Research Article GPU Acceleration of Melody Accurate Matching in Query-by-Humming

Size: px
Start display at page:

Download "Research Article GPU Acceleration of Melody Accurate Matching in Query-by-Humming"

Transcription

1 e Scentfc World Journal, Artcle ID , 7 pages Research Artcle GPU Acceleraton of Melody Accurate Matchng n Query-by-Hummng Lmn Xao, 1,2 Yao Zheng, 1,2,3 Wenq Tang, 1,2 Guangchao Yao, 1,2 and L Ruan 1,2 1 State Key Laboratory of Software Development Envronment, Behang Unversty, Bejng , Chna 2 School of Computer Scence and Engneerng, Behang Unversty, Bejng , Chna 3 Avaton Insttute, Bejng , Chna Correspondence should be addressed to Lmn Xao; xaolm@buaa.edu.cn and Yao Zheng; zyshren@163.com Receved 18 July 2013; Accepted 18 December 2013; Publshed 12 February 2014 Academc Edtors: Y. Chen and D. Wallom Copyrght 2014 Lmn Xao et al. Ths s an open access artcle dstrbuted under the Creatve Commons Attrbuton Lcense, whch permts unrestrcted use, dstrbuton, and reproducton n any medum, provded the orgnal work s properly cted. Wth the ncreasng scale of the melody database, the query-by-hummng system faces the trade-offs between response speed and retreval accuracy. Melody accurate matchng s the key factor to restrct the response speed. In ths paper, we present a GPU acceleraton method for melody accurate matchng, n order to mprove the response speed wthout reducng retreval accuracy. The method develops two parallel strateges (ntra-task parallelsm and nter-task parallelsm) to obtan accelerated effects. The effcency of our method s valdated through extensve experments. Evaluaton results show that our sngle GPU mplementaton acheves 20x to 40x speedup rato, when compared to a typcal general purpose CPU s executon tme. 1. Introducton Wth the development of nformaton technology, musc retreval technology has been wdely appled. Query-byhummng (QBH) s an mportant applcaton for musc retreval, where users can hum a melody to retreve the song [1 3]. Dfferent from the tradtonal musc retreval engne, whch searches the song based on the descrpton of the musc such as the snger or the song name, QBH retreves the song based on the content. Musc retreval s becomng more natural, smple, and user-frendly wth the advancement of QBH. Thus, QBH wll gve broader applcaton prospects for musc retreval [4 6]. Typcal QBH systems consst of three modules, ncludng feature extracton, melody database, and melody matchng [7]. The feature extracton module ncludes ptch extract and note segmentaton. Ptch extract means that the acoustc nput s frst put nto frames; then we obtan the ptch frequences from the frames; fnally we use (1) totransform them nto the representaton of semtone. Note segmentaton means that the obtaned ptch sequence s cut nto dfferent notes. The melody database module s responsble for the management and ndex MIDI melody nformaton. Consder semtone =12 log 2 ( freq ) (1) 440 The core of QBH s the melody matchng between the hummng melody and the melody database. From the vew of practcal applcaton, melody matchng algorthms can be dvded nto fast match and accurate match [8, 9]. Fast match manly removed the less lkely canddates from the melody database, n order to reduce the computaton of accurate match and mprove the system s response tme; accurate match then calculated the more exact smlartes between thehummngmelodyandthecanddatemelodes,soasto obtan the fnal lst of smlar songs. Hence, we can see that the characterstcs of accurate match are a large amount of calculatonandprecsecalculatons. There has to be a trade-off between response speed and accuracy, f the QBH system wants to mprove the retreval accuracy wthn acceptable tmes. However, wth the ncreasng scale of the melody database, the sze of canddate

