Fragments based tracking with adaptive multi-cue integration

Size: px
Start display at page:

Download "Fragments based tracking with adaptive multi-cue integration"

Transcription

1 Fragmets based tracig with adaptive multi-cue itegratio Abstract Lichua Gu *, Jiaxiao Liu, Chegi Wag School of Computer ad Iformatio, Ahui Agriculture Uiversity, Hefei Ahui, No.130 West Chagiag Road Chia, Received 1 Jauary 2014, I this paper, we address the issue of part-based tracig by proposig a ew fragmets-based tracer with multi-cue itegratio. First the target template is divided ito multiple fragmets to get the cotributio of each fragmet to the possible positios of the target i the curret frame, similarity measure is used i edge histogram ad HSV histogram i every fragmet, ad the weights of cues itegratio are computed adaptively. Secod we preset a fragmet template update mechaism with the discrimiatio betwee occlusios ad appearace chages. The template is uchaged whe the target is occluded ad some fragmets of template are updated i the case of appearace chages. I the experimets we use the idoor ad outdoor test videos which cotai the illumiatio chages, occlusios, ad the appearace chages of targets. The experimetal results show that our approach has strog robustess ad high tracig accuracy. Keywords: Target tracig, Multi-cue itegratio, Fragmets tracig, Template update 1 Itroductio The obect tracig i a accurate way is essetial for applicatios lie activity aalysis, ma-machie iteractio ad visual surveillace. However, i the actual process of tracig, target appearace chages (as show i Figure 1 (a)) ad occlusios (as show i Figure 1 (b)) will affect the stability of the tracig algorithm, which leads to iaccurate target tracig ad eve target lost. As a result, how to effectively deal with target appearace chages ad occlusios has always bee oe of the difficulties i target tracig problems. A commo solutio to the above problems is tracig target by the itegratio of multiple visual cues, amely each cue provides a lielihood value for possible positios of the target i the ext frame, ad these cues are itegrated accordig to the lielihood values to locate the fial output (Figure 2). I recet te years, a few of tracig algorithms usig multi-cue itegratio have bee proposed, say, Birchfield proposed usig itesity gradiet ad color histograms to trac people head [1]. The disadvatage of this method is that durig tracig the cues do ot always provide reliable iformatio about the target obect, sice the equal weights are assiged to each cue ad the cues weights are fixed i the whole video sequece. I order to solve this problem, Triesch ad vo der Malsburg proposed a ew itegratio framewor [2], which ca tae advatage of the ucertaity of each cue to adaptively adust each cue cotributio to the tracig results. Followig this framewor appeared may adaptive cues itegratio algorithms, as show i [3-8]. These algorithms itegrated a variety of cues, complemeted each other ad improved the accuracy of the tracig process to some extet. (a) (b) FIGURE 1 Difficulties i tracig coditios: (a) target s variable appearace; (b) occlusios Aother solutio to the above problems is to trac movig targets based o fragmets. I the tracig of * Correspodig author - glc@ahau.edu.c 126

2 (a) (b) (c) FIGURE 2 Cues based o differet features: (a) Origial image; (b) HSV histogram; (c) Edge histogram huma, for example, huma is divided ito the head, torso ad limbs [9-11]. This approach geerally required target model is ow or a priori premise ad is ot suitable for geeric target tracig. I order to solve this problem, some scholars put forward the geeral fragmet tracig algorithms [12, 13]. This id of method divided the target ito differet parts, but the divisio was arbitrary without cosiderig ay referece target model. I the process of tracig, this method weighted each fragmet, ad the cotributios of differet fragmets are combied through statistical method to get the fial output locatio of target. If there were target appearace chages or occlusios, the weight of correspodig fragmet would be small ad the impact o the overall goal would be small, thus the resolutio of target would be improved. The target tracig method based o template matchig has attracted icreasig attetio [14-17]. Traditioal image template matchig based tracig method is still widely used because of its simple computatios. This id of method extracted some cues as a template which will be ept uchaged, ad the looed for a regio whose cues were most similar to the template i curret frame. However, the target may be occluded durig motio, ad may also chage its appearace due to ow motio. I these cases the template eeds to be updated olie to trac the target accurately. The existig mai template update algorithms update template oce the chages of grey are detected i target regios. These algorithms are differet from oe aother oly i that the update rate or updated compoets are ot the same, without discrimiatio of appearace chages ad occlusios. The compariso of aforemetioed algorithms is give i Table 1. This paper's cotributios lie i: 1) our algorithm employs a adaptive cue itegratio scheme i each fragmet of the target. Similarity measure i edge histogram ad HSV histogram are used, the vote of each fragmet cotributes to the oit tracig result accordig to adaptive itegratio, ad the oes havig small values have little effect o the outcome. This allows us to achieve a better tracig accuracy i hadlig partial occlusios ad appearace chages. 2) Our algorithm ca update the fragmet template olie durig the tracig process. I the small fragmets the method detects the possible fragmets where appearace chages or occlusios occur, ad the taes correspodig update strategies accordig to the matchig of the small fragmets. The fragmet template overcomes the shortcomig of the traditioal template, which modelled the whole target without the spatial iformatio. The rest of this paper is orgaized as follows. The subsequet sectio describes the fragmet tracig algorithm. The a adaptive multi-cue itegratio model is give i Sectio 3. Sectio 4 presets a ew tracig algorithm of multi-cue itegratio based o fragmet, which is the mai cotributio of this paper. Sectio 5 is devoted to algorithm aalysis, icludig experimetal results o videos with differet tracig coditios, ad gives qualitative ad quatitative aalysis. The fial sectio cocludes this paper ad suggests the future directio of our research. 2 Fragmet tracig algorithm Fragmet tracig is a id of target tracig algorithm based o cues proposed by Adam [12]. The basic idea is dividig target widow ito multiple sub-fragmets ad represets the target accordig to the oit sub-fragmet histograms. This id of algorithm has good atiiterferece ad ati-occlusio, ad the computatio has othig to do with traced target size. Detailed descriptio is as follows: The divisio patter is show i Figure 3. The target template is: ( p ( ), p p p1,, T h b, p 127

