Background Removal in Image indexing and Retrieval

Size: px
Start display at page:

Download "Background Removal in Image indexing and Retrieval"

Transcription

1 Background Removal n Image ndexng and Retreval Y Lu and Hong Guo Department of Electrcal and Computer Engneerng The Unversty of Mchgan-Dearborn Dearborn Mchgan , U.S.A. Voce: , Fax: ylu@umch.edu Abstract Ths paper presents our research n mage content based ndexng and retreval, a key technology n dgtal mage lbrares. In most of the exstng mage content-based technques, mage features used for ndexng and retreval are global, features are computed over the entre mage. The major problem wth the global mage feature based retreval methods s that background features can be easly mstakenly taken as object features. When a user attempts to retreve mages usng color features, he/she usually means the color feature of an object or objects of nterests contaned n the mage. The approach we descrbe n ths paper utlzes color clusters for mage background analyss. Once the background regons are dentfed, they are removed from the mage ndexng procedure; therefore, no longer nterferng wth the meanngful mage content durng the retreval process. The algorthm conssts of three major steps of computaton, fuzzy clusterng, color mage segmentaton, and background analyss. 1. Introducton Dgtal magery s a convenent medum for descrbng and storng spatal, temporal, spectral and physcal components of nformaton contaned n a varety of domans. Large databases contan thousands of dgtal mages that can occupy ggabytes of space, and hence t s dffcult for a user to fnd useful nformaton usng smple browsng technques. Therefore, effcent and automatc algorthms are requred for ndexng and retrevng mages from databases. Conventonally mages are ndexed usng text nformaton, such as keywords, date, artst, etc. [Gro88, ag85, FJL97]. However, the nadequacy of textual descrptons s obvous because the complexty of the nformaton mbedded n mages (number and types of objects, ther attrbutes and spatal relatonshps) can not be syntheszed n a few key words. Untl recently wth the fast growng technology n computer vson, mage content based ndexng and retreval are the new areas of research and development [Ja93, PPS94, FS95, Gup6, FDW96]. There s a growng demand for systems usng pctoral nformaton for both commercal and cultural applcatons. Research n content-based mage retreval can be characterzed n the followng categores: query-by-vsual sample, pctoral queres, and/or lngustc queres. In a query by vsual sample, a user must submt a sample mage to the system, and then the system searches ts database to fnd the mages most smlar to the sample mage. The smlarty measure s usually taken from mage features such as color, shape and texture. The IBM QBIC system [FS95] and the Vrage mage search engne both use ths type of query[gup96]. Pctoral queres are sketch-based queres, and these systems usually provde sketch tools to allow the user to draw mage features, such as shape, sze, color, and texture that occur n the mages to be retreved. The system uses the sketch as a query and search for mages that contan the sketch. VsualSEEK, a content-based mage/vdeo retreval system at Columba Unversty [SmC96], uses pctoral queres extensvely. Belonge et al[bcg98] used blobworld to represent the composton of mage content. The blobworld descrptors uses -D ellpses, or blobs, each of whch possesses a number of attrbutes. Lngustc queres that drectly refer to mage contents are popular among many users. These queres clearly dentfy salent vsual propertes contaned n the desred mages usng the languages are understandable by most 1

