Implementation of Vector Directional Distance Rational Hybrid Filter Using TMS320C6416

Size: px
Start display at page:

Download "Implementation of Vector Directional Distance Rational Hybrid Filter Using TMS320C6416"

Transcription

1 06 The Internatonal Arab Journal of Informaton Technology, Vol. 7, No., Aprl 00 Implementaton of Vector Drectonal Dstance Ratonal Hybrd Flter Usng TMS0C646 Ans Boudabous, Lazhar khrj, and Nour Masmoud Laboratory of Electroncs and Informaton Technology, Tunsa Department of Electrcal and Computer Engneerng, Sultan Qaboos Unversty, Oman Abstract: Ths paper proposes a novel dgtal sgnal processng mplementaton of vector drectonal dstance ratonal hybrd flter for mpulsve, Gaussan and mxed nose suppresson and fne-detals preservaton n color mages. The Implementaton was done, ntally, based on dgtal sgnal processng, whch proves the need to a large executon tme for flterng based on vector approach. Then, we propose effcent optmzatons and approxmatons of the man functons whch have no nfluence on the fltered mage qualty. Indeed, we could ncrease the flterng speed compared to the ntal vector drectonal dstance ratonal hybrd flter mplementaton. Comparatve studes of mplemented vector drectonal dstance ratonal hybrd flter wth other nonlnear flters are reported. It demonstrates a good color mage qualty for dfferent nose ntenstes. Ths effcent system mplementaton can perform well n several mage processng applcatons. Keywords: Color mage, DSP, optmzaton, nonlnear flter. Receved July, 008; accepted November 5, 008. Introducton Nose removal s an mportant task n mage processng. Many flters are often appled to gray value mages. The extenson of the concept of scalar flterng to color mage processng s not smple and straght forward task. Wth the extensve use of color n modern applcatons, the demand for hgh qualty mages s also ncreasng. Consequently, the nterest n color percepton and processng has been rsng rapdly and the area of color mage processng has been the subject of extensve research recently [9, 0,, ]. Flterng s one of the most mportant elements of color mage processng systems. Its most mportant applcatons are nose removal, mage enhancement, and mage restoraton. Nose may be ntroduced to an mage due to mperfect formaton or storage systems, nosy communcaton channels, and other varous factors, such as atmospherc turbulence [8]. However, color mage flterng dd not have the same development n the past that other areas of dgtal sgnal processng have enjoyed. Another mportant factor s the dffculty n understandng and modelng the human percepton of color [8] due to the complexty of the Human Vsual System (HVS). Therefore, lnear flters used n mage flterng applcatons, cannot cope wth the nonlnearty of the mage formaton model and transmsson channels and cannot take nto account the nonlnearty of human vson [8]. Flters havng good edge and mage detal preservaton propertes are hghly sutable for dgtal mage flterng. Most of the classcal lnear dgtal mage flters have low pass characterstcs. They tend to blur edges and to destroy lnes, edges, and other fne mage detals []. These reasons have led researchers to use nonlnear flterng technques. These technques were consdered n many applcatons n dgtal mage processng. Many classes of nonlnear dgtal mage flterng technques have appeared n the lterature such as order statstcs flters, polynomal flters, morphologcal flters, and neural networks [, 9]. Indeed, nonlnear vector flterng technques have generated much research nterest due to ther mportance n color mage restoraton [8, 0, 7, 0]. A varety of flterng technques proposed to date are based on multvarate order statstcs [], whch take the advantage of color nter-channel dependence and avod unpleasant drawbacks (pxel value rearrangng and chromatc shft) of component-wse flterng technques. The well known vector flters nclude the Vector Medan Flter (VMF) [, ], the Vector Drectonal Flter (VDF) [4], and the Drectonal Dstance Flter (DDF) [9]. Detecton based flterng technques were desgned to remove mpulsve nose [4, 0]. Such a structure, whle effcent for mpulsve nose removal, s napproprate n dealng wth Gaussan nose or mxed nose contamnaton. In comparson, a class of flters based on ratonal functons was desgned to cope wth dfferent types of nose. Among these flters we menton the Vector Medan Ratonal Hybrd Flter (VMRHF) and the Vector Drectonal Dstance Ratonal Hybrd Flter (VDDRHF) [7, 9]. They consttute very accurate estmators n long and short taled nose dstrbutons and, at the same tme, preserve the chromatcty of the color mage. Moreover, they act n small wndow and requre lttle