2 2 The Scentfc World Journal melodes s also growng, whch results n an extended response tme and poor user experence. The usual method formprovngtheresponsespeedstoreducetheszeofthe canddate melodes, but ths may cause low retreval accuracy [10]. Therefore, many efforts are made to develop methods and technques that execute the melody matchng n hgh performance platforms, allowng the producton of accurate results n a shorter tme. Graphc Processng Unt (GPU) s one of the recent trends n hgh performance platforms, whch has demonstrated sgnfcant speedups for many scentfc applcatons [11, 12]. Researchers also have studed the melody matchng on GPUs [13, 14]. However, exstng works choose to mplement a coarse-graned parallelzaton of the melody matchng algorthm, where multple problem nstances are smply replcated onto each multprocessor of a GPU. In ths paper, by analyzng the exstng accurate matchng algorthms, we propose a GPU-based parallelzaton for the accurate matchng algorthm. Our proposed method develops two parallel strateges (ntratask parallelsm and ntertask parallelsm). By takng full advantage of multcomputng unts of GPU, our method can greatly decrease the computaton tme of the accurate match algorthm and mprove the system s response speed wthout reducng the retreval accuracy. The rest of ths paper s organzed as follows. Secton 2 summarzes the melody accurate matchng. In Secton 3, an outlne of the CUDA-based GPU computng platform s gven. Secton 4 descrbes the detals of our parallelzaton strategy for melody accurate matchng usng CUDA. Expermental results are analyzed n Secton 5. The last secton concludes the whole paper and ponts out some future works. 2. Melody Accurate Matchng n QBH 2.1. Problem Defnton. In a QBH system, the hummng nput s frstly put nto a ptch tracker frame by frame and then the output ptch sequence s converted to a semtone scale accordng to (1). The MIDI man melody extracted from the multtrack melody s put nto a note sequence. Defnton 1. The hummng ptch sequence at semtone scale of length m s denoted by Q = (q 1,q 2,...,q m ),where q represents the ptch correspondng to the th frame of hummng. Defnton 2. The MIDI note sequence matched wth Q s denoted by P = {(p 1,d 1 ), (p 2,d 2 ),...,(p n,d n )}, wherep and d represent the ptch and duraton of the th note, respectvely. Defnton 3. The dstance functon between the hummng sequence and the MIDI note sequence s denoted by dst(q,p j ),whereq and p j represent the ptch of hummng sequence and MIDI melody. Therefore, based on the above defntons, the problem of melody accurate matchng could be expressed, gven hummng sequence Q, MIDI note sequence P, and dstance functon,howtocalculatethemnmumdstancebetweenthe two sequences Melody Accurate Matchng Algorthms. In practcal hummng applcatons, dfferent people have dfferent vocal ranges, resultng n tonal nconsstency between hummng melody and standard melody but tonal change consstency; smlarly, dfferent people have dfferent sngng speed, resultng n sngng speed rato nconsstency between the two melodes. Thus, there exst speed change and ptch offset between hummng melody and standard MIDI melody. How toevaluatethespeedchangeandptchoffsetbeforeaccurate matchng s the dffculty of melody accurate matchng algorthms. Exstng matchng algorthms, ncludng lnear scalng (LS) [15], recursve algnment (RA) [9], and dynamc tme warpng (DTW) [16], take dfferent approaches to resolve the above dffculty and mprove retreval accuracy. Lnear scalng (LS) s a smple, ntutve, and effectve melody matchng algorthm. LS chooses dfferent scalng factors to stretch or compress the ptch contour of the hummng melody to more accurately match the MIDI melody. The defcency of LS s also qute obvous: local msmatch may deterorate the global matchng; the selecton of sutable stretchng coeffcent or ptch offset s dffcult; the length of hummng melody may affect the retreval accuracy. Recursve algnment (RA) uses recursve lnear scalng match to explore the optmal matchng results. Although t s derved from the LS, ths method overcomes the dsadvantage of usng sngle stretch factor or ptch offset throughout the entre melody matchng. Hence, RA s capable of capturng long dstance nformaton n human sngng. The drawbacks of ths approach are matchng the fner, hgher tme complexty; the segmentaton fragments based on mdpont, and the MIDI melodes do not necessarly satsfy the lnear relatonshp. Dynamc tme warpng (DTW) s one of the most effectve approaches to melody accurate matchng, whch s a frame-based dynamc programmng algorthm. In DTW, the canddate note sequences should be expanded to frame sequences. Compared to LS and RA, DTW has hgher accuracy but hgher complexty. The detals of DTW-based melody accuracy matchng are descrbed n followng secton. 3. GPU Computng wth CUDA Over the past few years, GPU has ganed sgnfcant popularty as a powerful tool for hgh performance computng. Wth the development of general programmng toolkts, such as Compute Unfed Devce Archtecture (CUDA), GPU becomes a general-purpose shared-memory many-core computng platform and plays mportant roles n applcatons such as computer vson, search, and sortng [17, 18]. CUDA s an extenson of C/C++ whch enables users to wrte scalable multthreaded programs for CUDA-enabled GPUs [19]. In the CUDA programmng model, an applcaton conssts of a sequental host program, whch can execute parallel programs known as kernels on GPUs. A kernel s

3 The Scentfc World Journal 3 Input: Hummng ptch sequence Q=(q 1,q 2,...,q m ) MIDI note sequence P = {(p 1,d 1 ), (p 2,d 2 ),...,(p n,d n )} Scalng factor set S=(sc 1, sc 2,...,sc k ) Output: Mnmum dstance s mn (1) T =LS(Q, P, S)//T s the best scalng hummng ptch sequence (2) Standardze hummng ptch sequence T (3) Transform note sequence P nto frame-based ptch sequence P =(p 1,p 2,...,p t ) (4) Standardze canddate ptch sequence P (5) Intalze ptch offset s pan (6) whle (s pan > 0.01) then (7) s left =dtw(t, P s pan ) (8) s mddle =dtw(t, P ) (9) s rght =dtw(t, P +s pan ) (10) s mn = mn{s left,s mddle,s rght } (11) s pan = s pan /2 (12) end whle (13) return s mn Algorthm 1: DTW-based match (Q, P, S). executed usng potentally large number of parallel threads. Thus, GPU acheves massve parallelsm through executng a large number of lghtweght threads concurrently. These threads are organzed n thread blocks and grds of thread blocks. Each thread runs the same sequental program and has a thread ID wthn ts thread block. The herarchcal organzaton nto blocks and grds has mplcatons for thread communcaton and synchronzaton. Threads wthn a thread block can communcate through a per-block shared memory and may synchronze usng barrers. However, threads located n dfferent blocks cannot communcate or synchronze drectly. The memory model n CUDA has fve types of memory: each thread has prvate local memory; each thread block has shared memory vsble to all threads of the block; all threads have access to the same global memory; there are also two addtonal read-only memores accessble by all threads: the constant, and texture memory. Thread creaton, schedulng, and management are performed entrely n hardware. In order to manage large number of threads, the GPU employs the SIMT (Sngle Instructon Multple Thread) archtecture [20] n whch the threads of ablockareexecutedngroupsof32calledwarps.thewarp can greatly mprove performance by havng threads n a warp execute the same code path and access memory n nearby addresses. 4. GPU-Parallel Melody Accurate Matchng As descrbed n Secton 2.2, DTW s an effcent melody matchng algorthm; however, the tme complexty s O(mt). Ths s the motvaton for our research, hopng that we can accelerate DTW computaton usng GPU wthout reducng the accuracy Melody Accurate Matchng Based on DTW. The melody accurate matchng based on DTW s outlned n Algorthm 1. Suppose that the query ptch sequence s denoted by Q = (q 1,q 2,...,q m ), and the note sequence s denoted by P = {(p 1,d 1 ), (p 2,d 2 ),...,(p n,d n )}.Thenwecanconstructam t DTW matrx D accordng to (2). D(, j) s the mnmum dstance startng from the left-most sde ( = 1) of the matrx to the current poston (, j). Consder D (, j) = dst (q,p j )+mn { D ( 2, j 1) D ( 1, j 1) { { D ( 1, j 2). The correspondng boundary condtons for the above recurson can be expressed as D (, 1) =, D(1,j)=dst (1,j), D (, 0) =, D(0,j)=, D (0, 0) =0. =2,...,m j=1,...,t =1,...,m j=1,...,t The cost of the optmal DTW path s defned as (2) (3) mn j=1 to t D(m,j). (4) After fndng the optmzng j,the optmal DTW path can be obtaned by back trackng. As prevously descrbed, tempo varaton and ptch transposton should be consdered before the accurate matchng. For most users, tempo varaton s attrbuted to lnear varaton. Researchers apply the LS algorthm to the hummngptchsequencebeforecomparngttothecanddate melody. The recurrent relaton shows that the optmal path exsts only when the hummng nput s wthn half to twce the sze of the canddate melody. Hence, the hummng