2 populaton. These mage features can be low level, such as red, or hgher level such as horzon or horse. The Dgtal Lbrary Intatve at the Unversty of Calforna at Berkeley has made a sgnfcant effort n ths drecton[wl96]. In most of the exstng mage content-based technques, mage features used for ndexng and retreval are global features, features are computed over the entre mage. For example, retrevng smlar mages n color, the most of the exstng technques use a color hstogram generated from the entre mage[hse95, JAV96]. The major problem wth the global color hstogram based retreval methods s that they do not possess spatal relatonshp between object regons and t does not dfferentate the color between the background and foreground. When a user attempts to retreve mages usng color features, he/she usually means the color feature of an object or objects of nterests contaned n the mage. For example, f a desgn engneer wants to retreve the mages of automobles usng color feature, he would lkely submt queres such as red cars or blue seat covers, etc. Therefore n many applcatons, t makes more sense to ndex and retreve mages based on the color nformaton of objects. If a retreval system only uses color as the feature for mage retreval, a user may get many red cars wth black background for a gven query black car. The dfferentaton of background from foreground s an ll-posed problem n mage analyss. For moton pctures, the background features n the consecutve mages do not change much, therefore we can use the dfference mage of two consecutve mages n the tme doman to dfferentate background from foreground. In stll mages, mage background s not well defned. Generally speakng, the background of an mage ncludes all the features n the mage that do not belong to the objects of nterests. In many mages, background can be easly recognzed by a human beng but dffcult for computer, because the human bengs have excellent object recognton and analyss abltes. In many mages mage background features are even ambguous to human observers. Our approach to ths problem s to elmnate regons that are certan to be backgrounds. For those mages that are ambguous to human percepton, we wll not nterfere. We do not ntend to use object recognton n mage retreval, whch s a very complcated classc computer vson problem. Instead, we are nvestgatng a computatonally effcency and yet effectve background removal algorthm that can facltate mage retreval. The approach we descrbe n ths paper utlzes color clusters for mage background analyss. Once the background regons are dentfed, they are removed from the mage ndexng procedure; therefore, they are no longer nterferng wth the meanngful mage content durng the retreval process. We have mplemented the algorthm at our web ste, and tested the system on mages from three dfferent collectons of mages, car mages, Earth and space mages, and crashed car mages. We have compared the algorthm wth a global color hstogram based approach and the results show that the algorthm wth background removal s very effectve n retrevng mages contanng a few major objects. Due to the lack of color prntng n ths publcaton, we have reduced number of mage examples and converted all color mages to gray scales. A color verson of the paper can be found at webste:[guo: put the address here].. Background regon analyss Fgure 1 gves an over vew of our system that performs background regon analyss. The system has three major computatonal components: fuzzy clusterng algorthm, color mage segmentaton, and background regon analyss. The fuzzy clusterng algorthm attempts to fnd smlar colors n the Luv space usng fuzzy logc. The algorthm consders a cluster of smlar colors as a fuzzy set, and represents the lkelness of a color pxel belongng to a fuzzy set by a fuzzy membershp a color mage Fuzzy clusterng n color hstogram doman CL 1 Mappng clusters CL to mage doman Color mage segmentaton CL 3 : a set of color regons Background regon analyss [BR 1,, BR k ] Fgure 1. Background feature of a color mage analyss

3 functon. The fuzzy clusterng algorthm yelds a decomposton of the 3D-color hstogram; namely a number of non-overlappng heaps n the 3D-color hstogram, CL 1. The labelng of the clusters n CL 1 results n a multthresholded mage. The detals of ths algorthm can be found n [ChL98]. In ths paper, we wll only dscuss the color mage segmentaton and the background analyss algorthm, and how the system can be used n mage content based ndexng and retreval..1 Color regon segmentaton The fuzzy clusterng algorthm mentoned above generates a set of clusters, CL 1, each of the clusters n CL 1 represents a sub-spectrum of colors n the Luv space. The mappng of clusters n CL 1 to the mage doman takes two steps. Frst we compute the central color of each cluster n CL 1 usng the formula below. Assume M clusters n CL 1 and each mage pxel belongs to one of the clusters. C( p )K denotes belongs to the Kth cluster, and ts Luv values are [l, u, v ]. The central color of cluster K s represented by ( K ) ( K ) ( K ) [ l, u, v ], whch s calculated as follows: l l ( K ) u v ( K ) ( K ) C( p ) K # of pxels n cluster K C( p ) K #of pxels n cluster K v #of pxels n cluster K u C( p ) K Each pxel n the mage doman s assgned the representatve color value of ts cluster. We obtan the second set of clusters, CL, by computng the connected pxels of the same cluster color. In general, one cluster n CL 1 may generate more than one clusters n CL, therefore, CL s usually much larger than CL 1. For example, n one of our experments, the algorthm generated, from a CL 1 wth CL1 64, the CL that contans more than fve hundred clusters. The clusters contaned n CL have the followng propertes: Pxels n the same cluster are spatally connected and are assgned of the same color: the representatve color of the cluster. Therefore, we sometmes refer the clusters n CL as color regons. The regons n CL are usually small and p p a meanngful mage regon contans several clusters n CL. The next step n the mage segmentaton s an agglomeratve process. It attempts to merge clusters n CL based on the color dstances among neghborng clusters n the spatal doman, cluster szes and the maxmum number of clusters n CL 3. Durng the agglomeratve process, the dstance between two clusters s defned as the color dstance between the representatve colors of the two clusters, and the color dstance functons we use are defned n the Luv space. For two dark colors,.e. the colors whose L, u and v components all have very small values, then color dstance s dependent on the L component only,.e., dst([ L1, u1, v1 ],[ L, u, v]) L1 L. For other colors, we use the Eucldean dstance. It s mportant to use a dfferent dstance functon for dark colors. As we found out emprcally that human percepton does not match the Eucldean dstance to dfferentate dark colors. For example n Fgure, regon A and regon B have very close colors accordng to human percepton and, therefore, they should be regarded as belongng to the same regon. However, we found that the Eucldean dstance between the two regons s large. For example, one typcal Luv value of a pxel n regon A s (39.45,11.96,-7.3), and one typcal Luv value of a pxel n regon B s (30.87,7., -4.78), and hhe Eucldean dstance of the two color pxels s dst 8.54, whch s qute large. However, the dark color dstance value s dst 8.58, whch s more consstent wth human percepton. If we used the Eucldean dstance functon n the mage segmentaton, the background regon n ths mage s very lkely to be dvded nto several smaller regons. Ths example suggests that for dark colors, the saturaton components are nsgnfcant. B A 3