3 p is the idex of fragmet, the target is divided ito fragmets, h (b) ad p are a histogram ad weight p respectively, where b=1, B, ad fragmet weight equals 1 for every fragmet. the histogram matchig distace betwee fragmet q, ad fragmet : x y, which is defied as d( p ) :( x, ( h ( b) h ( b)) p 2 B p q:( x, y ) b1 hp ( b) h ( ) q b :(, x, y ). (2) FIGURE 3 Vertical ad horizotal fragmet template Fragmet tracig is to fid the most similar regio to the template i curret frame. The tracig priciple is show i Figure 4. Assume that a previous estimate of the positio has bee got, ad each poit i the eighbourhood of this estimate is searched, the every coordiate poit i the searchig widow is a cadidate for the target i curret frame. The smaller d( p :( x, ) meas more similarity betwee the histograms. After ( xy, ) has traversed all the cadidate poits, we obtai the similarity betwee each cadidate target ad the curret template. The the curret target locatio ( xy ˆ, ˆ) is defied as: ( xˆ, yˆ ) arg mi( S( x, ), (3) ( xy, ) where is the set of all the cadidate locatios. To distiguish easily betwee occlusios ad appearace chages, after fidig the target locatio i curret frame accordig to the matchig distace betwee fragmet q: xˆ, yˆ i the target regio ad the correspodig fragmet p, we update the weight of p as p ( d( p ) exp( ), (4) :( xˆ, yˆ ) 2 d where is the variace of d( p d :( xˆ, yˆ )). For coveiece q: xˆ, yˆ is represeted as q i the sequel. FIGURE 4 Fragmet tracig priciple Let p =(dx,dy,h,w) be a rectagular patch i the template, whose cetre is displaced (dx,d from the template cetre, ad whose half width ad height are w ad h respectively. Let (x, be a hypothesized positio of the obect i the curret frame. The the patch p defies a correspodig rectagular patch i the image q ; x, y whose cetre is at (x+dx,y+d ad whose half width ad height are w ad h, respectively. Calculatig the histogram matchig distaces of each q : x, y i the cadidate regio ad the correspodig fragmet i the template, we ca get Formula (1) usig the liear weightig of all the matchig distaces to represet the similarity betwee cadidates ad target template. S( x, d( p, (1) 1 : ( x, ) where S( x, y ) is the similarity betwee the cadidate at the locatio ( xy, ) ad the template, ad d ( p :( x, ) is p 3 Adaptive multi-cue itegratio I complicated scee, it is difficult to obtai good tracig performace by sigle visual cue. To further improve tracig robustess, we eed itegrate other cues. Suppose F is a visual cue which has bee selected ad the fused cue is f fusio, the f fusio ca be represeted as: f where fusio F is the weight of, (5) F. It is clear that differet cues have differet cotributios to the ability of idetifyig the whole target due to the impact of the eviromet, so differet cues should have differet weights. Sice there may be target s appearace chages or occlusios i real tracig, the weight of each cue i 128

4 the each frame of video sequece should be updated dyamically. There are two problems to be solved: oe is how to evaluate the weights accordig to differet cotributio of each cue i differet tracig eviromet, the other is how to update the weights i a steady ad real-time way. 4 Proposed tracig algorithm 4.1 TRACKING PROCESS This paper proposes a ew method of fragmets based tracig with multi-cue itegratio ad template updatig. The algorithm describes the target obect by a template, ad carries out tracig by searchig for the image regio i each frame of the tracig sequece with multiple cues similar to the template. This estimatio process is doe by splittig the target obect ito multiple arbitrary patches with each of them describig a differet part of the target ad accordigly providig the multi-cue iformatio. The template is adaptively updated i accordace with whether occlusios or appearace chages occur to alleviate the template drift. The tracig schedule is show i Figure 5. Step3: The idividual votes of the template patches are combied to obtai the oit tracig result. The weights of the template fragmets are computed accordig to the formula (4), if p 0.5, the histograms of q chaged a lot compared to correspodig p, that meas there might be occlusios or appearace chages. We call such q as ivalid fragmet. The oit similarity S( x, is defied i formula (6), ad the we ca get the optimum positio of the target i curret frame usig formula (3): d( p ), 0.5, :( x, p p S( x, (6) 1 0, 0.5. p Step4: We udge each ivalid fragmet to see if it is caused by occlusio or appearace chage. If there is occlusio, the go to step 2 directly without template update ad start the ext frame tracig; otherwise update the template ad the go to step 2. The detail is described i sectio CUES HISTOGRAM COMPUTATION FIGURE 5 The flow chart of our algorithm The steps of our algorithm are as follows. Step1: Target template is divided ito multiple fragmets usig the patch layouts give i Figure 2, ad the each of these fragmets is represeted with histograms of differet cues (colour, edge), ad Itegral Histogram data structure [18] is used to estimate them i a fast way. Step2: The search widow size is set to be (2M 1) (2M 1) i curret frame, amely the search radius is set to be M pixels from the previous target positio (M =7 i this paper, ad the search widow is ). Matchig each template patch ad the correspodig image patch is the carried out by comparig their histograms usig the formula (13), ad the we obtai the values { d( p ) } by adaptive :( x, y ) weight itegratio. Each d( p :( x, ) idicates every fragmet votes o the possible positios of the target. The detail is give i sectio 4.3. I complicated scee, it is difficult to obtai good tracig performace by sigle visual cue. To improve tracig robustess, we eed itegrate other cues. It is very importat to select the strog discrimiate cues. This paper uses the itegratio of colour ad edge cues to costruct lielihood fuctio for the purpose of scatterig the impacts of all ids of chages. Colour is a maor cue to describe the target. There have bee may studies about the descriptio of target colour cues before. Ref. [19] proposed a colour histogram to describe the target colour cues, which has simple calculatio ad fast processig, ad is relatively robust i solvig problems such as partial occlusio, rotatio, etc. HSV histogram is employed to extract colour cue i each fragmet i this paper. I our experimets, the Hue, Saturatio ad Value are quatified to be 16, 4, ad 4. The three colour compoets are combied ito oe dimesioal cue accordig to the quatitative level: C 16H 4S V, (7) where C [0,1,,255], so we ca get 256 bis HSV histogram. Edge is aother commoly used cue to describe the target. This paper chooses gradiet directio histogram to describe the edge cue of the target [19]. The calculatio formulas are as follows: Horizotal gradiet G ( x, f ( x 1, f ( x 1,, (8) x 129