2 Implementaton of Vector Drectonal Dstance Ratonal Hybrd Flter Usng TMS0C number of operatons, resultng n smple and fast flter structures. The man goal of ths paper s to fnd a way on mplementng the VDDRHF for real tme applcaton purposes. To do so, we propose effcent optmzatons and approxmatons of the man functons whch have no nfluence on the fltered mage qualty yeldng to a much better and reduced consumng tme (up to 95% of the flterng speed) compared to the ntal VDDRHF s mplementaton. The paper s structured as follows. Secton presents an overvew of the nonlnear flter VDDRHF. A computer smulaton of ths flter s the object of secton. In secton 4, we provde the mplementaton by means of use of Dgtal Sgnal Processng (DSP) technology. Secton 5 descrbes the effcent optmzaton and approxmaton technques of the man functons of VDDRHF for mplementaton. Fnally, conclusons are drawn n secton 6. Image Lena Mandrll Butterfly Flower Table. Flterng results usng mpulsve nose %. Metrc Nosed VDF VMF VMRHF VDDRHF Bref Overvew of the Vector Nonlnear Flters.. Color Image Flterng There are two approaches to flter colour mages [4]. The frst approach breaks up the mage nto three components R, G and B and ensures the treatment ndependently. It s called margnal approach. The second s called vector approach as shown n Fgure, where colour mage pxels are consdered as - component vectors n the colour space that s more approprate for the human vsual system. Therefore, the nherent correlaton that exsts between the dfferent colour components s not gnored. decade n whch a lot of related work has been done. Ths secton descrbes the desgn of VDDRHF flter. The DSP mplementaton of some nonlnear flters lke VMF and VDF s publshed [6], where a floatng pont DSP technque (DSP C67) has been used. Table. Flterng results usng Gaussan nose ( σ =00). Image Lena Mandrll Butterfly Flower Metrc Nosed VDF VMF VMRHF VDDRHF In addton to ts detal preservng capablty, the VRF removes Gaussan nose and small magntude mpulsve nose [8]. The specfc of VDDRHF [9] s the combnaton between the drecton and the magntude process whch s sutable for human vsual system and can gve better balanced result between nose reducton and chromatcty retenton. The VDDRHF structure s made of two flterng stages as shown n Fgure. x (n) and Φ (n) are respectvely the nput (nosy pxel) and the output (fltered pxel) of the VDDF. y (n) s the fnal output (fltered pxel) of the VDDRHF gven by equaton. y ( n ) = Φ ( n ) + = h + k. D [ Φ β Φ ( n ) () ( n ), Φ ( n )] The term D[.] s an edge sensng term, expressed as, p p ( n), Φ( n)] = Φ Φ. θ( Φ, Φ) D[ Φ ().. VDDRHF Desgn Fgure. Vector approach. Nonlnear flterng technques were consdered as premature n dgtal mage processng. Research n ths area has been a dynamc development n the past Fgure. Block dagram of VDDRHF flter. To fnd Φ ( n), Φ ( n) and Φ ( n) we use the VDDF. They combne n the frst stage the L -norm crtera and angular dstance crtera to produce three output vectors n whch three vector drectonal dstance flter