4 Fgure. Illustraton of dark color dstance functon. Regon A and B look smlar from human percepton. Whle ther Luv dstance s large. We have explored the followng four dfferent methods n mergng clusters. In the frst three methods, a common parameter, max_cls s used to control the maxmum number of clusters n CL3. Method 1 consders the clusters that are adjacent and have smlar colors as the frst canddates for mergng. It uses a control parameter, cl_dff_th to denote color dfference threshold. It frst attempts to merge the neghborng clusters whose color dstances are below cl_dff_th. The order of mergng s rrelevant. If the number of clusters at the end of the mergng process s greater than max_cls, then the algorthm selects the smallest cluster and merges the cluster wth one of ts neghbors to whch t has the smallest color dstance. Ths mergng process repeats untl the number of clusters n CL3 s no more than max_cls. The second method consders the sze of clusters as the only mergng crteron. It selects the smallest cluster and merges the cluster wth one of ts neghbors to whch t has the smallest color dstance. The process s repeated untl the number of clusters n CL3 s no more than max_cls. The thrd method consders the color dstance as the most mportant crteron n spatal clusterng. However, the computaton n fndng the mnmum color dstance between two adjacent clusters s qute tme consumng f a large number of clusters exst. To allevate the computatonal burden, we frst repeatedly merges the smallest clusters wth ther neghbors n the closest color dstance untl the total number of clusters s reasonable, e.g In the second pass, we select the two adjacent clusters whose color dstance s the smallest wthn the entre mage and merges the two clusters. Ths process repeats untl the top max_cls clusters n sze contan a large percentage of the mage pxels (e.g. 99%). Snce the largest max_cl clusters already cover the major percent of the entre mage pxels, the small clusters below the top max_cl do not affect much the fnal segmentaton result. Therefore at the thrd pass, the algorthm repeatedly merges the smallest cluster wth ts closest neghbor n color dstance untl the total number of clusters n CL 3 s no more than max_cls. The last method we explored used two dfferent thresholds for color dstance functons. For clusters that have smaller sze, they are merged f ther color dstance s less than the larger threshold, and for larger clusters, they are merged f ther color dstance s less than the smaller threshold. The threshold for small clusters s determned by the mage sze. In all these methods, the center of the new regon s calculated after every merge, the sze of the regon has changed, and so do the neghbors. The representatve color of a new regon merged from C 1 and C s calculated usng the followng formula: L1 1 + L L + 1 u1 1 + u u + 1 v1 1 + v v + 1 where ( L 1, u1, v1 ) and ( L, u, v ) are the representatve colors of C 1 and C respectvely. ote the order of mergng may affect the clusterng result n the mage doman. The results from the spatal clusterng s a set of color regons, CL 3, n whch all the pxels n the same regon n CL 3 are spatally connected and have the same representatve color. From computatonal pont of vew, Method 1 and are more effcent than Method 3. Method 4 s the more effcent n terms of computaton than method 3 and more effectve than Method 1 and. However, n many cases, Method 3 generates better results than the others do. In the trade off of computatonal tme and segmentaton results, we chose to use method 4 n our background analyss algorthm. Fgure 4 shows an example of mage segmentaton results generated by Method 4. Fgure 4 (a) shows the orgnal mage, (b) shows the mage of clusters generated by mappng the clusters generated by the fuzzy clusterng algorthm drectly to the mage doman, and (c) shows the output from the segmentaton algorthm. For the purpose of demonstraton, the mage shown n (b) s drawn n pseudo colors. The colors shown n (c), the resultng mage, are true representatve colors of color regons. It can be seen, the sgnfcant color features are well captured after color mage segmentaton.. Image background analyss The background analyss algorthm s developed based on the followng hypotheses: 4

