AXON 2 A visual object recognition system for non-rigid objects

Size: px
Start display at page:

Download "AXON 2 A visual object recognition system for non-rigid objects"

Transcription

1 AXON 2 A visual object ecognition system fo non-igid objects PABLO ALVARADO, PEER DÖRFLER, JOCHEN WICKEL Depatment of echnical Compute Science RWH Aachen, Gemany alvaado ABSRAC his aticle descibes a viewe-centeed, featue-based object ecognition system capable to distinguish up to 200 diffeent objects. It has to cope with images of non-igid objects showing patial occlusion. heefoe, we suggest the exclusive use of colo and textue infomation. o accomplish this, we intoduce some novel featue types, which we have successfully tested on a vaiety of classification poblems. KEY WORDS object ecognition, compute vision, neual netwoks, colo constancy, featue extaction, steeable filtes 1. Intoduction he main task of patten ecognition is to find unambiguous identification labels fo some given data. his might be fo example the ecognition of phonemes in an audio steam o the identification of specific pattens in images. Visual ecognition deals with the poblem of finding labels fo eal-wold objects contained in one o moe images of a scene. An application tageted by ecognition systems is an intelligent electonic catalog [1]. hese can be useful fo example in logistics, when a manufactue needs to egiste objects emitted fom etailes. If these objects lack a poduct label, it is a tedious and time-consuming task to look up the object s poduct identification in a catalog manually. Ou system aims to alleviate this task by allowing the use to pesent the object to a camea system and let the object ecognize etieve a list of the most simila objects. Both systems pefom well on an set of about hunded diffeent objects, but they also show some dawbacks. SEEMORE uses intelligently chosen featues, but equies impovements in its classification methods. DyPERS uses local featues, with impede the discimination between vey simila objects, like the ones depicted in Fig. 5. Ou system aims to avoid these limitations by combining diffeent featues with obust and efficient classification algoithms. 1.2 System desciption he AXON 2 system follows a featue-based, viewe-centeed appoach. Usually, an impotant pat of the ecognition is the analysis of object shape. Most of the known shape featues, howeve, suffe fom a stong sensitivity to patial occlusion and defomation. Futhemoe, in lage object sets, thee ae fom pimitives which ae shaed by many objects, making it had to distinguish among them. Additionally ou main object set is constituted by non-igid objects. Due to this facts, AXON 2 elies exclusively on colo and textue. he poposed object ecognition system woks as depicted in Figue 1. It fist acquies an image fom a camea and feeds it to the colo constancy system. Afte that, diffeent kinds of featue vectos ae extacted (Section 2.). he next step consists of thei classification, using RBF neual netwoks (Section 3.), whose esults ae combined to yield the final ecognition esult. Image Acquisition Colo Constancy 1.1 Related wok Result Classification Featue Extaction Many object ecognition systems have been intoduced peviously. Fo a compehensive suvey of olde systems, see [2]. wo of the most inteesting ecent systems ae SEEMORE [3] and DyPERS [4]. SEEMORE is an anthopomophic featue-based, viewe-centeed system that uses a combination of colo and shape desciptions. Classification is pefomed using a neaest-neighbo appoach. DyPERS uses local featues which ae classified using socalled multidimensional eceptive field histogams. Use Inteface Object Recognition Kenel Figue 1. he achitectue of the AXON 2 system. he colo constancy and featue extaction algoithms ae descibed in Section 2., the classification algoithms in Section 3. he system has been tested in a vaiety of situations