4 4 The Scentfc World Journal Input: Hummng ptch sequence Q=(q 1,q 2,...,q m ) Frame-based canddate ptch sequence P =(p 1,p 2,...,p t ) Output:CostmatrxD (1) Intalze the frst two rows and columns of the matrx D (2) for j=2to t do (3) for =2to m parallel do (4) calculate D(, j) (5) end for (6) synchronze( ) (7) end for (8) return D Algorthm 2: Parallel mplementaton for calculatng the matrx D PDTW (Q, P). sequencecanbescaledseveraltmes,rangngfrom0.5to 2 tmes the orgnal length and compared to the canddate melody n order to acheve the best scalng factor. For ptch transposton, a heurstc method can be appled that shfts the entre hummng ptch sequence to a sutable poston n order to generate the mnmum DTW dstance. m 4.2. Parallelzaton Strategy for Melody Accurate Matchng. In ths secton, our proposed parallelzaton strateges for melody accurate matchng based on DTW are explaned. We nvestgate two approaches for parallelzng the melody accurate matchng usng CUDA: ntra-task parallelzaton and nter-task parallelzaton. Intra-task parallelzaton ndcates that each task s assgned to one thread block and all threads n the thread block cooperate to perform the task n parallel. Inter-task parallelzaton ndcates that each task s assgned to exactly one thread and all threads n a thread block perform the tasks n parallel. 0 0 j t Intra-Task Parallelzaton. As descrbed n the above secton, the DTW algorthm has quadratc tme complexty that lmts ts usefulness. The purpose of computaton durng the DTW algorthm s to fll the matrx D,whchcanbeeasly mplemented as a smple two nested loops. Accordng to (2), agvend(, j) can be computed only f D( 2,j 1),D( 1, j 1), andd( 1, j 2) have already been computed. For nstance, as shown n Fgure 1, D(2, 2) (red cell locates) depends on D(0, 1), D(1, 1), and D(1, 0) (green cells locate). Ths ndcates that all elements n column two are computable smultaneouslyftheelementsncolumnzeroandcolumn one have already been computed. Thus, every entry of the same column s computable and elements wthn the column canbecomputednparallel. Based on ths dea, we present an effcent parallel mplementaton for calculatng matrx D,as shown n Algorthm 2. The nput to the algorthm s two ptch sequences. The output of the algorthm s matrx D. Lnes3 5arethecoreofour parallel algorthm, whch calculates each element of a column n parallel. After calculatng all elements of each column, synchronzaton operaton s performed to ensure that all processors have fnshed current computaton (lne 6). Fgure 1: Column-wse fashon for computng matrx D. We can easly determne the tme complexty of Algorthm 2. Assume that there are p processors for computng smultaneously. Based on Algorthm 2, the total executon tme can be expressed as T (m, t) = t j=2 m 1 p =O( 1 mt). (5) p Inter-Task Parallelzaton. As descrbed n above the secton, nter-task parallelzaton ndcates that we compute n parallel all the scores of a hummng query wth every peceofmelodynthecanddatemelodyset.fornstance, f the canddate set contans N peces of melodes, we can launch N threads executng Algorthm 2 n parallel. The man challenges are the optmzaton of resource allocatons and memory operatons. Westorethehummngptchsequencentexturememory, whch s the cached, fast memory and shared among all threads. Because of the ncreasng scale of the melody