5 (a) (b) (c) Fgure 3. Example 1 of mage segmentaton. (a)the orgnal mage. (b) The mage after the color clusterng. There are 1 color clusters and 598 spatal clusters. (c) The segmentaton result. A background regon must be moderate n sze and must be connected to the mage borders. The man purpose of ths algorthm s to dentfy bg color regons that are not belongng to objects of nterests, snce only large background regons can hamper the retrevng results. Regons occupy mage corners are more lkely to be background. For a gven mage collecton, we may have specfed background colors. However, snce a background color can also appear n objects of nterests n the mages of the same collecton, and not all the mages have the same background colors, the pror knowledge of background colors should be used cautously. Assume there are M color regons n the mage, denoted by R 1, R, R M. Every regon s descrbed by { C, x 1, y 1, x,y }, where C s the representatve color of the th regon and (x 1, y 1 ), (x,y ) are the coordnates of the top-left and bottom-rght corners of the boundng box of the regon. We frst dvde the mage plane nto nne blocks shown n Fgure Fgure 4. An mage plane s dvded nto nne blocks for background analyss. The szes of the nne blocks are determned by the mage sze. In general we want the center block to be large enough to contan the mage contents. Based on our experence, the followng szes of the nne blocks are feasble. Let the wdth and heght of an mage be W and H respectvely. The center block 4 s 0.7W0.7H, the four corner blocks, 0,, 6 and 8, have sze of H0.15W, the four lateral blocks, 1 and 7, have sze of 0.7W0.15H, 3 and 5 of 0.15Hx0.7W. For each regon R j, we calculate ts percentage r j over each mage block, (0, 1,.8). From the percentage, we can obtan the number of corners and lateral regons one cluster occupes. Let c and l be the number of corners and the number of lateral regons a cluster occupes respectvely. We set these two varables as follows: c the number of corner regons whose occupancy rato s greater than a gven threshold θ. l the number of lateral regons whose occupancy rato s greater than a gven threshold. The algorthm can be summarzed nto the followng rules. Rule1. For a regon R j, f ts representatve color C j belongs to the specfed background color and the regon occupes more than two corners,.e. c >, then R j s marked as a background regon. Rule. For a regon R j, f ts representatve color C j belongs to the specfed background color and the regon occupes more than one corners and one lateral blocks,.e. ( c >0 and l >0), then R j s marked as a background regon. These two rules apply to an mage collecton wth specfed background colors. In certan applcatons ths specfcaton s possble. For examples, mages taken ndoors often have specfc background colors; n our Earth and Space Scence mage collecton, we found that black and whte often appear as background colors. For those mage collectons have no specfed background colors, we use the followng rules: Rule3. For an mage regon R j, f t occupes more than two corner blocks,.e. c >,, and ts percentage over the center block s small, then set R j as a background regon. Rule4. For an mage regon R j, f t occupes one or more corner and lateral blocks, and ts percentage over the center block s small,.e. r j4 < θ b, where θ b s a threshold, then set R j as a background regon. a 5

6 Rule5. For an mage regon R j, f t occupes at least one corner mage s a color photo of the Pacfc Ocean wth blue or one lateral regons,.e., c >0, or l >0, and most of R j s pxels are n the lateral and corner blocks, then set R j as a background regon. In many mages, background regons occupy the corner and lateral blocks, and therefore, the corner and lateral regons are emphaszed n the background regon analyss. Addtonally only large clusters can be recognzed as background, whch s controlled by θ n the computaton of c and l. For example, f we set θ to 0.85, then a mage regon s consdered as occupyng a corner only f one of r j0, r j, r j6, r j8 s above 85%. In Rule 3 and 4, n addton to the occupancy measure of corners and laterals, we also check occupaton percentage of the mage regon nsde the central block. If ths percentage s small, t mples that the center block contans some other larger regons and the current cluster may be regarded as background. Wthout ths addtonal condton, these rules can be detrmental to the mages that contan meanngful regons that occupy the central block as well as the corners and/or lateral blocks. For example, Fgure 5 shows an example. The orgnal a a color. The large regon connectng the corners and a good porton of the central block s the Pacfc Ocean, whch s the mportant content to the mage. As the result of our background analyss algorthm, ths mage s consdered to have no background regons. Fgure 6 shows an example of the background analyss algorthm. Fgure 6(b) shows the mage of clusters generated by the fuzzy clusterng algorthms n the color space. Fgure 6(c) shows the color regons after the segmentaton algorthm. For the purpose of llustraton we used pseudo colors to dsplay dfferent regons. The whte grds n (b) and (c) dvde the mage nto nne blocks. The labels n (c) ndcate the background regons found by the background analyss algorthm. Fgure 6 (d) shows the representatve colors of regons generated by the segmentaton algorthm. After the background analyss, we dentfed n ths mage three background regons, whch are labeled as 1, and 3 n (c). In ths example, although cluster 1, and 3 are not connected wth each other n the spatal doman, they are all consdered as background because of the same representatve background color, black. 6

7 (a) orgnal mage (b) Resultng mage from the Clusterng algorthm Fgure 5. An mage has no background regons. 1 (a) (b) (c) (d) 3 Fgure 6. Example of background analyss. (a) The orgnal mage. (b) The mage after of clusterng n pseudo colors. (c) The color regons after segmentaton n pseudo colors and 3 dentfed background regons. (d) The segmentaton result llustrated n the representatve colors. 1 (a) (b) (c) (d) Fgure 6. Example of background analyss. (a) Orgnal mage. (b) The mage of color clusters n pseudo colors. (c) The color regons after segmentaton n pseudo colors and 1 dentfed background regon. (d) Segmentaton results shown n representatve colors. [Guo: choose one of the two fgures based on the qualty of the converted gray scale mages.] 7