2 > A and with diffeent kinds of test data. Expeimental esults with seveal data sets ae pesented in Section Featue Extaction he example application discussed hee equies featue vectos to be invaiant against tanslation and otation (similaity tansfomations on the image plane) and against changes in the illuminant. Futhemoe, the objects to be ecognized have vey simila, non-igid foms. Due to the fact that the use pesents the objects to the system by hand, patial occlusion also has to be expected. Fo these easons only colo and textue featues ae utilized. he following sections conside the extaction of featue vectos fo the whole image (global featues), but the methods could be applied to smalle egions as well (local featues). Figue 2 shows the acquisition and featue extaction achitectue. Chomaticity Histogam Colo Histogams Image Acquisition Colo Constancy Segmentation MR Enegy Featue MR OGD Featue Figue 2. Achitectue fo the featue extaction 2.1 Colo constancy MR OCOGD Featue Finlayson et. al. [5, 6] have shown that a diagonal tansfomation suffices fo colo constancy if thee conditions ae met: the camea sensos exhibit naow-band colo sensitivity, the lighting geomety is constant, and the acquied scene contains lambetian sufaces only. he use of chomaticity channels to attain invaiance against lighting geomety imposes the loss of all infomation contained in the intensity channel, which can be indispensable fo the ecognition. Humans can indeed ecognize objects ignoing all colo infomation, exclusively using the intensity images. In the emainde of this pape will denote a channel containing specific infomation of an image, e. g. colo. Let be a nom which fulfills It can be easily shown that the nomaliza-. tion of each colo channel!, "# and $%! of the RGB colo space with thei espective noms yields a canonical image which is invaiant against changes in the illuminant colo. In [6], the nom '& ()+* (1) is used. &, - epesents the mean value of all pixels in the channel,. We pefe the nom./& (!02143! - (2) whee 1 is a constant and 3! ( the standad deviation of all elements of. he final ange of canonical values using nom (1) will highly depend on the distibution of the oiginal channel values. In Eq. (2) the standad deviation ensues that most of the nomalized values will be kept in a fixed inteval. In ou expeiments 156* is used. his way, if we assume a nomal distibution, the channel values will be mapped into an element of the inteval [0,1] with a pobability of 7879;: <. 2.2 Object segmentation o simplify image segmentation, a black backgound and a black glove fo holding the object ae used. he eal backgound is mapped into an ideal black backgound applying a simple egion gowing algoithm. A smalle image containing the object alone is passed to the next stages of the featue extaction. 2.3 Chomaticity Histogams Since the intoduction of colo histogamming by Swain and Ballad [7], this pinciple has been applied and enhanced in many object and image ecognition systems (e.g. [3, 8]). he chomaticity histogam exclusively contains infomation about colo, ignoing the intensity. his fact is used in [6] to poduce canonical images invaiant against changes of the lighting geomety, which ae eflected exclusively in the intensity channel. he chomaticity channels! ae defined as = >@?AB?C and D >@?AE?C. he 2D chomaticity histogam used hee contains 16 bins fo each coodinate. o incease the obustness against noise, the 2D histogam is convolved with a Gaussian *=FG* low-pass filte kenel (HJIK)+L MIN)O,IK)+L8P QHJIK)+L MIN)O,IK)+L8P ). Consideing that chomaticity values cove only one half of the histogam bins, the esulting vecto equies no moe than 136 dimensions. Scale invaiance is achieved by nomalizing the volume unde the histogam. All ideally black pixels ae ignoed; this way only the object pixels ae consideed. 2.4 Colo Histogams he chomaticity featue discads valuable infomation fo the ecognition. Consideing the pevious geneation of a canonical image invaiant against changes in the illuminant colo, the intensity infomation can also be obustly evaluated using simple colo histogams, geneated in the following way. Fou 1D histogams with 32 bins each ae calculated fo the ed, geen, blue and luminance channels, whee luminance is defined as RS