3 08 The Internatonal Arab Journal of Informaton Technology, Vol. 7, No., Aprl 00 outputs to elmnate mpulsve nose, preserve edges and color chromatcty. In the second stage a vector ratonal operaton acts on the above three output vectors to produce the fnal output vector. Furthermore, as ponted out n [7, 9], a vector ratonal flter performs well for relatvely hgh SNR Gaussan contamnated envronments. When both mpulsve and Gaussan noses are present, nether the vector ratonal flter nor VDDF performs well. For ths reason, t was necessary to use a hybrd structure flter. The most popular nonlnear multchannel flters are based on the orderng of vectors n a predefned flter wndow. The output of these flters s defned as the lowest ranked vector accordng to a specfc orderng technque [4]. Let y (X ) represent a multchannel mage and let W (n) be a wndow of fnte sze N, where the nosy mage vectors nsde the wndow are denoted as X j (n), j=,,,n and the central sample n) = X ( ( n) determnes the poston of the flter wndow. X ( N+) / If the Eucldan dstance (L -norm) between two vectors X, X s denoted as, ( X X ) / m, j = X X j = X X j l= N M = j = dx X j j d, then the scalar quantty ( ), s the dstance assocated wth the nosy vector X n W. An orderng of the M s ( M ( ) M ( )... M ) mples the same orderng scheme to the nput set W (n) resultng n the ordered sequence X ( ) X ( )... X. If the angle between two vectors as X, X s denoted j T X X j θ( X, X ) = cos, then the scalar j X. X j N A X, X s the angular dstance quantty = = θ( ) j assocated wth the nosy vector X n W. An orderng of the A s ( A( ) A( )... A( N ) ) mples the same orderng scheme to the nput set W (n) resultng n the ordered sequence X ( ) X ( )... X. Combned magntude and drectonal dstances: the scalar quantty, p p γ M A =. () s the dstance assocated wth the nosy vector X n W. The power parameter p s a desgn parameter ranged from 0 to. It controls the mportance of the angle crteron versus the dstance crteron n the overall flter process. In the two extremes, p=0 or p=, the operator behaves as ether magntude processng or drectonal processng, respectvely. The case of p=0.5 s gvng equal mportance to both j crtera. For an ntermedate value of p (0<p<) both, magntude dstance (to process nose attenuaton) and angular dstance (to process chromatcty retenton) are consdered, whch ft better the concept of human vsual system [9]. An orderng of the γ s ( γ( ) γ( )... γ ) mples the same orderng scheme to the nput set W (n) resultng n the ordered sequence X ( ) X ( )... X. Nonlnear ranked type multchannel estmators defne the vector X ( ) W( n) assocated wth the mnmum aggregated dstance γ ( ) { γ, γ,..., γ N } as the flter output, whch s called a Vector Drectonal Dstance Flter (VDDF). Therefore, the VDDF conssts of computng and comparng the values of γ nsde the sldng flterng wndow W (n) and the output s the vector X k for whch γ k reaches t s mnmum.. Computer Smulatons In ths secton we compare the subjectve qualty, the objectve qualty of VDDRHF wth other well known algorthms. For ths purpose two metrcs, gvng a dstncton between two dgtal mages, have been used: the Peak Sgnal to Nose Rato () and the Structural SIMlarty Metrc () [0]. The s defned as: = 0 log 0 M max N = j = o, j M f N, j (4) where max s the range of allowable gray scale (55 for 8 bts). MxN s the number of vector samples, and f, j o, j are the ampltudes of the orgnal and fltered vectors. Some tradtonal metrcs lke Mean Absolute Error (MAE) and Mean Square Error (MSE) are proved to be nconsstent wth human eye percepton. Therefore, the use of metrc as a method for measurng the smlarty between two mages wll go perfectly wth the human vsual system. It s an mproved verson of the unversal mage qualty ndex []. The s defned as follow: µµ + + I J+ C σσ I J C σ IJ C = (5) µ + C + + C I µ J σi σj C σσ I J µ Iµ J s the weghted local lumnance, σ Iσ s the J weghted local contrast, σ s the weghted local correlaton (structure) between orgnal mage fltered mage j IJ o, and j f and C s a stablzng constant., The proposed algorthms were tested usng mages