8 Fgure 6 shows another example of mage segmentaton. In ths example, the algorthm found one bg background regon(see (c)) whch has representatve color whte (see (d)). 3. Image ndexng and retreval usng background analyss We have mplemented the background analyss algorthm n IMAGE-SEEKER, a content-based mage retreval system runnng on the web ste IMAGE- SEEKER supports ffteen dfferent colors that are commonly used n mage queres. To mplement these queres, we generate the color hstogram usng ffteen domnant colors for each mage and ndex the mage wth the color hstogram. Durng the mage retreval stage, the system takes a color query and looks for mages whose domnant colors match the color query. The background analyss can be used before the color hstogram generaton. If the algorthm dentfed any background regons, we wll remove these regons from the mage, and then compute the color hstogram from the remanng regons n the mage. Currently, our mage lbrary conssts of fve dfferent mage collectons: mages of cars, nteror and exteror mages of a Honda vehcle model, crashed cars, face mages, and mages of Earth and Space. We have tested the background analyss algorthm on all these collectons usng dfferent color queres. We found the algorthm s partcular effectve on mages of cars and crashed cars, and mages of Earth and Space Scence. Due to the lmt of space and the absence of color prntng, we are not able to show the mage retreval results n the paper. Readers may want to referred to the extended verson of ths paper that contans a number of examples mage retreval after background removal. 4. Concluson In ths paper, we presented an algorthm for background regon analyss n color magery and the experment results on the algorthm mplemented n a content-based mage retreval lbrary at a Web ste. We showed that the algorthm was effectvely used n content-based mage retreval on a number of dfferent mage collectons. The background removal algorthm can also be used n combnaton wth mage features other than colors. For examples, we can utlze the dentfed background regons to ndex mage background, dfferentate n-door from out-door mages, acheve more accurate retreval usng shape or texture after removng the dentfed background regons. 5. References [BCG98] Serge Belonge, Chad Carson, Hayt Greenspan, and Jtendra Malk, Color- and Texture-Based Image Segmentaton Usng EM and Its Applcaton to Content-Based Image Retreval, ICCV, [ChL98] Te Q Chen and Y Lu, Color Image Segmentaton n Color and Spatal Domans submtted to IEEE Trans. PAMI, 1998 [FDW96] Usama M. Fayyad, S. G. Djorgovsk, and cholas Wer, From Dgtzed Images to Onlne Catalogs, AI magazne, pp , [FJL97] C. O. Frost, R. Johnson, S. Lee, and Y. Lu, Research on Human-centered Desgn and Evaluaton of Multmode Image Retreval, Fourth DELOS Workshop San Mnato, August [FS95] M. Flckner, H. Sawhney, W. black and etc, Query by Image and Vdeo Content, IEEE Computer, pp. - 3, September [Gro88]W. I. Grosky, An archtecture for ntegratng mages and text n a database envronment, n: Database machnes, A. K. Sood and A. H. Quresh, (edt.), ATO ASI Seres, vol. F4, 1988 [Gup96] Amarnath Gupta, The Vrage Image Search Engne: An open framework for mage management, SPIE conference on "Storage and Retreval for Stll Image and Vdeo Databases IV," Feb. 1, [HSE95] J. Hafner, H. S. Sawhney, W. Equtz, et al., Effcent color hstogram ndexng for quadratc form dstance functons, IEEE trans. on PAMI, Vol. 17, o. 7, pp , July 1995 [Ja93] R. Jan, SF Workshop on Vsual Informaton Management Systems, SIGmod Record, vol., o. 3, Dec. 1993, pp [JAV96] A. K. Jan and Adtya Valaya, Image retreval usng color and shape, Pattern Recognton, vol. 9, no. 8, pp , 1996 [ag85] G. agy, Image database, Image and Vson Computng, vol. 3, no. 3, pp , 1985 [PPS94] A. Pentland, R. W. Pcard, and S. Sclaroff, Photobook; Tools for Content-Based Manpulaton of Image Databases, Proc. Storage and Retreval for Image and Vdeo Databases II, Vol., 185, SPIE 1994, pp [SmC94] J. R. Smth and Shh-Fu Chang, Quad-tree segmentaton for texture-based mage query, Proceedngs of nd annual ACM multmeda conference, San Francsco, Ca, Oct [Wl96] Robert Wlensky, Toward Work-Centered Dgtal Informaton Servces, IEEE Computer, pp , May

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

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

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

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

1. Introduction. Abstract

1. Introduction. Abstract Image Retreval Usng a Herarchy of Clusters Danela Stan & Ishwar K. Seth Intellgent Informaton Engneerng Laboratory, Department of Computer Scence & Engneerng, Oaland Unversty, Rochester, Mchgan 48309-4478