5 The Scentfc World Journal 5 Input: Hummng ptch sequence Q Canddate melody set N=(P 1,P 2,...,P l ) Scalng factor set S=(sc 1, sc 2,...,sc k ) Output: Melody smlarty lst (1) Load Q, N to GPU devce (2) for =1to l parallel do (3) T =LS(Q, P,S) (4) Standardze hummng ptch sequence T (5) Transform P nto frame-based ptch sequence P (6) Standardze canddate ptch sequence P (7) Intalze ptch offset s pan (8) whle (s pan > 0.01) then (9) s left =PDTW(T, P s pan ) (11) s mddle =PDTW(T, P ) (12) s rght =PDTW(T, P +s pan ) (13) s mn = mn{s left,s mddle,s rght } (14) s pan =s pan /2 (15) end whle (16) end for (17) Load the matrces back to host (18) Calculate the melody smlarty lst Algorthm 3: GPU-based parallel melody accurate matchng. database,thecanddatemelodysetusuallycontanstoomany peces of melodes to be stored n any of the cached memores (texture, constant and shared memory). Therefore, we store the canddate melody set n the global memory. To gan maxmum bandwdth and best performance, all threads n a half-warpshouldaccessthenotesequencesnglobalmemory n a coalesced pattern. In order to adopt the coalesced pattern, everynotesequencesstorednanarrayofdouble-float, a CUDA structure contanng two 32-bt floats storng the ptch and duraton of each note. Then we organze all note sequences n memory as a one-dmensonal array, such that ts frst N entres correspond to the frst note of each sequence; the next N entres correspond to the second note of each sequence, and so forth. Each thread performs the computatons on ts own matrx D. In order to mnmze the amount of requred memory, we only store the current column and forward two columns of each matrx n shared memory. Fnally, n order to allow smultaneous read/wrte operatons by the actve threads, we store the matrces usng the same strategy as the note sequences Implementaton wth CUDA. Basedontheaboveparallelzaton strateges, we present an acceleraton method for melody accurate matchng, as shown n Algorthm3. Specfcally, ntra-task parallelzaton s performed n a thread block; nter-task parallelzaton s performed among thread blocks. 5. Performance Evaluaton In order to evaluate the performance of our GPU-parallel melody accurate matchng algorthm, several experments Table 1: Evaluaton envronment. CPU (AthlonII X4) GTX285 GTX480 Number of Core (30SM) 480 (15SM) SP clock 3.2 GHz GHz GHz Compler GCC CUDA SDK 4.0 CUDA SDK 4.0 were carred out. We frst ntroduce the sngng corpus used n ths study. Then, we compare the performance and accuracy between our parallel mplementaton and the exstng sequental mplementaton Evaluaton Envronment. Evaluaton envronment s shown n Table 1. We use an AMD AthlonII X4 3.2 GHz for CPU, NVIDIA GTX285 wth 240 SPs (Stream processor), and GTX480 wth 480 SPs for GPUs. To evaluate our proposed acceleraton method, we used the publcly avalable MIREX (musc nformaton retreval evaluaton exchange) QBSH corpus [21], whch has been used for the evaluaton of QBSH for many tmes. The corpus ncludes 48 MIDI fles and 4431 sngng or hummng clps. Each clp has duraton of 8 s, wth an 8 khz samplng rate and 8 bt resoluton. The frame sze s 256 and the overlap s 0, resultng n a ptch rate of 8000/256 = Hence, an 8 s sngng clp s converted to a ptch vector of =250 elements n semtones. We add MIDI fles to MIREX corpus to compose the MIDI database. These MIDI fles are collectedfrommidiarchvesonthenternet.

6 6 The Scentfc World Journal Tme (s) CPU GTX GTX K 5 K 10 K Scale of MIDI database (1K = 1000) Fgure 2: Comparson of executon tmes wth dfferent scale of MIDI database. Speedup rato K 5 K 10K Scale of MIDI database (1K = 1000) GTX285 GTX480 Fgure 3: Speedup rato obtaned wth the usage of GPUs Performance Results. We ntroduce the speedup rato to evaluate the performance of GPU-parallel melody accurate matchng algorthm. Top-M ht rate and mean recprocal rank(mrr)areusedtoevaluatetheaccuracyofourproposed algorthm. MRR s a standard metrcs n MIREX, whch can be denoted as n MRR = 1 1, (6) n rank where rank means the rank of the correct song for the th query. Fgure2 shows the executon tmes of dfferent hardware settngs wth respect to the varyng scale of MIDI database. We observe that the performance s senstve to the problem sze and hardware settngs. The results ndcate that the more cores are on GPU, the greater the performance s mproved. As shown n Fgure 3, the speedup rato over CPU seral versonrangesfrom20to40ongpus.theresultsalso ndcate that the fluctuaton of speedup rato s very small as the scale of MIDI database ncreases. To verfy whether the accelerated strategy affects the retreval accuracy, we evaluate the accuracy of our proposed parallel algorthm smultaneously. We selected 829 sngng clps from the corpus to test the performance. As shown =1 Table 2: Retreval results for dfferent methods. Method MRR Top-1 Top-3 Top-5 Top-10 Seral verson Parallel verson n Table 2, our proposed parallel algorthm almost does not affect the retreval accuracy. 6. Concluson and Future Work Ths paper presents a GPU-parallel melody accurate matchng method for query-by-hummng. The method uses two parallelzaton strateges (ntra-task parallelzaton and ntertask parallelzaton) to accelerate melody matchng. The expermental results show that our proposed method can obtan 20x to 40x speedups wthout reducng retreval accuracy. For future work, we wll further optmze the parallel program to mprove the performance. Second, we wll perform exhaustve comparson wth other QBH acceleraton methods. Conflct of Interests The authors declare that there s no conflct of nterests regardng the publcaton of ths paper. Acknowledgments Ths research was funded by the H-tech Research and Development Program of Chna (863 Program) under Grant no. 2011AA01A205, the Natonal Natural Scence Foundaton of Chna under Grant nos , , and , the Doctoral Fund of Mnstry of Educaton of Chna under Grant no , Bejng Natural Scence Foundaton undergrantno ,andthefundofthestatekey Laboratory of Software Development Envronment under Grant no. SKLSDE-2012ZX-23. References [1] A. Ghas, J. Logan, D. Chamberln, and B. C. Smth, Query by hummng: muscal nformaton retreval n an audo database, n Proceedngs of the 3rd Internatonal Multmeda Conference and Exhbton (MULTIMEDIA 95), pp , November [2] G.P.Nam,T.T.T.Luong,andH.H.Nam, Intellgentquery by hummng system based on score level fuson of multple classfers, EURASIP Journal on Sgnal Processng, vol. 2011, no. 21, pp. 1 11, [3] Q. Wang, Z. Guo, G. Lu, J. Guo, and Y. Lu, Query by hummng by usng localty senstve hashng based on combnaton of ptch and note, n Proceedngs of the IEEE Internatonal Conference on Multmeda and Expo, pp , [4] K. Km, K. R. Park, S.-J. Park, S.-P. Lee, and M. Y. Km, Robust query-by-sngng/hummng system aganst background nose envronments, IEEE Transactons on Consumer Electroncs,vol. 57, no. 2, pp , 2011.