4 Implementaton of Vector Drectonal Dstance Ratonal Hybrd Flter Usng TMS0C wth artfcally njected mpulse nose, Gaussan nose or mxed nose. The mplementaton of nose and dfferent flters s developed usng ANSI C Language. All of the orgnal mages were 4 bt, RGB color mages. A fxed probablty of an mpulse has been assumed for every byte n the mage. If a byte was replaced by an mpulse, the mpulse could take any value n the range [0, 55] wth unform probablty [4]. Tables, and show the comparatve results between the VDDRHF flter and others common flters lke VMF, VDF and VMRHF usng varous types of addtve nose. Table. Flterng results usng mxed nose (%, σ = 00 ). of on board devces that sut a wde varety of applcaton envronments. The code composer studo development tools are bundled wth the 646 DSK, ntegrated development envronment for C/C++ and assembly programmng [5]. In addton to a hgh clock rate, C64x DSPs can do more work each cycle wth bult n extensons. These extensons nclude new nstructons to accelerate performance n key applcaton areas such as dgtal communcatons nfrastructure, vdeo and mage processng. The DSP kernel s composed of two data paths: A and B, each, contanng four functonal unts (S, D, M, L, S, D, M and L) whch allow a maxmum of eght nstructons n one cycle. Fgure shows dfferent blocks of DSP archtecture: Image Metrc Nosed Vdf Vmf Vmrhf Vddrhf Lena Mandrll Butterfly Flower Theoretcally, VDDRHF flter seems more complcated than other flters. But we wll notce that t s more powerful n mage qualty. Ths smulatons show that the VDDRHF flter has better and than other algorthms. We noted that the mage qualty dffers accordng to the parameter p. we can test ths for varous mages. In order to fnd the best value of parameter p, we used an teratve optmzaton algorthm. The MSE s consdered as a cost functon. The parameter p changes slghtly by usng dfferent mages as shown by Table 4. Table 4. Optmum value of parameter p usng dfferent color mage. Image Lena Peppers Mandrll Flower Butterfly p Values 0,5 0,4 0,06 0,5 0, To smplfy the mplementaton, we choose an approprate value of p (p=0.5) that seems sutable for all tested mages. The choce s based on the desgn aspect of the power p that needs a smple shftng operaton, whch can be executed very fast by DSP. 4. DSP Board and Archtecture DSP offers technologes for nnovatve system development and applcaton desgn. The DSP s able to provde a hgh number of MIPS. The prmary features of the DSK are: 70 MHz, AIC stereo codec, four poston users DIP swtch and four user LEDs, flash and SDRAM. The DSK comes wth a full complment Fgure. Internal DSP archtecture The code development was done through seres of steps as follows: Step : we comple and profle natve C/C++ code (It valdates orgnal C/C++ code and determnes whch loops are the most mportant n terms of MIPS requrements). Step : we add const declaratons, loop, memory bank, and data algnment nformaton (It reduces potental ponter alasng problems, allows loops wth ndefnte teraton counts to execute eplogs, uses pragmas to pass count nformaton to the compler, uses memory bank pragmas and asserts ntrnsc to pass memory bank and algnment nformaton to the compler). Step : we optmze C code usng other C6000 ntrnsc and other methods (It facltates the use of some C6000-nstructons that are not easly represented n C, optmzes data flow bandwdth, uses word access for short data and double word access for word). Step 4: lnear assembly can be wrtten and parttonng nformaton can be added [5]. One Tmer has been used to measure the number of cycles. In fact, the most delcate parts are the SDRAM well managng, the cache memory and the fle manpulatons. In our present work, we are nterested by the valdaton of steps and and.

5 0 The Internatonal Arab Journal of Informaton Technology, Vol. 7, No., Aprl Implementaton of VDDRHF Flter The frst mplementaton of VDDRHF flter requres a y large number of cycles (due to the use of x functons and Arc cosne operatons). Therefore, our goal s to reduce CPU tme as much as possble usng dfferent approxmatons and optmzaton technques. One way s to use Taylor seres expanson of sn to substtute cos functon as follow: π π cos ( x) = sn ( x) x (6) Va smulaton, we can vary the order of the expanson of sn. Indeed, the calculated values are very close to zero, then, they have angles that are close toπ. We can clearly notce that when x s close to 0, the dfference between the theoretcal functon and ts approxmaton s neglgble (error s about.8x0-4 ). In addton, we can use the expanson of the power functon usng the followng equaton: x wth; a = e alogx ( alogx) + alogx+ ( alogx) +! + (7) x x ln( + x ) x + + (8)! By changng the varable to X = +x then x = X- and equaton 7 becomes: X a = e a log X ( X ) ( a. ( X ) ( X ) + a. ( X ) ( X ) + ) ( X ) (9) Ths approxmaton s used to calculate the VDDF dstances, but not to calculate the edge sensng term D[.] as n () snce t has nfluence on the and values, thus, the mage qualty. Moreover, the optmzaton s obtaned by usng technques gven by TI wth ts software CCS II, whch are ntrnsc and standard assembly nstructons [5, 5]. On one hand the ntrnsc technque of C646 has many advantages such as: the reducton of compled code and executon tme, more comprehensble code, varous nstructons, easy to ntegrate n the C code, syntax very close to the standard assembler and good documentaton on the Code composer level. On the other hand, the dsadvantage of the «ntrnsc» technque resdes n the fact that t does not replace all the assembly nstructons. The C64x provdes a number of nstructons whch combne common operatons together. These nstructons reduce the overall nstructon count n the code. Thereby, reducng code sze and ncreasng code densty. Also, they tend to smplfy programmng. Some used macro operatons are lsted n Table 5. Table 5. Descrpton of some ntrnsc nstructons. Intrnsc Instructons _MPY and _MPYU _DOTP _DOTPU4 Descrpton Multples the 6 LSBs of data by the 6 LSBs of data and returns the result. Values can be sgned or unsgned. Performs two 6x6 multples and adds the products together. Performs four 8x8 multples and adds products together. We can also explan the operatons as shown n Fgure 4: 6 a_h a_l 0 DOTP b_h b_l 0 = 0 5 a_h * b_h + a_l 0 * b_l 0 Fgure 4. DOTP Descrpton. Another ntrnsc nstructon _DOTPU4 as shown n Fgure 5 s used n our work when the nteger values are between 0 and 55. Usng _DOTPU4 nstructon, we can load, smultaneously, 4 pxels and calculate ther product and summaton n one cycle. Ths nstructon was used for dstance/norm calculatons. To extract the pxel values after flterng, we use the functon named _ROUNDF whch returns value rounded to the nearest nteger. 4 6 a a a a4 4 DOTPU4 6 b b b b4 a * b + a * b + a * b + a4 * b4 = Fgure 5. DOTPU4 Descrpton All these deas have no nfluence on the mage qualty, yeldng to a faster executon tme. Usng the kt C646 DSK, Table 6 shows the average CPU tme for each approxmaton usng Lena mage corrupted by % of mpulsve nose (sze: 76x44x). In addton, Table 6 shows the and values of the mplementaton results usng TMS0C646 DSK board wth a C646 DSP clocked at 70 MHz. We can conclude that the and of the mplemented flter are clearly dentcal for the three stages of optmzaton. 0