5 Vertical gradiet q:( x, y ) d( p ) 1 :( x, y ) d( p ) :( x, y ), (14) G ( x, f ( x, y 1) f ( x, y 1), (9) y Gradiet magitude m x y G x y G x y, (10) 2 2 (, ) x(, ) y(, ) Gradiet directio ( x, arcta( G ( x, / G ( x, ), (11) y where ( xy, ) [ / 2, / 2]. (x, should be mapped ito 0 ~ 2. ( x, Gx 0 ( x, 2 ( x, Gx 0 ad Gy 0 ( x, Gx 0 ad Gy 0 x. (12) [0,2 ) is divided ito bis with the width of each bi beig 2 / (=9 i the paper). Each bi value is got by coutig the pixels umber i it ad we ca obtai the edge histogram. 4.3 ADAPTIVE WEIGHT INTEGRATION I order to adapt to the eviromet ad the target appearace chages, the weight of each feature i each frame must be updated i real time. Ref. [21] proposed Kolmogorov-Smirov similarity measure criteria, K-S distace for short, to calculate the degree of similarity betwee the cadidate obect ad target template. Its value represets the differece of two histograms; the smaller K-S value meas smaller differece ad higher similarity. Based o K-S distace, the matchig distace for oe cue is defied as follows: d( p ) :( x, y ) B p q :( x, y ) b1 :( x, y ) ( h ( b) h ( b)) h ( b) h ( b) p q:( x, y ) 2. (13) d( p ) is the histogram matchig distace for cue betwee fragmet q, p ad fragmet x y :. We ca chage the weight of correspodig cue by formula (14), so as to automatically chage the ifluece of differet cues to global fuctios. A adaptive weight itegratio formula is defied as i (15): :( x, q :( x, y ) :( x, y ) 1 d( p ) d( p ). (15) It ca be see that automatically chagig the differet cues weights ca better reflect the cotributios of cues to the result. I experimets =1, 2, colour is the 1 1 first cue, edge is the secod, d( p ) ad 2 2 (, ) :( x, y ) :( x, y ) d p q are the histograms matchig distaces for colour ad edge cues betwee fragmet p ad fragmet q : x, y respectively. Whe the obect begis to eter the occlusio or chages appearace, the edge will chage 2 2 greatly, the weight of d( p ) is larger, the weight of 1 1 (, ) :( x, y ) :( x, y ) d p q is smaller, ad the the cotributio of edge cue to d( p :( x, ) is larger. Therefore, this adaptive itegratio method ca better reflect actual situatio durig tracig. 4.4 TEMPLATE ONLINE UPDATE I order to esure real-time tracig effect, this paper tracs the target by dividig the target ito fragmets, udges states of fragmets, ad adusts update strategy accordig to whether there is occlusio or appearace chage. The distace betwee the fragmet ad the template will be icreased because of occlusio or appearace chage. The template does ot have to be updated i the case of occlusio while the template eeds to be updated i the case of appearace chage. Fragmet weights of the template are give i Sectio 2. After the experimets we foud that if p 0.5, the q is a ivalid fragmet. Through the aalysis we foud that if the ivalid fragmet is caused by occlusio, the H compoet i HSV space of the fragmet will be distributed with greater probability i the previous frame i the bacgroud regio; if it is caused by target's ow chages, the H compoet i HSV space of the fragmet will be distributed with greater probability i the previous frame i the target regio. Because obstacles oly appear aroud the target, we select a "rig" bacgroud regio aroud the target to determie whether the target is occluded or ot, where the area of the aular regio is about 2 times of target area. The distributio of the H compoet i HSV space of the ivalid fragmet is bac proected oto the target ad aular bacgroud i the previous frame, so we ca obtai the distributio regio i the previous frame, ad the calculate respectively the probabilities of the ivalid fragmet belogig to the target regio ad the bacgroud regio. This paper uses the mai H compoet i HSV space of the ivalid 130

6 fragmet for bac proectio, ad the proectio method is the same as the oe used i Camshift [2]. As show i Figure 6, bac proectio image is a biary image, where white pixels deote the pixels whose grey values are equal to those of the mai H compoet of ivalid fragmets. I bac proectio image, we cout the umber of the foregroud pixels i the target area S ad the umber of 0 the foregroud pixels i the whole image S. If the t probability p of the H compoet of the ivalid 0 fragmet belogig to the target satisfies formula (16), we thi the target appearace chages. th is set to 0.8 after may experimets. If the ivalid fragmet is determied as occlusio, the template will ot be updated; if the ivalid fragmet is determied as appearace chage, the iformatio of the curret frame will be updated to the template. So that we ca rapidly get the target chage iformatio ad suppress bacgroud disturbace to the template at the same time, which obviously improves the tracig performace: S0 p0 log S S t 0 th. (16) The err( x, y ) is the locatio error betwee ob ob the target locatio ( x, y ) ad the true locatio ( x, y ). true true ob ob 5.1 THE QUALITATIVE ANALYSIS Test Sequece 1. Figure 7 shows the simulatio results of a video i a campus amed campus. The video sequece cotais pixel colour images. I the video, from frame 272 to frame 377, the target eters ito dar regio from bright regio ad later returs to the bright regio. The illumiatio has a great chage. At frame 347, the tracig widow of the fragmet-based tracer loses the target; the colour-texture based meashift tracer does ot lose the target, but the estimatio locatio of the target has already deviated from the actual locatio. At frame 387, the tracig widow of the colour-texture based mea-shift tracer loses the target too while the tracig widows of our method ad the decetralized template tracer ever lose the target. FIGURE 6 Bac proectio images uder appearace chage or the occlusio 5 Experimets ad aalysis We use the followig four groups of experimets to verify the effectiveess of the algorithm. The experimetal eviromets are Microsoft visual studio 2008 ad Opecv2.0. The results are aalysed i Matlab Qualitative ad quatitative aalysis are made for the results of the four methods, the fragmet-based tracer [12], the colour-texture based mea-shift tracer [7], the decetralized template tracer [17], ad the method proposed i this paper (our method). I the figures, the widow of our method is labelled as 1, the tracig widow of the decetralized template tracer is labelled as 2, the tracig widow of the colour-texture based mea-shift tracer is labelled as 3, ad the tracig widow of the fragmet-based tracer is labelled as 4. The qualitative aalysis is doe through observig test images ad the quatitative aalysis is doe through measurig locatio errors. Let 2 2 ( xob xtrue ) ( yob ytrue ) err( xob, yob ). (17) 2 FIGURE 7 Tracig results of video Campus (Frames: 246, 272, 302, 347, 377, 387) Test Sequece 2. Figure 8 shows the simulatio results of video hall, which comes from CAVIAR database. The video sequece cotais pixel colour images. The mai difficulty with this sequece is that the target is occluded by aother people. Whe the target is occluded at the video frame 191, both the fragmet-based tracer ad the decetralized template 131

7 tracer lose the target from frame 208 to frame 221, ad later o the colour-texture based mea-shift tracer loses the target too, while our method has ept tracig the target. To sum up, our method ca always trac the target o matter if there is occlusio or ot. The video sequece cotais pixel colour images, where the ma s head i the sequece has irregular traslatio ad rotatio movemets. At frame 100 ad frame 110 whe the face is approximately frot face, the four methods trac the target accurately. Whe the bac face appears at frame 130, the fragmet-based tracer ad the colour-texture based mea-shift tracer lose the target immediately, ad the fragmet-based tracer caot recover from the error eve for a side face at frame 131. Our method ad the decetralized template tracer have more robustess. Test Sequece 4. Figure 10 shows the simulatio results of video ma2 which comes from SPEVI database. FIGURE 8 Tracig results of video hall (Frames: 163, 191, 208, 221, 240, 267) Test Sequece 3. Figure 9 shows the simulatio results of video ma1 which comes from SPEVI database. FIGURE 9 Tracig results of video ma1 (Frames: 100, 110, 117, 130, 141, 145) FIGURE 10 Tracig results of video ma2 (Frames: 368, 384, 398, 410, 474, 580, 590, 601) The video sequece cotais pixel colour images. The target first rotates ad the the illumiatio chages sharply, followed by the part occlusio i the sequece. At frame 398 whe the target rotates, both the colour ad gradiet of the target chage, ad the fragmet-based tracer with oe sigle cue tracig loses the target whe the target turs bac, the colour-texture based mea-shift tracer with fixed template tracig loses the target too, ad the decetralized template tracer brigs about a certai tracig drift while our method achieves accurate 132