7 The Scentfc World Journal 7 [5] H.-M. Yu, W.-H. Tsa, and H.-M. Wang, A query-by-sngng system for retrevng Karaoke musc, IEEE Transactons on Multmeda,vol.10,no.8,pp ,2008. [6] J.-S. R. Jang and H.-R. Lee, A general framework of progressve flterng and ts applcaton to query by sngng/hummng, IEEE Transactons on Audo, Speech and Language Processng, vol. 16, no. 2, pp , [7] S. Jo and C. D. Yoo, Melody extracton from polyphonc audo basedonpartcleflter, nproceedngs of the Internatonal SymposumonMuscInformatonRetreval,pp ,2010. [8] Y. Zhu and D. Shasha, Warpng ndexes wth envelope transforms for query by hummng, n Proceedngs of the ACM SIGMOD Internatonal Conference on Management of Data,pp , June [9] G. C. Yao, Y. Zheng, L. M. Xao, L. Ruan, and Y. N. L, Effcent vocal melody extracton from polyphonc musc sgnals, Electroncs and Electrcal Engneerng,vol.19,no.6,pp , [10] J.Hou,D.-N.Jang,W.-X.Cao,Y.Qn,T.F.Zheng,andY.Lu, Effectveness of N-gram fast match for query-by-hummng systems, n Proceedngs of the IEEE Internatonal Conference on Multmeda and Expo (ICME 09),pp ,July2009. [11] L. M. Xao, Y. Zheng, W. Q. Tang, and G. C. Yao L, A GPUaccelerated large-scale musc smlarty retreval method, n Proceedngs of the IEEE Internatonal Conference on Internet of Thngs,2013. [12]L.M.Xao,Y.Zheng,W.Q.Tang,G.C.Yao,andL.Ruan, Parallelzng dynamc tme warpng algorthm usng prefx computatons on GPU, n Proceedngs of the 15th IEEE Internatonal Conference on Hgh Performance Computng and Communcatons,2013. [13] D. Sart, A. Mueen, W. Najjar, E. Keogh, and V. Nennattrakul, Acceleratng dynamc tme warpng subsequence search wth GPUs and FPGAs, n Proceedngs of the 10th IEEE Internatonal Conference on Data Mnng (ICDM 10), pp , December [14] Y. D. Zhang, K. Adl, and J. Glass, Fast spoken query detecton usng lower-bound dynamc tme warpng on graphcal processng unts, n Proceedngs of the IEEE Internatonal Conference on Acoustcs, Speech and Sgnal Process, pp , [15] J. S. R. Jang, H. R. Lee, and M. Y. Kao, Content-based musc retreval usng lnear scalng and branch-and-bound tree search, n Proceedngs of the Internatonal Conference on Multmeda&Expo, pp , [16] J.S.R.JangandM.Y.Gao, Aquery-by-Sngngsystembased on dynamc programmng, n Proceedngs of the Internatonal Workshop Intellgent System Resolutons, pp , [17] E. Lndholm, J. Nckolls, S. Oberman, and J. Montrym, NVIDIA Tesla: a unfed graphcs and computng archtecture, IEEE Mcro,vol.28,no.2,pp.39 55,2008. [18] NVda Corporaton, NVda s Next Generaton CUDA Compute Archtecture: Ferm, verson 1.1, content/pdf/ferm whte papers/nvidia Ferm Compute Archtecture Whtepaper.pdf. [19] J. Nckolls, I. Buck, M. Garland, and K. Skadron, Scalable parallel programmng wth CUDA, Queue,vol.6,no.2,pp.40 53, [20] NVda Corporaton, NVda CUDA Programmng Gude, verson 4. 2, 2012, [21] MIREX, Query by Sngng/Hummng, 2012, by Sngng/Hummng.

8 Journal of Industral Engneerng Multmeda The Scentfc World Journal Appled Computatonal Intellgence and Soft Computng Internatonal Journal of Dstrbuted Sensor Networks Fuzzy Systems Modellng & Smulaton n Engneerng Submt your manuscrpts at Journal of Computer Networks and Communcatons Artfcal Intellgence Internatonal Journal of Bomedcal Imagng Artfcal Neural Systems Internatonal Journal of Computer Engneerng Computer Games Technology Software Engneerng Internatonal Journal of Reconfgurable Computng Robotcs Computatonal Intellgence and Neuroscence Human-Computer Interacton Journal of Journal of Electrcal and Computer Engneerng

Parallelism for Nested Loops with Non-uniform and Flow Dependences

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

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

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

More information

Cluster Analysis of Electrical Behavior

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

More information

An Optimal Algorithm for Prufer Codes *

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

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

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

More information

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

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

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

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

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

More information

Efficient Distributed File System (EDFS)

Efficient Distributed File System (EDFS) Effcent Dstrbuted Fle System (EDFS) (Sem-Centralzed) Debessay(Debsh) Fesehaye, Rahul Malk & Klara Naherstedt Unversty of Illnos-Urbana Champagn Contents Problem Statement, Related Work, EDFS Desgn Rate

More information

Parallel matrix-vector multiplication

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

More information

Array transposition in CUDA shared memory

Array transposition in CUDA shared memory Array transposton n CUDA shared memory Mke Gles February 19, 2014 Abstract Ths short note s nspred by some code wrtten by Jeremy Appleyard for the transposton of data through shared memory. I had some

More information

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

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

More information

A Binarization Algorithm specialized on Document Images and Photos

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

More information

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

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

More information

The Shortest Path of Touring Lines given in the Plane

The Shortest Path of Touring Lines given in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 262 The Open Cybernetcs & Systemcs Journal, 2015, 9, 262-267 The Shortest Path of Tourng Lnes gven n the Plane Open Access Ljuan Wang 1,2, Dandan He

More information

Music/Voice Separation using the Similarity Matrix. Zafar Rafii & Bryan Pardo

Music/Voice Separation using the Similarity Matrix. Zafar Rafii & Bryan Pardo Musc/Voce Separaton usng the Smlarty Matrx Zafar Raf & Bryan Pardo Introducton Muscal peces are often characterzed by an underlyng repeatng structure over whch varyng elements are supermposed Propellerheads