6 Implementaton of Vector Drectonal Dstance Ratonal Hybrd Flter Usng TMS0C646 Table 6. Average CPU tme for dfferent optmzaton steps of VDDRHF. VDDRHF Executon Tme (ms) Usng Orgnal Source Fle Usng cos Developmen t Usng y x Development Usng Intrnsc Instructons (a) (b) (c) (d) (e) (f) (g) (h) () (j) (k) (l) (m) (n) (p) Fgure 6. Orgnal (a, d, g, j, m), nosy (b, e, h, k, n) (mpulsve nose %) and fltered mages (c, f,, l, p) by optmzed VDDRHF flter usng C646 DSP. Therefore, wth these optmzatons, we ganed n the tme crteron whle keepng the same mage qualty. Dfferent test mages are dsplayed n Fgure 6. The orgnal color mages were 4-bt n RGB color space. These mages are contamnated by an mpulsve nose (mages (b), (e), (h), (k), (n)). The fltered mages by the mplemented flter are dsplayed also n Fgure 6 (mages (c), (f), (), (l), (p)). 6. Conclusons In ths paper, a novel DSP mplementaton of VDDRHF flter for mpulsve, Gaussan and mxed nose suppresson and fne-detal preservaton n color mage s presented. The desgned VDDRHF demonstrates a good color mage qualty for dfferent ntensty and varous noses compared to other well known nonlnear flters. Implemented n DSP (TMS0C646 70Mhz), the VDDRHF gves large executon tme. Therefore, the challengng task was n desgnng and developng approxmaton technques whch answer hgher processng speed and have no nfluence on the mage qualty. Subsequently, our approxmatons mprove the flterng speed compared to the ntal algorthm (0 tmes faster) and would be useful n some colour mage applcatons that do not need fast dsplay, lke medcal magng, tomography, etc., Future work wll be focused and challenged on reducng further the processng tme to deal more wth real tme applcatons usng combned HardWare/SoftWare (HW/SW) mplementaton n codesgn context. References [] Astola J. and Kuosmanen P., Fundamentals of Nonlnear Dgtal Flterng, CRC Press, 997. [] Astola J., Haavsto P., and Neuvo Y., Vector Medan Flter, n Proceedngs of IEEE, USA, pp , 990. [] Caselles V., Sapro G., and Chung D., Vector Medan Flters Inf Sup Operatons, and Coupled PDE s: Theoretcal Connectons, Computer Journal of Mathematcal Imagng and Vson, vol., no., pp. 09-9, 000. [4] Chekh F., Hamla R., Gabbouj M., and Astola J., Impulsve Nose Removal n Hghly Corrupted Color Images, n Proceedngs of Internatonal Conference on Image Processng, Swtzerland, pp , 996. [5] Chen J., Code Composer Studo Tutoral, Techncal Reference TMS0C646 DSK, 00. [6] Domínguez L. and Ponomaryov V., Non Lnear Flters for Colour Imagng Implemented by DSP, n Proceedngs of XI Conference on Electrcal Engneerng, Mexco, pp. 7-9, 005. [7] Khrj L. and Gabbouj M., Generalzed Class of Nonlnear Type Hybrd Flters, Computer Journal of IEE Electroncs Letters, vol. 8, no. 5, pp , 00. [8] Khrj L. and Gabbouj M., Vector Medan Ratonal Hybrd Flters for Multchannel Image Processng, n Proceedngs of ISCAS, New York, pp. 90-9, 999. [9] Khrj L., Vector Drectonal Dstance Ratonal Hybrd Flters for Color Image Restoraton,