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

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

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

More information

Cluster Analysis of Electrical Behavior

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

More information

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

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

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

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

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

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

Query Clustering Using a Hybrid Query Similarity Measure

Query Clustering Using a Hybrid Query Similarity Measure Query clusterng usng a hybrd query smlarty measure Fu. L., Goh, D.H., & Foo, S. (2004). WSEAS Transacton on Computers, 3(3), 700-705. Query Clusterng Usng a Hybrd Query Smlarty Measure Ln Fu, Don Hoe-Lan

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

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

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

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

Chapter 6 Programmng the fnte element method Inow turn to the man subject of ths book: The mplementaton of the fnte element algorthm n computer programs. In order to make my dscusson as straghtforward

More information

Semantic Image Retrieval Using Region Based Inverted File

Semantic Image Retrieval Using Region Based Inverted File Semantc Image Retreval Usng Regon Based Inverted Fle Dengsheng Zhang, Md Monrul Islam, Guoun Lu and Jn Hou 2 Gppsland School of Informaton Technology, Monash Unversty Churchll, VIC 3842, Australa E-mal:

More information

A Multi-step Strategy for Shape Similarity Search In Kamon Image Database

A Multi-step Strategy for Shape Similarity Search In Kamon Image Database A Mult-step Strategy for Shape Smlarty Search In Kamon Image Database Paul W.H. Kwan, Kazuo Torach 2, Kesuke Kameyama 2, Junbn Gao 3, Nobuyuk Otsu 4 School of Mathematcs, Statstcs and Computer Scence,

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

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

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

Visual Thesaurus for Color Image Retrieval using Self-Organizing Maps

Visual Thesaurus for Color Image Retrieval using Self-Organizing Maps Vsual Thesaurus for Color Image Retreval usng Self-Organzng Maps Chrstopher C. Yang and Mlo K. Yp Department of System Engneerng and Engneerng Management The Chnese Unversty of Hong Kong, Hong Kong ABSTRACT

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

Pictures at an Exhibition

Pictures at an Exhibition 1 Pctures at an Exhbton Stephane Kwan and Karen Zhu Department of Electrcal Engneerng Stanford Unversty, Stanford, CA 9405 Emal: {skwan1, kyzhu}@stanford.edu Abstract An mage processng algorthm s desgned

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

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram Shape Representaton Robust to the Sketchng Order Usng Dstance Map and Drecton Hstogram Department of Computer Scence Yonse Unversty Kwon Yun CONTENTS Revew Topc Proposed Method System Overvew Sketch Normalzaton

More information

Efficient Content Representation in MPEG Video Databases

Efficient Content Representation in MPEG Video Databases Effcent Content Representaton n MPEG Vdeo Databases Yanns S. Avrths, Nkolaos D. Doulams, Anastasos D. Doulams and Stefanos D. Kollas Department of Electrcal and Computer Engneerng Natonal Techncal Unversty

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

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

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

USING GRAPHING SKILLS

USING GRAPHING SKILLS Name: BOLOGY: Date: _ Class: USNG GRAPHNG SKLLS NTRODUCTON: Recorded data can be plotted on a graph. A graph s a pctoral representaton of nformaton recorded n a data table. t s used to show a relatonshp

More information

Virtual Machine Migration based on Trust Measurement of Computer Node

Virtual Machine Migration based on Trust Measurement of Computer Node Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on

More information

Color Image Segmentation Using Multispectral Random Field Texture Model & Color Content Features

Color Image Segmentation Using Multispectral Random Field Texture Model & Color Content Features Color Image Segmentaton Usng Multspectral Random Feld Texture Model & Color Content Features Orlando J. Hernandez E-mal: hernande@tcnj.edu Department Electrcal & Computer Engneerng, The College of New

More information

Local Quaternary Patterns and Feature Local Quaternary Patterns

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

More information

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

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

A Clustering Algorithm for Key Frame Extraction Based on Density Peak

A Clustering Algorithm for Key Frame Extraction Based on Density Peak Journal of Computer and Communcatons, 2018, 6, 118-128 http://www.scrp.org/ournal/cc ISSN Onlne: 2327-5227 ISSN Prnt: 2327-5219 A Clusterng Algorthm for Key Frame Extracton Based on Densty Peak Hong Zhao

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

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

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

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

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

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

Efficient Mean Shift Algorithm based Color Images Categorization and Searching

Efficient Mean Shift Algorithm based Color Images Categorization and Searching 152 Effcent Mean Shft Algorthm based Color Images Categorzaton and Searchng 1 Dr S K Vay, 2 Sanay Rathore, 3 Abhshek Verma and 4 Hemra Sngh Thakur 1 Professor, Head of Dept Physcs, Govt Geetanal Grl s

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

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

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

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

Long-Term Moving Object Segmentation and Tracking Using Spatio-Temporal Consistency