8 tracig. At frame 474, the lamp is tured o, which maes the illumiatio chage sharply ad the fragmetbased tracer lose the target agai, while the other three algorithms eep tracig the target. At frame 580 whe the target is occluded by aother movig obect, the other three algorithms fail to trac the target accurately while our method still tracs well, sice our method ot oly uses adaptive cue itegratio scheme durig fragmet tracig, but also employs the olie updatig template by distiguishig appearace chages ad occlusios. 5.2 THE QUANTITATIVE ANALYSIS Figures 7-10 show the qualitative aalysis results. We also provide the error plots for each sequece i Figure 11. FIGURE 11 Error plots for the video sequeces used i the quatitative aalysis It ca be see from these results that the proposed algorithm outperforms the fragmet-based tracer i terms of tracig accuracy. The reaso for this maily stems from our adaptive cue itegratio scheme, which removes the assumptio o the degree of potetial occlusios for the fragmet-based tracer, ad replaces its competitive approach to cue itegratio with a more cooperative strategy. The colour-texture based mea-shift tracer, which uses the oit colour ad LBP cues to trac target, outperforms the sigle cue tracig approaches ad is robust to the complex eviromet. However, it yields, i may cases, poor tracig results whe the target becomes occluded by the other obects i the scee. Our method ad the fragmet-based tracer cope with these ids of occlusio better tha most of the multi-cue itegratio tracers. The reaso stems from tracig the target i fragmets. The decetralized template tracer, which uses olie updatig template to trac target, is proved to be very robust agaist severe appearace chages sice it ca lear a ew obect model at each frame. Our method gives better results tha the decetralized template tracer for the video sequeces because of the fact that the decetralized template tracer updates target template at each frame without discrimiatio of appearace chages ad occlusios, ad caot prevet the template from driftig, eve if it uses short-term ad log-term appearace models As a result, for most of the video sequeces, our algorithm provides the best results. 6 Coclusios Video target tracig algorithm is faced with may problems, such as illumiatio chage, occlusios, appearace chages, etc. A ew tracig algorithm of multi-cue itegratio based o fragmet has bee proposed i this paper. The ew algorithm associates the image fragmets describig the differet parts of the target obect with multi-cue itegratio, where the weight of each cue is dyamically adusted durig tracig. I this way, the vote of each fragmet cotributes to the oit tracig result accordig to its adaptive weight itegratio. Furthermore, the algorithm uses the template updatig for target tracig, desigs the correspodig 133

9 template update algorithm, ad successfully implemets the target tracig uder complex bacgroud. We have demostrated the potetial ad the effectiveess of the proposed approach o various challegig video sequeces with differet tracig scearios. As revealed by our experimets, the proposed approach wors geerally better tha the fragmet-based tracer, the multi-cue itegratio tracers ad template-based tracers for the sequeces i the complex bacgroud of illumiatio, occlusio, ad the appearace chage. Acowledgemets The authors wish to tha the helpful commets ad suggestios from my teachers ad colleagues. This wor was supported by the Natioal Natural Sciece Foudatio of chia(grat No ),Natioal Sciece ad Techology Support Program(Grat No. 2013BAJ10B12), he Key Techologies R & D Program of AHui Provice (Grat No ) ad by the Natural Sciece Foudatio of AHui Provice(Grat No MF89). Refereces [1] Birchfield S 1998 Computer Visio ad Patter Recogitio (CVPR) Sata Barbara, CA:IEEE Computer Society [2] Triesch J, vo der Malsburg C 2001 Neural Computatio 13(9) [3] Yuru Wag, Xiaglog Tag, Qig Cui 2012 Patter Recogitio 45(12) [4] Brasett P, Mihaylova L, Bull D, Caagaraah N 2007 Image Visio Computatio 25(8) [5] Wag J Q, Yagi Y 2008 IEEE Trasactio o Image Processig 17(2) [6] Jia G, Tia Y, Wag Y 2010 Dyamic multi-cue tracig with detectio resposes associatio Proceedigs of the ACM Multimedia NY USA: ACM New Yor [7] Nig J, Zhag L, Zhag D, Wu C 2009 Iteratioal Joural of Patter Recogitio ad Artificial Itelligece 23(7), [8] Li Yig, Liag Shi, Bai Bedu, Feg David 2012 Multimedia Tools ad Applicatios 6(3) 1-21 [9] Nehum S M S, Ho J, Yag M H 2008 Visual tracig with histograms ad articulatig blocs Proc. Cof. Computer Visio ad Patter Recogitio (CVPR), Achorage, AK :CVPR [10] Nicel K, Stiefelhage R 2008 Dyamic itegratio of geeralized cues for perso tracig Proc. Eur. Cof. Computer Visio (Frace) (ECCV): Marseille [11] Chocaligam P, Pradeep N, Birchfield S 2009 Adaptive fragmets-based tracig of o-rigid obects usig level sets Proc. It. Cof. Computer Visio (Japa) (ICCV) :Kyoto [12] Adam A, Rivli E, Shimshoi I 2006 Robust fragmets-based tracig usig the itegral histogram Proc. Cof. Computer Visio ad Patter Recogitio (CVPR) [13] Erdem E, Dubuisso S 2012 Computer Visio ad Image Uderstadig 116(7) [14] Porili F, Tuzel O, Meer P 2006 Covariace tracig usig model update based o Lie algebra. Proceedigs of IEEE Computer Society Coferece o Computer Visio ad Patter Recogitio:New Yor, USA [15] Collis R T, Liu Y X, Leordeau M 2005 IEEE Trasactios o Patter Aalysis ad Machie Itelligece 27(10) [16] Wag Yog-Zhog, Liag Ya, Zhao Chu-Hui, Pa Qua 2008 Acta Automatica Siica 34 (4) [17] Firouzi Hadi, Naara Homayou 2012 Patter Recogitio 45(12) [18] Porili F 2005 Itegral histogram: A fast way to extract histograms i Cartesia spaces Proc. IEEE Cof. o Computer Visio ad Patter Recogitio (CVPR) [19] Comaiciu D, Ramesh V, Meer P 2003 IEEE Trasactios o Patter Aalysis ad Machie Itelligece 25(5) [20] Yag C, Duraiswami R, Davis L 2005 Fast multiple target tracig via a hierarchical particle filter Proceedigs of the Teth IEEE Iteratioal Coferece o Computer Visio: Beiig, Chia [21] Simard R, L Ecyuer P 2011 Joural of Statistical Software 39(11) 1-18 [22] Fashig M, Tomasi C 2005 IEEE Trasactios o Patter Aalysis ad Machie Itelligece 27(3) Authors Lichua Gu, October 1974, Sichua, Chia Curret positio, grades: Associate Professor of Computer Sciece Departmet of Ahui Agricultural Uiversity Uiversity studies: BSc, MSc degrees i computer sciece i HeFei Uiversity of techology, Chia, i 1997 ad Scietific iterest: research iterests, maily icludig machie learig, image processig, ad patter recogitio. Chegi Wag Curret positio, grades: master studies i agricultural iformatizatio i AHAU Uiversity Uiversity studies: BE i Logistics Egieerig, AHAU Uiversity (2013) Scietific iterests: research iterests iclude cogitive computig ad logistics system optimizatio Jiaxiao Liu Curret positio, grades: master studies i Agricultural Iformatics at Uiversity. Uiversity studies: BE i Logistics Egieerig (2013) from AHAU Uiversity. Scietific iterests: research directio icludes differet aspects of Agricultural logistics 134