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

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

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

A Similarity Measure Method for Symbolization Time Series

A Similarity Measure Method for Symbolization Time Series Research Journal of Appled Scences, Engneerng and Technology 5(5): 1726-1730, 2013 ISSN: 2040-7459; e-issn: 2040-7467 Maxwell Scentfc Organzaton, 2013 Submtted: July 27, 2012 Accepted: September 03, 2012

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

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

More information

Fast Computation of Shortest Path for Visiting Segments in the Plane

Fast Computation of Shortest Path for Visiting Segments in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 4 The Open Cybernetcs & Systemcs Journal, 04, 8, 4-9 Open Access Fast Computaton of Shortest Path for Vstng Segments n the Plane Ljuan Wang,, Bo Jang

More information

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

Positive Semi-definite Programming Localization in Wireless Sensor Networks

Positive Semi-definite Programming Localization in Wireless Sensor Networks Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer

More information

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.

More information

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

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

More information

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,

More information

A fast algorithm for color image segmentation

A fast algorithm for color image segmentation Unersty of Wollongong Research Onlne Faculty of Informatcs - Papers (Arche) Faculty of Engneerng and Informaton Scences 006 A fast algorthm for color mage segmentaton L. Dong Unersty of Wollongong, lju@uow.edu.au

More information

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole Appled Mathematcs, 04, 5, 37-3 Publshed Onlne May 04 n ScRes. http://www.scrp.org/journal/am http://dx.do.org/0.436/am.04.584 The Research of Ellpse Parameter Fttng Algorthm of Ultrasonc Imagng Loggng

More information

Research and Application of Fingerprint Recognition Based on MATLAB

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

More information

Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier

Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier Some materal adapted from Mohamed Youns, UMBC CMSC 611 Spr 2003 course sldes Some materal adapted from Hennessy & Patterson / 2003 Elsever Scence Performance = 1 Executon tme Speedup = Performance (B)

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

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

More information

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

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

More information

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

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

More information

The Codesign Challenge

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

More information

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

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

More information

Boundary-Based Time Series Sorting

Boundary-Based Time Series Sorting JOURNAL OF ELECTRONIC SCIENCE AND TECHNOLOGY OF CHINA, VOL. 6, NO. 3, SEPTEMBER 2008 323 Boundary-Based Tme Seres Sortng Jun-Ku L, Yuan-Zhen Wang, and Ha-Bo L Abstract In many applcatons, t s desrable

More information

Classifier Selection Based on Data Complexity Measures *

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

More information

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

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

More information

Machine Learning: Algorithms and Applications

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

More information

An Image Fusion Approach Based on Segmentation Region

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

More information

High-Boost Mesh Filtering for 3-D Shape Enhancement

High-Boost Mesh Filtering for 3-D Shape Enhancement Hgh-Boost Mesh Flterng for 3-D Shape Enhancement Hrokazu Yagou Λ Alexander Belyaev y Damng We z Λ y z ; ; Shape Modelng Laboratory, Unversty of Azu, Azu-Wakamatsu 965-8580 Japan y Computer Graphcs Group,

More information

Wavefront Reconstructor

Wavefront Reconstructor A Dstrbuted Smplex B-Splne Based Wavefront Reconstructor Coen de Vsser and Mchel Verhaegen 14-12-201212 2012 Delft Unversty of Technology Contents Introducton Wavefront reconstructon usng Smplex B-Splnes

More information

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

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

More information

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images

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

More information

Scheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research

Scheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research Schedulng Remote Access to Scentfc Instruments n Cybernfrastructure for Educaton and Research Je Yn 1, Junwe Cao 2,3,*, Yuexuan Wang 4, Lanchen Lu 1,3 and Cheng Wu 1,3 1 Natonal CIMS Engneerng and Research

More information

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

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

More information

Distributed Resource Scheduling in Grid Computing Using Fuzzy Approach

Distributed Resource Scheduling in Grid Computing Using Fuzzy Approach Dstrbuted Resource Schedulng n Grd Computng Usng Fuzzy Approach Shahram Amn, Mohammad Ahmad Computer Engneerng Department Islamc Azad Unversty branch Mahallat, Iran Islamc Azad Unversty branch khomen,

More information

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

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

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

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

More information

3D vector computer graphics

3D vector computer graphics 3D vector computer graphcs Paolo Varagnolo: freelance engneer Padova Aprl 2016 Prvate Practce ----------------------------------- 1. Introducton Vector 3D model representaton n computer graphcs requres

More information

Shape-adaptive DCT and Its Application in Region-based Image Coding

Shape-adaptive DCT and Its Application in Region-based Image Coding Internatonal Journal of Sgnal Processng, Image Processng and Pattern Recognton, pp.99-108 http://dx.do.org/10.14257/sp.2014.7.1.10 Shape-adaptve DCT and Its Applcaton n Regon-based Image Codng Yamn Zheng,

More information

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

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

More information

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

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

More information

Security Vulnerabilities of an Enhanced Remote User Authentication Scheme

Security Vulnerabilities of an Enhanced Remote User Authentication Scheme Contemporary Engneerng Scences, Vol. 7, 2014, no. 26, 1475-1482 HIKARI Ltd, www.m-hkar.com http://dx.do.org/10.12988/ces.2014.49186 Securty Vulnerabltes of an Enhanced Remote User Authentcaton Scheme Hae-Soon

More information

ELEC 377 Operating Systems. Week 6 Class 3

ELEC 377 Operating Systems. Week 6 Class 3 ELEC 377 Operatng Systems Week 6 Class 3 Last Class Memory Management Memory Pagng Pagng Structure ELEC 377 Operatng Systems Today Pagng Szes Vrtual Memory Concept Demand Pagng ELEC 377 Operatng Systems