3 Ÿ V.UJVW >YXAXC!Z[?.\^]_W >YXAXC!Z. Again, to educe the effects of noise, a a IbFc* low-pass filte kenel (H IN)L,IK)+O4,IN)L P ) is applied to each histogam. he fou histogams ae concatenated into a 128-dimensional vecto. 2.5 Multiesolutional Enegy Featue Using the concepts fo textue channels pesented in [9], a textue-enegy image is geneated. Smith s suggestions fo scale and otation invaiance ae also consideed. A channel is fist tansfomed into its wavelet epesentation with fou esolutions, using a 9-tap symmetic kenel de#fh g9 g O h IN _i.g9 g j g87 i.g49 g :*8L +g49 L IL :N ^g49 :^7@:L4 g9 g :* L4 4i.g49 g8j g 74 g49 g8o h@i^p. he low-pass channel is ignoed, which means that only the thee following bands ae consideed (nine channels in total). he enegy channels ae calculated applying k Kk to each pixel of each spectal channel. hey ae then combined into a single image using the multiscaled addition defined in [10]. hese enegy images ae calculated fo the thee colo channels, " and $. Histogams with 32 bins fo each channel ae concatenated into a 96-dimensional featue vecto. he same noise-eduction measues as fo the colo featues ae applied. Anothe featue appoximates the multiesolutional enegy calculating the enegy of the pixels of a high-pass filteed image. he filte subtacts a low-pass filteed channel fom the oiginal. he used low-pass kenel is a O lmfno l Gaussian with vaiance 31. A 96-dimensional featue vecto is geneated following the same pinciples as fo the multiesolutional enegy featue. 2.6 Multiesolutional OGD Featue he infomation contained in each specific spectal band of a multiesolutional decomposition of an image has been successfully applied (fo example [11, 12]). hese featues, howeve, have a limited applicability in 3D object ecognition due to thei lack of invaiance against scaling and otation. Rotation invaiant and scaling equivaiant featues can be geneated exploiting the steeability popety [13] of oiented Gaussian deivatives (OGD). Let (o_ be a spectal component of a channel p extacted using a steeable filte o, which can be expessed as: o q sutwv s x8 s o (3) whee s x_ denotes the intepolation functions and o s the basis filtes. Let yzo! be the powe of the channel (o! defined as y o q o U Xsutwv sutv U Xs x_ y o^{; } s x8 os with U Xs x8b~ U x_ s x8, yzo {[ } b~(ou (os! and (os = p n o s whee denotes the convolution opeato. Eq. (4) indicates that the powe is also steeable using intepolation functions U X s x8 and basis powe channels y ou Xs. Let mo > be the enegy fo a egion ƒ Fˆ : > o W ] XŠZ( N> y o Œ U Xsutwv (4) U Xs x8 U Xs (5) with U Xs W ] XŠZ( N> yzo^{; }. he total enegy mo > is theefoe also steeable. With U Xs Ž s XU, it follows: > o U twv U XU x8 U XU 0O v U s U Xs x8 U Xs (6) With Eq. (6) and the steeability of the Gaussian deivatives, it can also be poven that the total enegy of a egion in os fo a -th ode Gaussian deivative filte is given by: > o 6 Y m0 U twv UN ;ON ^xši5 U (7) whee the U do not depend on the steeing angle x, i. e. the desciptos U ae otation invaiant. able 1 shows the values of these tems fo the fist and second oiented gaussian deivatives. he oiginal channel can be analyzed applying this method at diffeent esolutions using fo example dyadic pyamid achitectues. his way, diffeent fequency bands can be consideed in the final featue vecto. We employ a dyadic pyamid of OGDs with standad deviation 3 œ3,o fo each level used. Let ž;mo denote the total enegy of a channel, computed by Eq. (7) using a - and th ode OGD steeed towads x with a vaiance 3 a scale Ÿ O_ v. It can be easily poven that W V v Z o ž[mo?, i. e. the space-scaling of an image by Ÿ shifts the enegy in the esolution axis by = ŸK and W V v Z inceases the enegy level by Ÿ. Fo the constucted featue vecto a 2nd ode OGD dyadic pyamid with fou levels ( GL ) is geneated. Fo this task, a I* FªI* OGD filte kenel with vaiance 3 O is employed. In ode to enhance the esolution axis, a second pyamid with 3z«O3 is also ceated. he coefficients U in able 1 ae calculated fo each level of both pyamids. he final vecto is composed of all desciptos