Long-Term Moving Object Segmentation and Tracking Using Spatio-Temporal Consistency Long-Term Movng Obect Segmentaton Trackng Usng Spato-Temporal Consstency D Zhong Shh-Fu Chang {dzhong, sfchang}@ee.columba.edu Department of Electrcal Engneerng, Columba Unversty, NY, USA Abstract The

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

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

Video Content Representation using Optimal Extraction of Frames and Scenes

Video Content Representation using Optimal Extraction of Frames and Scenes Vdeo Content Representaton usng Optmal Etracton of rames and Scenes Nkolaos D. Doulam Anastasos D. Doulam Yanns S. Avrths and Stefanos D. ollas Natonal Techncal Unversty of Athens Department of Electrcal

More information

Searching Large Image Databases using Color Information

Searching Large Image Databases using Color Information Searchng Large Image Databases usng Color Informaton Ioan Racu CMSC350: Artfcal Intellgence Wnter Quarter 2004 Department of Computer Scence Unversty of Chcago racu@cs.uchcago.edu ABSTRACT The goal of

More information

Accurate Overlay Text Extraction for Digital Video Analysis

Accurate Overlay Text Extraction for Digital Video Analysis Accurate Overlay Text Extracton for Dgtal Vdeo Analyss Dongqng Zhang, and Shh-Fu Chang Electrcal Engneerng Department, Columba Unversty, New York, NY 10027. (Emal: dqzhang, sfchang@ee.columba.edu) Abstract

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

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

A Fuzzy Image Matching Algorithm with Linguistic Spatial Queries

A Fuzzy Image Matching Algorithm with Linguistic Spatial Queries Fuzzy Matchng lgorthm wth Lngustc Spatal Queres TZUNG-PEI HONG, SZU-PO WNG, TIEN-HIN WNG, EEN-HIN HIEN epartment of Electrcal Engneerng, Natonal Unversty of Kaohsung Insttute of Informaton Management,

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

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

A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION

A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION Mhaela Gordan *, Constantne Kotropoulos **, Apostolos Georgaks **, Ioanns Ptas ** * Bass of Electroncs Department,

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

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

CHAPTER 2 DECOMPOSITION OF GRAPHS

CHAPTER 2 DECOMPOSITION OF GRAPHS CHAPTER DECOMPOSITION OF GRAPHS. INTRODUCTION A graph H s called a Supersubdvson of a graph G f H s obtaned from G by replacng every edge uv of G by a bpartte graph,m (m may vary for each edge by dentfyng

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

SCALABLE AND VISUALIZATION-ORIENTED CLUSTERING FOR EXPLORATORY SPATIAL ANALYSIS

SCALABLE AND VISUALIZATION-ORIENTED CLUSTERING FOR EXPLORATORY SPATIAL ANALYSIS SCALABLE AND VISUALIZATION-ORIENTED CLUSTERING FOR EXPLORATORY SPATIAL ANALYSIS J.H.Guan, F.B.Zhu, F.L.Ban a School of Computer, Spatal Informaton & Dgtal Engneerng Center, Wuhan Unversty, Wuhan, 430079,

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

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

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

Lecture #15 Lecture Notes

Lecture #15 Lecture Notes Lecture #15 Lecture Notes The ocean water column s very much a 3-D spatal entt and we need to represent that structure n an economcal way to deal wth t n calculatons. We wll dscuss one way to do so, emprcal

More information

Editorial Manager(tm) for International Journal of Pattern Recognition and

Editorial Manager(tm) for International Journal of Pattern Recognition and Artfcal Intellgence Edtoral Manager(tm) for Internatonal Journal of Pattern Recognton and Manuscrpt Draft Manuscrpt Number: Ttle: TEXT LOCALIZATION IN COMPLEX COLOR DOCUMENTS Artcle Type: Research Paper

More information

A Computer Vision System for Automated Container Code Recognition

A Computer Vision System for Automated Container Code Recognition A Computer Vson System for Automated Contaner Code Recognton Hsn-Chen Chen, Chh-Ka Chen, Fu-Yu Hsu, Yu-San Ln, Yu-Te Wu, Yung-Nen Sun * Abstract Contaner code examnaton s an essental step n the contaner

More information

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn

More information

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

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

More information

(1) The control processes are too complex to analyze by conventional quantitative techniques.