3D Model Retrieval Method Based on Sample Prediction

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

More information

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

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

More information

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

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

More information

Stone Images Retrieval Based on Color Histogram

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

More information

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem

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

More information

Dynamic Programming and Curve Fitting Based Road Boundary Detection

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

More information

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

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

More information

New HSL Distance Based Colour Clustering Algorithm

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

More information

Pattern Recognition Systems Lab 1 Least Mean Squares

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

More information

Evaluation scheme for Tracking in AMI

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

More information

Performance Plus Software Parameter Definitions

Performance Plus Software Parameter Definitions Performace Plus+ Software Parameter Defiitios/ Performace Plus Software Parameter Defiitios Chapma Techical Note-TG-5 paramete.doc ev-0-03 Performace Plus+ Software Parameter Defiitios/2 Backgroud ad Defiitios

More information

Adaptive Resource Allocation for Electric Environmental Pollution through the Control Network

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

More information

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

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

More information

IMP: Superposer Integrated Morphometrics Package Superposition Tool

IMP: Superposer Integrated Morphometrics Package Superposition Tool IMP: Superposer Itegrated Morphometrics Package Superpositio Tool Programmig by: David Lieber ( 03) Caisius College 200 Mai St. Buffalo, NY 4208 Cocept by: H. David Sheets, Dept. of Physics, Caisius College

More information

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

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

More information

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

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

More information

Accuracy Improvement in Camera Calibration

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

More information

Image Segmentation EEE 508

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

More information

Improving Template Based Spike Detection

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

More information

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

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

More information

Neuro Fuzzy Model for Human Face Expression Recognition

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

More information

Auto-recognition Method for Pointer-type Meter Based on Binocular Vision

Auto-recognition Method for Pointer-type Meter Based on Binocular Vision JOURNAL OF COMPUTERS, VOL. 9, NO. 4, APRIL 204 787 Auto-recogitio Method for Poiter-type Meter Based o Biocular Visio Biao Yag School of Istrumet Sciece ad Egieerig, Southeast Uiversity, Najig 20096, Chia

More information

New Fuzzy Color Clustering Algorithm Based on hsl Similarity

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

More information

. Written in factored form it is easy to see that the roots are 2, 2, i,

. Written in factored form it is easy to see that the roots are 2, 2, i, CMPS A Itroductio to Programmig Programmig Assigmet 4 I this assigmet you will write a java program that determies the real roots of a polyomial that lie withi a specified rage. Recall that the roots (or

More information

Ones Assignment Method for Solving Traveling Salesman Problem

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

More information

Cubic Polynomial Curves with a Shape Parameter

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

More information

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

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

More information

ANN WHICH COVERS MLP AND RBF

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

More information

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

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

More information

THIN LAYER ORIENTED MAGNETOSTATIC CALCULATION MODULE FOR ELMER FEM, BASED ON THE METHOD OF THE MOMENTS. Roman Szewczyk

THIN LAYER ORIENTED MAGNETOSTATIC CALCULATION MODULE FOR ELMER FEM, BASED ON THE METHOD OF THE MOMENTS. Roman Szewczyk THIN LAYER ORIENTED MAGNETOSTATIC CALCULATION MODULE FOR ELMER FEM, BASED ON THE METHOD OF THE MOMENTS Roma Szewczyk Istitute of Metrology ad Biomedical Egieerig, Warsaw Uiversity of Techology E-mail:

More information

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

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

More information

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON

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

More information

A Note on Least-norm Solution of Global WireWarping

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

More information

The identification of key quality characteristics based on FAHP

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

More information

High-Speed Recognition Algorithm Based on BRISK and Saliency Detection for Aerial Images

High-Speed Recognition Algorithm Based on BRISK and Saliency Detection for Aerial Images Research Joural of Applied Scieces, Egieerig ad Techology 5(23): 5469-5473, 2013 ISSN: 2040-7459; e-issn: 2040-7467 Maxwell Scietific Orgaizatio, 2013 Submitted: November 29, 2012 Accepted: Jauary 17,

More information

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

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

More information

Analysis Metrics. Intro to Algorithm Analysis. Slides. 12. Alg Analysis. 12. Alg Analysis

Analysis Metrics. Intro to Algorithm Analysis. Slides. 12. Alg Analysis. 12. Alg Analysis Itro to Algorithm Aalysis Aalysis Metrics Slides. Table of Cotets. Aalysis Metrics 3. Exact Aalysis Rules 4. Simple Summatio 5. Summatio Formulas 6. Order of Magitude 7. Big-O otatio 8. Big-O Theorems

More information

Fundamentals of Media Processing. Shin'ichi Satoh Kazuya Kodama Hiroshi Mo Duy-Dinh Le

Fundamentals of Media Processing. Shin'ichi Satoh Kazuya Kodama Hiroshi Mo Duy-Dinh Le Fudametals of Media Processig Shi'ichi Satoh Kazuya Kodama Hiroshi Mo Duy-Dih Le Today's topics Noparametric Methods Parze Widow k-nearest Neighbor Estimatio Clusterig Techiques k-meas Agglomerative Hierarchical

More information

Improving Face Recognition Rate by Combining Eigenface Approach and Case-based Reasoning

Improving Face Recognition Rate by Combining Eigenface Approach and Case-based Reasoning Improvig Face Recogitio Rate by Combiig Eigeface Approach ad Case-based Reasoig Haris Supic, ember, IAENG Abstract There are may approaches to the face recogitio. This paper presets a approach that combies