7 The Internatonal Arab Journal of Informaton Technology, Vol. 7, No., Aprl 00 Computer Journal of Engneerng Research, vol., no., pp. -, 005. [0] Km J. and Wlls S., Fast Vector Medan Flter Implementaton Usng the Color Pack Instructon Set, n Proceedngs of Dgtal Sgnal Processng Workshop of IEEE, Germany, pp. 9-4, 00. [] Koschan A. and Abd M., A Comparson of Medan Flter Technques for Nose Removal n Color Images, n Proceedngs of 7 th Workshop on Color Image Processng, Germany, pp , 00. [] Lukac B., Applcaton of the Adaptve Center Weghted Vector Medan Framework for the Enhancement of Cdna Mcroarray Images, Internatonal Computer Journal of Applcaton Mathematc Computng Scence, vol., no., pp. 69-8, 00. [] Lukac R., Platanots N., Smolka B., and Venetsanopoulos A., Weghted Vector Medan Optmzaton, n Proceedngs of 4 th EURASIP Conference Vdeo-Image and Multmeda Communcatons, USA, pp. 7-, 00. [4] Lukac R., Smolka B., Martn K., Platanots N., and Venetsanopoulos N., Vector Flterng for Color Imagng, Computer Journal of IEEE Sgnal Processng Magazne, vol. 5, no., pp , 005. [5] Pedra M. and Frtsh A., Texas Instrument Code Composer Studo Tutoral Spru89f, Techncal Reference TMS0C6000, 000. [6] Ptas I. and Venetsanopoulos A., Order Statstcs n Dgtal Image Processng n Proceedngs of the IEEE, Canada, pp. 89-9, 99. [7] Ptas I. and Venetsanopoulos A., Nonlnear Dgtal Flters: Prncples and Applcatons, Klumer Academc, New York, 990. [8] Trahanas P., Ptas I., and Venetsanopoulos N., Color Image Processng, Academc Press, 994. [9] Vnayagarnoorthy S., Order Statstcs Flterng of Colour Images: A Perceptual Approach, Master Thess, Unversty of Toronto, 997. [0] Wang S., L Y., Chung F., and Xu M., An Iteratve Self Adaptve Algorthm to Impulse Nose Flterng for Color Images, Internatonal Computer Journal of Informaton Technology, vol., no. 0, pp. 5-59, 005. [] Wang Z., Bovk A., Shekh H., and Smoncell P., Image Qualty Assessment: From Error Vsblty to Structural Smlarty, Computer Journal of IEEE Transactons on Image Processng, vol., no. 4, pp , 004. Ans Boudabous receved BEng degree n electrcal engneerng from the Natonal Engneerng School of Sfax n 00, and hs MS degree n electroncs from Unversty of Sfax n 004, Actually, he s lecturer at the Unversty of Sfax, Tunsa and he s persuadng hs PhD studes wth the Laboratory of Electroncs and Informaton Technology. Lazhar khrj receved hs BS degree n electroncs, and hs MS and PhD degrees n electrcal engneerng from Unversty of Tuns II, Tunsa, n 990, 99 and 999, respectvely. In 00, he receved the Doctor of technology degree n nformaton technology from Sgnal Processng Insttute, Tampere Unversty of Technology, Fnland. Nour Masmoud receved electrcal engneerng degree from the Faculty of Scences and Technques- Sfax, Tunsa, n 98, the DEA degree from the Natonal Insttute of Appled Scences-Lyon and Unversty Claude Bernard- Lyon, France n 984. From 986 to 990, he prepared hs thess at the Laboratory of Power Electroncs at the Natonal School Engneerng of Sfax. He receved hs PhD degree from the Natonal School Engneerng of Tuns, Tunsa n 990.