(1) The control processes are too complex to analyze by conventional quantitative techniques. Chapter 0 Fuzzy Control and Fuzzy Expert Systems The fuzzy logc controller (FLC) s ntroduced n ths chapter. After ntroducng the archtecture of the FLC, we study ts components step by step and suggest a

More information

Comparison Study of Textural Descriptors for Training Neural Network Classifiers

Comparison Study of Textural Descriptors for Training Neural Network Classifiers Comparson Study of Textural Descrptors for Tranng Neural Network Classfers G.D. MAGOULAS (1) S.A. KARKANIS (1) D.A. KARRAS () and M.N. VRAHATIS (3) (1) Department of Informatcs Unversty of Athens GR-157.84

More information

PRÉSENTATIONS DE PROJETS

PRÉSENTATIONS DE PROJETS PRÉSENTATIONS DE PROJETS Rex Onlne (V. Atanasu) What s Rex? Rex s an onlne browser for collectons of wrtten documents [1]. Asde ths core functon t has however many other applcatons that make t nterestng

More information

Image Alignment CSC 767

Image Alignment CSC 767 Image Algnment CSC 767 Image algnment Image from http://graphcs.cs.cmu.edu/courses/15-463/2010_fall/ Image algnment: Applcatons Panorama sttchng Image algnment: Applcatons Recognton of object nstances

More information

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Lecture 13: High-dimensional Images

Lecture 13: High-dimensional Images Lec : Hgh-dmensonal Images Grayscale Images Lecture : Hgh-dmensonal Images Math 90 Prof. Todd Wttman The Ctadel A grayscale mage s an nteger-valued D matrx. An 8-bt mage takes on values between 0 and 55.

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

Novel Fuzzy logic Based Edge Detection Technique

Novel Fuzzy logic Based Edge Detection Technique Novel Fuzzy logc Based Edge Detecton Technque Aborsade, D.O Department of Electroncs Engneerng, adoke Akntola Unversty of Tech., Ogbomoso. Oyo-state. doaborsade@yahoo.com Abstract Ths paper s based on

More information

Photo management applications

Photo management applications Techncal Note PR-TN 7/698 Issued: /7 Photo management applcatons M.A. Peters; P.M.F. Fonseca Phlps Research Europe PR-TN 7/698 Authors address M.A. Peters WB 4 marc.a.peters@phlps.com P.M.F. Fonseca WB

More information

Real-Time View Recognition and Event Detection for Sports Video

Real-Time View Recognition and Event Detection for Sports Video Real-Tme Vew Recognton and Event Detecton for Sports Vdeo Authors: D Zhong and Shh-Fu Chang {dzhong, sfchang@ee.columba.edu} Department of Electrcal Engneerng, Columba Unversty For specal ssue on Multmeda

More information

Correlative features for the classification of textural images

Correlative features for the classification of textural images Correlatve features for the classfcaton of textural mages M A Turkova 1 and A V Gadel 1, 1 Samara Natonal Research Unversty, Moskovskoe Shosse 34, Samara, Russa, 443086 Image Processng Systems Insttute

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

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School

More information

EFFICIENT H.264 VIDEO CODING WITH A WORKING MEMORY OF OBJECTS

EFFICIENT H.264 VIDEO CODING WITH A WORKING MEMORY OF OBJECTS EFFICIENT H.264 VIDEO CODING WITH A WORKING MEMORY OF OBJECTS A Thess presented to the Faculty of the Graduate School at the Unversty of Mssour-Columba In Partal Fulfllment of the Requrements for the Degree

More information

Brushlet Features for Texture Image Retrieval

Brushlet Features for Texture Image Retrieval DICTA00: Dgtal Image Computng Technques and Applcatons, 1 January 00, Melbourne, Australa 1 Brushlet Features for Texture Image Retreval Chbao Chen and Kap Luk Chan Informaton System Research Lab, School

More information

Life Tables (Times) Summary. Sample StatFolio: lifetable times.sgp

Life Tables (Times) Summary. Sample StatFolio: lifetable times.sgp Lfe Tables (Tmes) Summary... 1 Data Input... 2 Analyss Summary... 3 Survval Functon... 5 Log Survval Functon... 6 Cumulatve Hazard Functon... 7 Percentles... 7 Group Comparsons... 8 Summary The Lfe Tables

More information

Real-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution

Real-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution Real-tme Moton Capture System Usng One Vdeo Camera Based on Color and Edge Dstrbuton YOSHIAKI AKAZAWA, YOSHIHIRO OKADA, AND KOICHI NIIJIMA Graduate School of Informaton Scence and Electrcal Engneerng,

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

The Effect of Similarity Measures on The Quality of Query Clusters

The Effect of Similarity Measures on The Quality of Query Clusters The effect of smlarty measures on the qualty of query clusters. Fu. L., Goh, D.H., Foo, S., & Na, J.C. (2004). Journal of Informaton Scence, 30(5) 396-407 The Effect of Smlarty Measures on The Qualty of

More information

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

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

More information

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

Description of NTU Approach to NTCIR3 Multilingual Information Retrieval

Description of NTU Approach to NTCIR3 Multilingual Information Retrieval Proceedngs of the Thrd NTCIR Workshop Descrpton of NTU Approach to NTCIR3 Multlngual Informaton Retreval Wen-Cheng Ln and Hsn-Hs Chen Department of Computer Scence and Informaton Engneerng Natonal Tawan

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