More information

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

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

More information

Python Programming: An Introduction to Computer Science

Python Programming: An Introduction to Computer Science Pytho Programmig: A Itroductio to Computer Sciece Chapter 6 Defiig Fuctios Pytho Programmig, 2/e 1 Objectives To uderstad why programmers divide programs up ito sets of cooperatig fuctios. To be able to

More information

The Closest Line to a Data Set in the Plane. David Gurney Southeastern Louisiana University Hammond, Louisiana

The Closest Line to a Data Set in the Plane. David Gurney Southeastern Louisiana University Hammond, Louisiana The Closest Lie to a Data Set i the Plae David Gurey Southeaster Louisiaa Uiversity Hammod, Louisiaa ABSTRACT This paper looks at three differet measures of distace betwee a lie ad a data set i the plae:

More information

MATHEMATICAL METHODS OF ANALYSIS AND EXPERIMENTAL DATA PROCESSING (Or Methods of Curve Fitting)

MATHEMATICAL METHODS OF ANALYSIS AND EXPERIMENTAL DATA PROCESSING (Or Methods of Curve Fitting) MATHEMATICAL METHODS OF ANALYSIS AND EXPERIMENTAL DATA PROCESSING (Or Methods of Curve Fittig) I this chapter, we will eamie some methods of aalysis ad data processig; data obtaied as a result of a give

More information

A Study on the Performance of Cholesky-Factorization using MPI

A Study on the Performance of Cholesky-Factorization using MPI A Study o the Performace of Cholesky-Factorizatio usig MPI Ha S. Kim Scott B. Bade Departmet of Computer Sciece ad Egieerig Uiversity of Califoria Sa Diego {hskim, bade}@cs.ucsd.edu Abstract Cholesky-factorizatio

More information

Parabolic Path to a Best Best-Fit Line:

Parabolic Path to a Best Best-Fit Line: Studet Activity : Fidig the Least Squares Regressio Lie By Explorig the Relatioship betwee Slope ad Residuals Objective: How does oe determie a best best-fit lie for a set of data? Eyeballig it may be

More information

DETECTION OF LANDSLIDE BLOCK BOUNDARIES BY MEANS OF AN AFFINE COORDINATE TRANSFORMATION

DETECTION OF LANDSLIDE BLOCK BOUNDARIES BY MEANS OF AN AFFINE COORDINATE TRANSFORMATION Proceedigs, 11 th FIG Symposium o Deformatio Measuremets, Satorii, Greece, 2003. DETECTION OF LANDSLIDE BLOCK BOUNDARIES BY MEANS OF AN AFFINE COORDINATE TRANSFORMATION Michaela Haberler, Heribert Kahme

More information

Using a Dynamic Interval Type-2 Fuzzy Interpolation Method to Improve Modeless Robots Calibrations

Using a Dynamic Interval Type-2 Fuzzy Interpolation Method to Improve Modeless Robots Calibrations Joural of Cotrol Sciece ad Egieerig 3 (25) 9-7 doi:.7265/2328-223/25.3. D DAVID PUBLISHING Usig a Dyamic Iterval Type-2 Fuzzy Iterpolatio Method to Improve Modeless Robots Calibratios Yig Bai ad Dali Wag

More information

Low Complexity H.265/HEVC Coding Unit Size Decision for a Videoconferencing System

Low Complexity H.265/HEVC Coding Unit Size Decision for a Videoconferencing System BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 15, No 6 Special Issue o Logistics, Iformatics ad Service Sciece Sofia 2015 Prit ISSN: 1311-9702; Olie ISSN: 1314-4081 DOI:

More information

FEATURE BASED RECOGNITION OF TRAFFIC VIDEO STREAMS FOR ONLINE ROUTE TRACING

FEATURE BASED RECOGNITION OF TRAFFIC VIDEO STREAMS FOR ONLINE ROUTE TRACING FEATURE BASED RECOGNITION OF TRAFFIC VIDEO STREAMS FOR ONLINE ROUTE TRACING Christoph Busch, Ralf Dörer, Christia Freytag, Heike Ziegler Frauhofer Istitute for Computer Graphics, Computer Graphics Ceter

More information

Sorting in Linear Time. Data Structures and Algorithms Andrei Bulatov

Sorting in Linear Time. Data Structures and Algorithms Andrei Bulatov Sortig i Liear Time Data Structures ad Algorithms Adrei Bulatov Algorithms Sortig i Liear Time 7-2 Compariso Sorts The oly test that all the algorithms we have cosidered so far is compariso The oly iformatio

More information

Performance Comparisons of PSO based Clustering

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

More information

Ontology-based Decision Support System with Analytic Hierarchy Process for Tour Package Selection

Ontology-based Decision Support System with Analytic Hierarchy Process for Tour Package Selection 2017 Asia-Pacific Egieerig ad Techology Coferece (APETC 2017) ISBN: 978-1-60595-443-1 Otology-based Decisio Support System with Aalytic Hierarchy Process for Tour Pacage Selectio Tie-We Sug, Chia-Jug Lee,

More information

Python Programming: An Introduction to Computer Science

Python Programming: An Introduction to Computer Science Pytho Programmig: A Itroductio to Computer Sciece Chapter 1 Computers ad Programs 1 Objectives To uderstad the respective roles of hardware ad software i a computig system. To lear what computer scietists

More information

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

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

More information

Title: Robust Registration of Multimodal Remote Sensing Images Based on Structural Similarity.

Title: Robust Registration of Multimodal Remote Sensing Images Based on Structural Similarity. 7 IEEE. Persoal use of this material is permitted. Permissio from IEEE must be obtaied for all other uses, i ay curret or future media, icludig repritig/republishig this material for advertisig or promotioal

More information

Hui Xiao School of Environmental Science, Nanjing Xiaozhuang University, Nanjing , China

Hui Xiao School of Environmental Science, Nanjing Xiaozhuang University, Nanjing , China doi:0.3/00.39.. Cotiuous knn Queries i Dyamic Road Networks Hui Xiao School of Evirometal Sciece, Najig Xiaozhuag Uiversity, Najig 7, Chia Abstract Cotiuous knn queries have bee widely studied i recet

More information

Lecture 28: Data Link Layer

Lecture 28: Data Link Layer Automatic Repeat Request (ARQ) 2. Go ack N ARQ Although the Stop ad Wait ARQ is very simple, you ca easily show that it has very the low efficiecy. The low efficiecy comes from the fact that the trasmittig

More information

Image based Cats and Possums Identification for Intelligent Trapping Systems

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

More information

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

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

More information

*Corresponding author. Keywords: Power quality, Assessment system, Harmonic evaluation, Comprehensive evaluation.