8

9 06 The Internatonal Arab Journal of Informaton Technology, Vol. 7, No., Aprl 00

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

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

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

More information

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

Edge Detection in Noisy Images Using the Support Vector Machines

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

More information

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 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

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

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

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

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

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

More information

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

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

More information

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

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

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

More information

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

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

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 Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 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 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

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

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

TN348: Openlab Module - Colocalization

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

More information

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

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

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

More information

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

Symmetrical recursive median filter for region smoothing without edge distortion

Symmetrical recursive median filter for region smoothing without edge distortion Int'l Conf. IP, Comp. Vson, and Pattern Recognton IPCV'16 171 Symmetrcal recursve medan flter for regon smoothng wthout edge dstorton A. Raj Laboratory of Images, Sgnals and Intellgent Systems Pars Est

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

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

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

More information

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

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

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

More information

Hybrid Non-Blind Color Image Watermarking

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

More information

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

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

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

More information

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

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

Efficient Architecture and Implementation of Vector Median Filter in Co-Design Context

Efficient Architecture and Implementation of Vector Median Filter in Co-Design Context RADIOEGIEERIG, VOL. 16, O. 3, SEPTEMBER 007 113 Effcent Archtecture and Implementaton of Vector Medan Flter n Co-Desgn Context Ans BOUDABOUS 1, Lazhar KHRIJI, A. BE ATITALLAH 1,3, P. KADIOIK 3, our MASMOUDI

More information

S1 Note. Basis functions.

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

More information

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

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

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

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples

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

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson

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

User Authentication Based On Behavioral Mouse Dynamics Biometrics

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

More information

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

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

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

RADIX-10 PARALLEL DECIMAL MULTIPLIER

RADIX-10 PARALLEL DECIMAL MULTIPLIER RADIX-10 PARALLEL DECIMAL MULTIPLIER 1 MRUNALINI E. INGLE & 2 TEJASWINI PANSE 1&2 Electroncs Engneerng, Yeshwantrao Chavan College of Engneerng, Nagpur, Inda E-mal : mrunalngle@gmal.com, tejaswn.deshmukh@gmal.com

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

Wishing you all a Total Quality New Year!

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

More information

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

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

Lecture 5: Multilayer Perceptrons

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

More information

Smoothing Spline ANOVA for variable screening

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

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Overvew 2 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Introducton Mult- Smulator MASIM Theoretcal Work and Smulaton Results Concluson Jay Wagenpfel, Adran Trachte Motvaton and Tasks Basc Setup

More information

Enhanced AMBTC for Image Compression using Block Classification and Interpolation

Enhanced AMBTC for Image Compression using Block Classification and Interpolation Internatonal Journal of Computer Applcatons (0975 8887) Volume 5 No.0, August 0 Enhanced AMBTC for Image Compresson usng Block Classfcaton and Interpolaton S. Vmala Dept. of Comp. Scence Mother Teresa

More information

THE PULL-PUSH ALGORITHM REVISITED

THE PULL-PUSH ALGORITHM REVISITED THE PULL-PUSH ALGORITHM REVISITED Improvements, Computaton of Pont Denstes, and GPU Implementaton Martn Kraus Computer Graphcs & Vsualzaton Group, Technsche Unverstät München, Boltzmannstraße 3, 85748

More information

Specifications in 2001

Specifications in 2001 Specfcatons n 200 MISTY (updated : May 3, 2002) September 27, 200 Mtsubsh Electrc Corporaton Block Cpher Algorthm MISTY Ths document shows a complete descrpton of encrypton algorthm MISTY, whch are secret-key

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

Feature Reduction and Selection

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

More information

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

Parallel Inverse Halftoning by Look-Up Table (LUT) Partitioning

Parallel Inverse Halftoning by Look-Up Table (LUT) Partitioning Parallel Inverse Halftonng by Look-Up Table (LUT) Parttonng Umar F. Sddq and Sadq M. Sat umar@ccse.kfupm.edu.sa, sadq@kfupm.edu.sa KFUPM Box: Department of Computer Engneerng, Kng Fahd Unversty of Petroleum

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

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

More information

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

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data Malaysan Journal of Mathematcal Scences 11(S) Aprl : 35 46 (2017) Specal Issue: The 2nd Internatonal Conference and Workshop on Mathematcal Analyss (ICWOMA 2016) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES

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

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

Virtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory

Virtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory Background EECS. Operatng System Fundamentals No. Vrtual Memory Prof. Hu Jang Department of Electrcal Engneerng and Computer Scence, York Unversty Memory-management methods normally requres the entre process

More information

Enhanced Watermarking Technique for Color Images using Visual Cryptography

Enhanced Watermarking Technique for Color Images using Visual Cryptography Informaton Assurance and Securty Letters 1 (2010) 024-028 Enhanced Watermarkng Technque for Color Images usng Vsual Cryptography Enas F. Al rawashdeh 1, Rawan I.Zaghloul 2 1 Balqa Appled Unversty, MIS