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

More information

Speedup of Type-1 Fuzzy Logic Systems on Graphics Processing Units Using CUDA

Speedup of Type-1 Fuzzy Logic Systems on Graphics Processing Units Using CUDA Speedup of Type-1 Fuzzy Logc Systems on Graphcs Processng Unts Usng CUDA Durlabh Chauhan 1, Satvr Sngh 2, Sarabjeet Sngh 3 and Vjay Kumar Banga 4 1,2 Department of Electroncs & Communcaton Engneerng, SBS

More information

Research Article A High-Order CFS Algorithm for Clustering Big Data

Research Article A High-Order CFS Algorithm for Clustering Big Data Moble Informaton Systems Volume 26, Artcle ID 435627, 8 pages http://dx.do.org/.55/26/435627 Research Artcle A Hgh-Order Algorthm for Clusterng Bg Data Fanyu Bu,,2 Zhku Chen, Peng L, Tong Tang, 3 andyngzhang

More information

Analysis on the Workspace of Six-degrees-of-freedom Industrial Robot Based on AutoCAD

Analysis on the Workspace of Six-degrees-of-freedom Industrial Robot Based on AutoCAD Analyss on the Workspace of Sx-degrees-of-freedom Industral Robot Based on AutoCAD Jn-quan L 1, Ru Zhang 1,a, Fang Cu 1, Q Guan 1 and Yang Zhang 1 1 School of Automaton, Bejng Unversty of Posts and Telecommuncatons,

More information

Mining User Similarity Using Spatial-temporal Intersection

Mining User Similarity Using Spatial-temporal Intersection www.ijcsi.org 215 Mnng User Smlarty Usng Spatal-temporal Intersecton Ymn Wang 1, Rumn Hu 1, Wenhua Huang 1 and Jun Chen 1 1 Natonal Engneerng Research Center for Multmeda Software, School of Computer,

More information

Structural Analysis of Musical Signals for Indexing and Thumbnailing

Structural Analysis of Musical Signals for Indexing and Thumbnailing Structural Analyss of Muscal Sgnals for Indexng and Thumbnalng We Cha Barry Vercoe MIT Meda Laboratory {chawe, bv}@meda.mt.edu Abstract A muscal pece typcally has a repettve structure. Analyss of ths structure

More information

Support Vector Machines

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

More information

Querying by sketch geographical databases. Yu Han 1, a *

Querying by sketch geographical databases. Yu Han 1, a * 4th Internatonal Conference on Sensors, Measurement and Intellgent Materals (ICSMIM 2015) Queryng by sketch geographcal databases Yu Han 1, a * 1 Department of Basc Courses, Shenyang Insttute of Artllery,

More information

GSLM Operations Research II Fall 13/14

GSLM Operations Research II Fall 13/14 GSLM 58 Operatons Research II Fall /4 6. Separable Programmng Consder a general NLP mn f(x) s.t. g j (x) b j j =. m. Defnton 6.. The NLP s a separable program f ts objectve functon and all constrants are

More information

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity Journal of Sgnal and Informaton Processng, 013, 4, 114-119 do:10.436/jsp.013.43b00 Publshed Onlne August 013 (http://www.scrp.org/journal/jsp) Corner-Based Image Algnment usng Pyramd Structure wth Gradent

More information

An Image Compression Algorithm based on Wavelet Transform and LZW

An Image Compression Algorithm based on Wavelet Transform and LZW An Image Compresson Algorthm based on Wavelet Transform and LZW Png Luo a, Janyong Yu b School of Chongqng Unversty of Posts and Telecommuncatons, Chongqng, 400065, Chna Abstract a cylpng@63.com, b y27769864@sna.cn

More information

Concurrent Apriori Data Mining Algorithms

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

More information

Performance Assessment and Fault Diagnosis for Hydraulic Pump Based on WPT and SOM

Performance Assessment and Fault Diagnosis for Hydraulic Pump Based on WPT and SOM Performance Assessment and Fault Dagnoss for Hydraulc Pump Based on WPT and SOM Be Jkun, Lu Chen and Wang Zl PERFORMANCE ASSESSMENT AND FAULT DIAGNOSIS FOR HYDRAULIC PUMP BASED ON WPT AND SOM. Be Jkun,

More information

A Deflected Grid-based Algorithm for Clustering Analysis

A Deflected Grid-based Algorithm for Clustering Analysis A Deflected Grd-based Algorthm for Clusterng Analyss NANCY P. LIN, CHUNG-I CHANG, HAO-EN CHUEH, HUNG-JEN CHEN, WEI-HUA HAO Department of Computer Scence and Informaton Engneerng Tamkang Unversty 5 Yng-chuan

More information

Video Proxy System for a Large-scale VOD System (DINA)

Video Proxy System for a Large-scale VOD System (DINA) Vdeo Proxy System for a Large-scale VOD System (DINA) KWUN-CHUNG CHAN #, KWOK-WAI CHEUNG *# #Department of Informaton Engneerng *Centre of Innovaton and Technology The Chnese Unversty of Hong Kong SHATIN,

More information

Music Information Retrieval Schemes in Peer-to-Peer Environments

Music Information Retrieval Schemes in Peer-to-Peer Environments Journal of Computer Scence 1 (3): 369-375, 2005 ISSN 1549-3636 Scence Publcatons, 2005 Musc Informaton Retreval Schemes n Peer-to-Peer Envronments Chaokun Wang, Janzhong L and Shengfe Sh School of Computer

More information

Simulation Based Analysis of FAST TCP using OMNET++

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

More information

Performance Study of Parallel Programming on Cloud Computing Environments Using MapReduce

Performance Study of Parallel Programming on Cloud Computing Environments Using MapReduce Performance Study of Parallel Programmng on Cloud Computng Envronments Usng MapReduce Wen-Chung Shh, Shan-Shyong Tseng Department of Informaton Scence and Applcatons Asa Unversty Tachung, 41354, Tawan