*Corresponding author. Keywords: Power quality, Assessment system, Harmonic evaluation, Comprehensive evaluation. 7 Iteratioal Coferece o Eergy, Power ad Evirometal Egieerig (ICEPEE 7) ISBN: 978--6595-456- Study of the Power Quality Comprehesive Evaluatio Method Zhi-mi ZHAN, Peg-fei CHAI, Bi LUO, Xig-bo LIU, Yua-li

More information

1.2 Binomial Coefficients and Subsets

1.2 Binomial Coefficients and Subsets 1.2. BINOMIAL COEFFICIENTS AND SUBSETS 13 1.2 Biomial Coefficiets ad Subsets 1.2-1 The loop below is part of a program to determie the umber of triagles formed by poits i the plae. for i =1 to for j =

More information

The isoperimetric problem on the hypercube

The isoperimetric problem on the hypercube The isoperimetric problem o the hypercube Prepared by: Steve Butler November 2, 2005 1 The isoperimetric problem We will cosider the -dimesioal hypercube Q Recall that the hypercube Q is a graph whose

More information

Bezier curves. Figure 2 shows cubic Bezier curves for various control points. In a Bezier curve, only

Bezier curves. Figure 2 shows cubic Bezier curves for various control points. In a Bezier curve, only Edited: Yeh-Liag Hsu (998--; recommeded: Yeh-Liag Hsu (--9; last updated: Yeh-Liag Hsu (9--7. Note: This is the course material for ME55 Geometric modelig ad computer graphics, Yua Ze Uiversity. art of

More information

Human Activity Tracking for Wide-Area Surveillance

Human Activity Tracking for Wide-Area Surveillance Huma Activity Trackig for Wide-Area Surveillace Patrick D. O Malley Michael C. Nechyba A. Atoio Arroyo pomalley@mil.ufl.edu echyba@mil.ufl.edu arroyo@mil.ufl.edu Machie Itelligece Laboratory Departmet

More information

EFFECT OF QUERY FORMATION ON WEB SEARCH ENGINE RESULTS

EFFECT OF QUERY FORMATION ON WEB SEARCH ENGINE RESULTS Iteratioal Joural o Natural Laguage Computig (IJNLC) Vol. 2, No., February 203 EFFECT OF QUERY FORMATION ON WEB SEARCH ENGINE RESULTS Raj Kishor Bisht ad Ila Pat Bisht 2 Departmet of Computer Sciece &

More information

Fire Recognition in Video. Walter Phillips III Mubarak Shah Niels da Vitoria Lobo.

Fire Recognition in Video. Walter Phillips III Mubarak Shah Niels da Vitoria Lobo. Fire Recogitio i Video Walter Phillips III Mubarak Shah Niels da Vitoria Lobo {wrp65547,shah,iels}@cs.ucf.edu Computer Visio Laboratory Departmet of Computer Sciece Uiversity of Cetral Florida Orlado,

More information

( n+1 2 ) , position=(7+1)/2 =4,(median is observation #4) Median=10lb

( n+1 2 ) , position=(7+1)/2 =4,(median is observation #4) Median=10lb Chapter 3 Descriptive Measures Measures of Ceter (Cetral Tedecy) These measures will tell us where is the ceter of our data or where most typical value of a data set lies Mode the value that occurs most

More information

Face Anti-spoofing based on Deep Stack Generalization Networks

Face Anti-spoofing based on Deep Stack Generalization Networks Xi Nig,, Weiju Li,, Meili Wei, Liju Su, ad Xiaoli Dog, Istitute of Semicoductors, Chiese Academy of Scieces, 00083, Beijig, Chia Cogitive Computig Techology Wave Joit Lab, 00083, Beijig, Chia Keywords:

More information

Real-Time Secure System for Detection and Recognition the Face of Criminals

Real-Time Secure System for Detection and Recognition the Face of Criminals Available Olie at www.ijcsmc.com Iteratioal Joural of Computer Sciece ad Mobile Computig A Mothly Joural of Computer Sciece ad Iformatio Techology IJCSMC, Vol. 4, Issue. 9, September 2015, pg.58 83 RESEARCH

More information

Floristic Quality Assessment (FQA) Calculator for Colorado User s Guide

Floristic Quality Assessment (FQA) Calculator for Colorado User s Guide Floristic Quality Assessmet (FQA) Calculator for Colorado User s Guide Created by the Colorado atural Heritage Program Last Updated April 2012 The FQA Calculator was created by Michelle Fik ad Joaa Lemly

More information

A New Network-based Algorithm for Human Activity Recognition in Videos

A New Network-based Algorithm for Human Activity Recognition in Videos IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, A New Network-based Algorithm for Huma Activity Recogitio i Videos Weiyao Li, Yuazhe Che, Jiaxi Wu, Hali Wag, Bi Sheg, ad Hogxiag Li Abstract

More information

Software development of components for complex signal analysis on the example of adaptive recursive estimation methods.

Software development of components for complex signal analysis on the example of adaptive recursive estimation methods. Software developmet of compoets for complex sigal aalysis o the example of adaptive recursive estimatio methods. SIMON BOYMANN, RALPH MASCHOTTA, SILKE LEHMANN, DUNJA STEUER Istitute of Biomedical Egieerig

More information

Efficient Eye Location for Biomedical Imaging using Two-level Classifier Scheme

Efficient Eye Location for Biomedical Imaging using Two-level Classifier Scheme 828 Iteratioal Joural of Cotrol, Automatio, ad Systems, vol. 6, o. 6, pp. 828-835, December 2008 Efficiet Eye Locatio for Biomedical Imagig usig Two-level Classifier Scheme Mi Youg Nam, Xi Wag, ad Phill

More information

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

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

More information

Handwriting Stroke Extraction Using a New XYTC Transform

Handwriting Stroke Extraction Using a New XYTC Transform Hadwritig Stroke Etractio Usig a New XYTC Trasform Gilles F. Houle 1, Kateria Bliova 1 ad M. Shridhar 1 Computer Scieces Corporatio Uiversity Michiga-Dearbor Abstract: The fudametal represetatio of hadwritig

More information

Algorithms for Disk Covering Problems with the Most Points

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

More information

Fast Fourier Transform (FFT) Algorithms

Fast Fourier Transform (FFT) Algorithms Fast Fourier Trasform FFT Algorithms Relatio to the z-trasform elsewhere, ozero, z x z X x [ ] 2 ~ elsewhere,, ~ e j x X x x π j e z z X X π 2 ~ The DFS X represets evely spaced samples of the z- trasform

More information

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation

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

More information

Mapping Publishing and Mapping Adaptation in the Middleware of Railway Information Grid System

Mapping Publishing and Mapping Adaptation in the Middleware of Railway Information Grid System Mappig Publishig ad Mappig Adaptatio i the Middleware of Railway Iformatio Grid ystem You Gamei, Liao Huamig, u Yuzhog Istitute of Computig Techology, Chiese Academy of cieces, Beijig 00080 gameiu@ict.ac.c

More information

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

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

More information

CIS 121 Data Structures and Algorithms with Java Spring Stacks, Queues, and Heaps Monday, February 18 / Tuesday, February 19

CIS 121 Data Structures and Algorithms with Java Spring Stacks, Queues, and Heaps Monday, February 18 / Tuesday, February 19 CIS Data Structures ad Algorithms with Java Sprig 09 Stacks, Queues, ad Heaps Moday, February 8 / Tuesday, February 9 Stacks ad Queues Recall the stack ad queue ADTs (abstract data types from lecture.