More information

Dynamic wetting property investigation of AFM tips in micro/nanoscale

Dynamic wetting property investigation of AFM tips in micro/nanoscale Dynamc wettng property nvestgaton of AFM tps n mcro/nanoscale The wettng propertes of AFM probe tps are of concern n AFM tp related force measurement, fabrcaton, and manpulaton technques, such as dp-pen

More information

Review of approximation techniques

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

More information

COMPLEX WAVELET TRANSFORM-BASED COLOR INDEXING FOR CONTENT-BASED IMAGE RETRIEVAL

COMPLEX WAVELET TRANSFORM-BASED COLOR INDEXING FOR CONTENT-BASED IMAGE RETRIEVAL COMPLEX WAVELET TRANSFORM-BASED COLOR INDEXING FOR CONTENT-BASED IMAGE RETRIEVAL Nader Safavan and Shohreh Kasae Department of Computer Engneerng Sharf Unversty of Technology Tehran, Iran skasae@sharf.edu

More information

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

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

More information

A 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

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

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

REMOTE SENSING REQUIREMENTS DEVELOPMENT: A SIMULATION-BASED APPROACH

REMOTE SENSING REQUIREMENTS DEVELOPMENT: A SIMULATION-BASED APPROACH REMOTE SENSING REQUIREMENTS DEVEOPMENT: A SIMUATION-BASED APPROAC V. Zanon a, B. Davs a, R. Ryan b, G. Gasser c, S. Blonsk b a Earth Scence Applcatons Drectorate, Natonal Aeronautcs and Space Admnstraton,

More information

Active Contours/Snakes

Active Contours/Snakes Actve Contours/Snakes Erkut Erdem Acknowledgement: The sldes are adapted from the sldes prepared by K. Grauman of Unversty of Texas at Austn Fttng: Edges vs. boundares Edges useful sgnal to ndcate occludng

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

Problem Set 3 Solutions

Problem Set 3 Solutions Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,

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

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

Newton-Raphson division module via truncated multipliers

Newton-Raphson division module via truncated multipliers Newton-Raphson dvson module va truncated multplers Alexandar Tzakov Department of Electrcal and Computer Engneerng Illnos Insttute of Technology Chcago,IL 60616, USA Abstract Reducton n area and power

More information

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

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

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

An Entropy-Based Approach to Integrated Information Needs Assessment Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology

More information

Fitting: Deformable contours April 26 th, 2018

Fitting: Deformable contours April 26 th, 2018 4/6/08 Fttng: Deformable contours Aprl 6 th, 08 Yong Jae Lee UC Davs Recap so far: Groupng and Fttng Goal: move from array of pxel values (or flter outputs) to a collecton of regons, objects, and shapes.

More information

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

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

More information

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

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Unsupervised Learning

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

More information

Video Denoising Algorithm in Sliding 3D DCT domain

Video Denoising Algorithm in Sliding 3D DCT domain Dmytro Rusanovskyy and Karen Egazaran, ACIVS 2005, Antwerp, Belgum Vdeo Denosng Algorthm n Sldng 3D DCT doman Dmytro Rusanovskyy and Karen Egazaran Insttute of Sgnal Processng Tampere Unversty of Technology,

More information

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

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

More information

CMPS 10 Introduction to Computer Science Lecture Notes

CMPS 10 Introduction to Computer Science Lecture Notes CPS 0 Introducton to Computer Scence Lecture Notes Chapter : Algorthm Desgn How should we present algorthms? Natural languages lke Englsh, Spansh, or French whch are rch n nterpretaton and meanng are not

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

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

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION

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

More information

ANALYSIS OF ADAPTIF LOCAL REGION IMPLEMENTATION ON LOCAL THRESHOLDING METHOD

ANALYSIS OF ADAPTIF LOCAL REGION IMPLEMENTATION ON LOCAL THRESHOLDING METHOD Nusantara Journal of Computers and ts Applcatons ANALYSIS F ADAPTIF LCAL REGIN IMPLEMENTATIN N LCAL THRESHLDING METHD I Gust Agung Socrates Ad Guna 1), Hendra Maulana 2), Agus Zanal Arfn 3) and Dn Adn

More information

Computational ghost imaging using a fieldprogrammable

Computational ghost imaging using a fieldprogrammable Computatonal ghost magng usng a feldprogrammable gate array IKUO HOSHI, * TOMOYOSHI SHIMOBABA, TAKASHI KAKUE, AND TOMOYOSHI ITO 1 Graduate School of Engneerng, Chba Unversty, 1-33, Yayo-cho, Inage-ku,

More information

Module Management Tool in Software Development Organizations

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

More information