More information

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

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

More information

The Research of Support Vector Machine in Agricultural Data Classification

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

More information

UB at GeoCLEF Department of Geography Abstract

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

More information

Motivation. EE 457 Unit 4. Throughput vs. Latency. Performance Depends on View Point?! Computer System Performance. An individual user wants to:

Motivation. EE 457 Unit 4. Throughput vs. Latency. Performance Depends on View Point?! Computer System Performance. An individual user wants to: 4.1 4.2 Motvaton EE 457 Unt 4 Computer System Performance An ndvdual user wants to: Mnmze sngle program executon tme A datacenter owner wants to: Maxmze number of Mnmze ( ) http://e-tellgentnternetmarketng.com/webste/frustrated-computer-user-2/

More information

An Influence of the Noise on the Imaging Algorithm in the Electrical Impedance Tomography *

An Influence of the Noise on the Imaging Algorithm in the Electrical Impedance Tomography * Open Journal of Bophyscs, 3, 3, 7- http://dx.do.org/.436/ojbphy.3.347 Publshed Onlne October 3 (http://www.scrp.org/journal/ojbphy) An Influence of the Nose on the Imagng Algorthm n the Electrcal Impedance

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

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

More information

Application of VCG in Replica Placement Strategy of Cloud Storage

Application of VCG in Replica Placement Strategy of Cloud Storage Internatonal Journal of Grd and Dstrbuted Computng, pp.27-40 http://dx.do.org/10.14257/jgdc.2016.9.4.03 Applcaton of VCG n Replca Placement Strategy of Cloud Storage Wang Hongxa Computer Department, Bejng

More information

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

More information

Private Information Retrieval (PIR)

Private Information Retrieval (PIR) 2 Levente Buttyán Problem formulaton Alce wants to obtan nformaton from a database, but she does not want the database to learn whch nformaton she wanted e.g., Alce s an nvestor queryng a stock-market

More information

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

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

More information

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

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

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

More information

A Resources Virtualization Approach Supporting Uniform Access to Heterogeneous Grid Resources 1

A Resources Virtualization Approach Supporting Uniform Access to Heterogeneous Grid Resources 1 A Resources Vrtualzaton Approach Supportng Unform Access to Heterogeneous Grd Resources 1 Cunhao Fang 1, Yaoxue Zhang 2, Song Cao 3 1 Tsnghua Natonal Labatory of Inforamaton Scence and Technology 2 Department

More information

Object-Based Techniques for Image Retrieval

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

More information

Efficient Content Distribution in Wireless P2P Networks

Efficient Content Distribution in Wireless P2P Networks Effcent Content Dstrbuton n Wreless P2P Networs Qong Sun, Vctor O. K. L, and Ka-Cheong Leung Department of Electrcal and Electronc Engneerng The Unversty of Hong Kong Pofulam Road, Hong Kong, Chna {oansun,

More information

A Robust Method for Estimating the Fundamental Matrix

A Robust Method for Estimating the Fundamental Matrix Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.

More information

Detection of an Object by using Principal Component Analysis

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

More information

Real-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems

Real-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems Real-tme Fault-tolerant Schedulng Algorthm for Dstrbuted Computng Systems Yun Lng, Y Ouyang College of Computer Scence and Informaton Engneerng Zheang Gongshang Unversty Postal code: 310018 P.R.CHINA {ylng,

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

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

More information

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints TPL-ware Dsplacement-drven Detaled Placement Refnement wth Colorng Constrants Tao Ln Iowa State Unversty tln@astate.edu Chrs Chu Iowa State Unversty cnchu@astate.edu BSTRCT To mnmze the effect of process

More information

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval Fuzzy -Means Intalzed by Fxed Threshold lusterng for Improvng Image Retreval NAWARA HANSIRI, SIRIPORN SUPRATID,HOM KIMPAN 3 Faculty of Informaton Technology Rangst Unversty Muang-Ake, Paholyotn Road, Patumtan,

More information

Improved Resource Allocation Algorithms for Practical Image Encoding in a Ubiquitous Computing Environment

Improved Resource Allocation Algorithms for Practical Image Encoding in a Ubiquitous Computing Environment JOURNAL OF COMPUTERS, VOL. 4, NO. 9, SEPTEMBER 2009 873 Improved Resource Allocaton Algorthms for Practcal Image Encodng n a Ubqutous Computng Envronment Manxong Dong, Long Zheng, Kaoru Ota, Song Guo School

More information

Maintaining temporal validity of real-time data on non-continuously executing resources

Maintaining temporal validity of real-time data on non-continuously executing resources Mantanng temporal valdty of real-tme data on non-contnuously executng resources Tan Ba, Hong Lu and Juan Yang Hunan Insttute of Scence and Technology, College of Computer Scence, 44, Yueyang, Chna Wuhan

More information

TF 2 P-growth: An Efficient Algorithm for Mining Frequent Patterns without any Thresholds

TF 2 P-growth: An Efficient Algorithm for Mining Frequent Patterns without any Thresholds TF 2 P-growth: An Effcent Algorthm for Mnng Frequent Patterns wthout any Thresholds Yu HIRATE, Ego IWAHASHI, and Hayato YAMANA Graduate School of Scence and Engneerng, Waseda Unversty {hrate, ego, yamana}@yama.nfo.waseda.ac.jp

More information

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on

More information

Association Rule Mining with Parallel Frequent Pattern Growth Algorithm on Hadoop

Association Rule Mining with Parallel Frequent Pattern Growth Algorithm on Hadoop Assocaton Rule Mnng wth Parallel Frequent Pattern Growth Algorthm on Hadoop Zhgang Wang 1,2, Guqong Luo 3,*,Yong Hu 1,2, ZhenZhen Wang 1 1 School of Software Engneerng Jnlng Insttute of Technology Nanng,

More information