CELLULAR AUTOMATA BASED PATH-PLANNING ALGORITHM FOR AUTONOMOUS MOBILE ROBOTS. Rami Al-Hmouz, Tauseef Gulrez & Adel Al-Jumaily

CELLULAR AUTOMATA BASED PATH-PLANNING ALGORITHM FOR AUTONOMOUS MOBILE ROBOTS. Rami Al-Hmouz, Tauseef Gulrez & Adel Al-Jumaily CELLULAR AUTOMATA BASED PATH-PLANNING ALGORITHM FOR AUTONOMOUS MOBILE ROBOTS Ram Al-Hmouz, Tauseef Gulrez & Adel Al-Jumaly Informaon and Communcaons Group ARC Cenre of Ecellence n Auonomous Sysems Unversy

More information

CS 268: Packet Scheduling

CS 268: Packet Scheduling Pace Schedulng Decde when and wha pace o send on oupu ln - Usually mplemened a oupu nerface CS 68: Pace Schedulng flow Ion Soca March 9, 004 Classfer flow flow n Buffer managemen Scheduler soca@cs.bereley.edu

More information

/99 $10.00 (c) 1999 IEEE

/99 $10.00 (c) 1999 IEEE Recognzng Hand Gesure Usng Moon Trajecores Mng-Hsuan Yang and Narendra Ahuja Deparmen of Compuer Scence and Beckman Insue Unversy of Illnos a Urbana-Champagn, Urbana, IL 611 fmhyang,ahujag@vson.a.uuc.edu

More information

ABSTRACT KEYWORDS. Bonus-malus systems, frequency component, severity component. 1. INTRODUCTION

ABSTRACT KEYWORDS. Bonus-malus systems, frequency component, severity component. 1. INTRODUCTION EERAIED BU-MAU YTEM ITH A FREQUECY AD A EVERITY CMET A IDIVIDUA BAI I AUTMBIE IURACE* BY RAHIM MAHMUDVAD AD HEI HAAI ABTRACT Frangos and Vronos (2001) proposed an opmal bonus-malus sysems wh a frequency

More information

5th International Conference on Advanced Design and Manufacturing Engineering (ICADME 2015)

5th International Conference on Advanced Design and Manufacturing Engineering (ICADME 2015) 5h Inernaonal onference on Advanced Desgn and Manufacurng Engneerng (IADME 5 The Falure Rae Expermenal Sudy of Specal N Machne Tool hunshan He, a, *, La Pan,b and Bng Hu 3,c,,3 ollege of Mechancal and

More information

[ ] 2. [ ]3 + (Δx i + Δx i 1 ) / 2. Δx i-1 Δx i Δx i+1. TPG4160 Reservoir Simulation 2018 Lecture note 3. page 1 of 5

[ ] 2. [ ]3 + (Δx i + Δx i 1 ) / 2. Δx i-1 Δx i Δx i+1. TPG4160 Reservoir Simulation 2018 Lecture note 3. page 1 of 5 TPG460 Reservor Smulaon 08 page of 5 DISCRETIZATIO OF THE FOW EQUATIOS As we already have seen, fne dfference appromaons of he paral dervaves appearng n he flow equaons may be obaned from Taylor seres

More information

Dynamic Team Decision Theory

Dynamic Team Decision Theory Dynamc Team Decson Theory EECS 558 Proec Repor Shruvandana Sharma and Davd Shuman December, 005 I. Inroducon Whle he sochasc conrol problem feaures one decson maker acng over me, many complex conrolled

More information

Let s treat the problem of the response of a system to an applied external force. Again,

Let s treat the problem of the response of a system to an applied external force. Again, Page 33 QUANTUM LNEAR RESPONSE FUNCTON Le s rea he problem of he response of a sysem o an appled exernal force. Agan, H() H f () A H + V () Exernal agen acng on nernal varable Hamlonan for equlbrum sysem

More information

Efficient Asynchronous Channel Hopping Design for Cognitive Radio Networks

Efficient Asynchronous Channel Hopping Design for Cognitive Radio Networks Effcen Asynchronous Channel Hoppng Desgn for Cognve Rado Neworks Chh-Mn Chao, Chen-Yu Hsu, and Yun-ng Lng Absrac In a cognve rado nework (CRN), a necessary condon for nodes o communcae wh each oher s ha

More information

Extended MHT Algorithm for Multiple Object Tracking

Extended MHT Algorithm for Multiple Object Tracking Eended MHT Algorhm for Mulple Obec Trackng Long Yng, Changsheng Xu aonal Lab of Paern Recognon, Insue of Auomaon, Chnese Academy of Scences, Beng, Chna {lyng, csu}@nlpr.a.ac.cn Wen Guo, Elecronc Engneerng

More information