More information

GPUMP: a Multiple-Precision Integer Library for GPUs

GPUMP: a Multiple-Precision Integer Library for GPUs GPUMP: a Multiple-Precisio Iteger Library for GPUs Kaiyog Zhao ad Xiaowe Chu Departmet of Computer Sciece, Hog Kog Baptist Uiversity Hog Kog, P. R. Chia Email: {kyzhao, chxw}@comp.hkbu.edu.hk Abstract

More information

Lecture 18. Optimization in n dimensions

Lecture 18. Optimization in n dimensions Lecture 8 Optimizatio i dimesios Itroductio We ow cosider the problem of miimizig a sigle scalar fuctio of variables, f x, where x=[ x, x,, x ]T. The D case ca be visualized as fidig the lowest poit of

More information

Road Boundary Detection in Complex Urban Environment based on Low- Resolution Vision

Road Boundary Detection in Complex Urban Environment based on Low- Resolution Vision Road Boudary Detectio i Complex Urba Eviromet based o Low- Resolutio Visio Qighua We, Zehog Yag, Yixu Sog, Peifa Jia State Key Laboratory o Itelliget Techology ad Systems, Tsighua Natioal Laboratory for

More information

Learning to Shoot a Goal Lecture 8: Learning Models and Skills

Learning to Shoot a Goal Lecture 8: Learning Models and Skills Learig to Shoot a Goal Lecture 8: Learig Models ad Skills How do we acquire skill at shootig goals? CS 344R/393R: Robotics Bejami Kuipers Learig to Shoot a Goal The robot eeds to shoot the ball i the goal.

More information

The measurement of overhead conductor s sag with DLT method

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

More information

Shape Completion and Modeling of 3D Foot Shape While Walking Using Homologous Model Fitting

Shape Completion and Modeling of 3D Foot Shape While Walking Using Homologous Model Fitting Shape Completio ad Modelig of 3D Foot Shape While Walkig Usig Homologous Model Fittig Yuji YOSHIDA* a, Shuta SAITO a, Yoshimitsu AOKI a, Makiko KOUCHI b, Masaaki MOCHIMARU b a Faculty of Sciece ad Techology,

More information

Cache-Optimal Methods for Bit-Reversals

Cache-Optimal Methods for Bit-Reversals Proceedigs of the ACM/IEEE Supercomputig Coferece, November 1999, Portlad, Orego, U.S.A. Cache-Optimal Methods for Bit-Reversals Zhao Zhag ad Xiaodog Zhag Departmet of Computer Sciece College of William

More information

FREQUENCY ESTIMATION OF INTERNET PACKET STREAMS WITH LIMITED SPACE: UPPER AND LOWER BOUNDS

FREQUENCY ESTIMATION OF INTERNET PACKET STREAMS WITH LIMITED SPACE: UPPER AND LOWER BOUNDS FREQUENCY ESTIMATION OF INTERNET PACKET STREAMS WITH LIMITED SPACE: UPPER AND LOWER BOUNDS Prosejit Bose Evagelos Kraakis Pat Mori Yihui Tag School of Computer Sciece, Carleto Uiversity {jit,kraakis,mori,y

More information

Arithmetic Sequences

Arithmetic Sequences . Arithmetic Sequeces COMMON CORE Learig Stadards HSF-IF.A. HSF-BF.A.1a HSF-BF.A. HSF-LE.A. Essetial Questio How ca you use a arithmetic sequece to describe a patter? A arithmetic sequece is a ordered

More information

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

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

More information

Lecture Notes 6 Introduction to algorithm analysis CSS 501 Data Structures and Object-Oriented Programming

Lecture Notes 6 Introduction to algorithm analysis CSS 501 Data Structures and Object-Oriented Programming Lecture Notes 6 Itroductio to algorithm aalysis CSS 501 Data Structures ad Object-Orieted Programmig Readig for this lecture: Carrao, Chapter 10 To be covered i this lecture: Itroductio to algorithm aalysis

More information

Descriptive Statistics Summary Lists

Descriptive Statistics Summary Lists Chapter 209 Descriptive Statistics Summary Lists Itroductio This procedure is used to summarize cotiuous data. Large volumes of such data may be easily summarized i statistical lists of meas, couts, stadard

More information

Octahedral Graph Scaling

Octahedral Graph Scaling Octahedral Graph Scalig Peter Russell Jauary 1, 2015 Abstract There is presetly o strog iterpretatio for the otio of -vertex graph scalig. This paper presets a ew defiitio for the term i the cotext of

More information

Alpha Individual Solutions MAΘ National Convention 2013

Alpha Individual Solutions MAΘ National Convention 2013 Alpha Idividual Solutios MAΘ Natioal Covetio 0 Aswers:. D. A. C 4. D 5. C 6. B 7. A 8. C 9. D 0. B. B. A. D 4. C 5. A 6. C 7. B 8. A 9. A 0. C. E. B. D 4. C 5. A 6. D 7. B 8. C 9. D 0. B TB. 570 TB. 5

More information

Computers and Scientific Thinking

Computers and Scientific Thinking Computers ad Scietific Thikig David Reed, Creighto Uiversity Chapter 15 JavaScript Strigs 1 Strigs as Objects so far, your iteractive Web pages have maipulated strigs i simple ways use text box to iput

More information

Chapter 3 Classification of FFT Processor Algorithms

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

More information

Position and Velocity Estimation by Ultrasonic Sensor

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

More information

Chapter 5. Functions for All Subtasks. Copyright 2015 Pearson Education, Ltd.. All rights reserved.

Chapter 5. Functions for All Subtasks. Copyright 2015 Pearson Education, Ltd.. All rights reserved. Chapter 5 Fuctios for All Subtasks Copyright 2015 Pearso Educatio, Ltd.. All rights reserved. Overview 5.1 void Fuctios 5.2 Call-By-Referece Parameters 5.3 Usig Procedural Abstractio 5.4 Testig ad Debuggig

More information

Counting Regions in the Plane and More 1

Counting Regions in the Plane and More 1 Coutig Regios i the Plae ad More 1 by Zvezdelia Stakova Berkeley Math Circle Itermediate I Group September 016 1. Overarchig Problem Problem 1 Regios i a Circle. The vertices of a polygos are arraged o

More information