4 ± ½ Ì Ê º Å Ä Ì Ä V able 1. Coefficients fo the enegy with 1st and 2nd OGD ± ³ ± à ½m Æ-Ç È+Æ-É ±³ 1st OGD # q ²±³! µ±b,,¹º(¼n ¾½m - ćä!å Å, 2nd OGD ²±m³ ±# ¹,º-¼n 5½m 4 ± E 4ćÄ + ±# Å 5º, E ½m Æ-Ç È+Æ É Å Æ Ç È+Æ-É Äzº Å Å + Å,¹º Ä,¼n 5½ ¾ Å U soted fist by 3 and then by the index. heefoe, the geneated vecto has 24 dimensions. his pocedue can conside colo infomation if it is applied to each of the, " and $ channels. he final featue vecto has 72 dimensions. 2.7 Opponent Colo Multiesolutional Featue Until now, the colo infomation has been consideed using the, " and $ colo channels sepaately. Anothe appoach to colo analysis is inspied by the opponent colo theoy of human vision. Hee, eceptive fields oganized in a cente-suound fashion compae the stimulus of the cente (e. g. ed) with the stimulus in the suounding (e. g. geen). his can be modeled using the scheme shown in Fig. 3. Colo Image Colo Space cente suound OGD 2 OGD 2 + Figue 3. Geneation of an opponent colo channel opponent colo channel If the enegy o \K Í fo the opponent colo channel is calculated as descibed by Eq. (6), it follows: \K 4Í o 6 \ o 02 Ío ig \Î Í o (8) Compaing this tem with the pevious featues, it is clea that the new infomation gained compaing both cente and suound channels is contained in the opponent colo en- \Î Í. Fo the OGD, this tem can be also expessed egy o as \Î%Í o Ž 0 U twv U + 8 [ON ^xši¾ U (9) Fo the 1st ode OGD, the new desciptos ae: Ž Ï 02 e \+Í ^Ï v Ï ig e Ï \+Í ^Ï v 'Ð8Ñ Ò Xe ^Ï Ð Ó Ï ig e Ï with ^Ï e ^Ï \+Í Ï Xe ^Ï W ] XŠZ N> W ] XŠZ N> W ] XŠZ N> ^Ï \ ^Ï Í e \ ^Ï! e Í Ï ^Ï \ e Í ^Ï!!0 0 Ï Xe Ï ^Ï Í e \ ^Ï Combining the desciptos and v of both input channels (cente and suound " ) and the new opponent colo channel, a featue vecto with 48 dimensions is geneated. 3. Classification he classification in AXON 2 follows a hieachical appoach as shown in Figue 4. In the fist stage, Radial Basis Function (RBF) neual netwoks can be found, one fo each of the featue types descibed in Section 2. he position of the basis functions is detemined by the OLVQ3 taining algoithm [14]. he esult of the output laye is computed using a sigmoid activation function. In the second stage, a combination module geneates a final classification esult fom the answes of the individual netwoks by linea combination [15]. 4. Applications and Results he applicability of object ecognition systems as inteface tools fo accessing electonic catalogues obviously depends on thei ecognition coectness. Howeve, fo inteactive applications it is not geneally necessay to pefom an unique identification. As outlined in the intoduction, we ae pimaily concened with finding simila objects. heefoe, it is sufficient if the system pesents the most simila objects at the top of the esult list. Fo that pupose,

5 Featue1 Featue2 Featue3 RBF netwoks Combine Figue 4. Achitectue fo the classification we use the concept of -best ecognition, which descibes the pobability of an object to be at position o bette in the esult list. AXON 2 has been tested on thee diffeent object sets. he fist one is the COIL database [16], consisting of 100 objects, each one epesented by 72 views. 36 of these views wee used fo taining, the emaining ones wee used fo testing. he second set is composed of views of 42 plush animals. aining was pefomed using 35 andom views fo each object, the test set consisted of 840 andom views, 20 fo each object; both sets wee disjunct. he thid set contains views of 202 plush animals. 40 views, taken fom camea points unifomly distibuted on an uppe hemisphee above the object, seved as taining set. he test set consisted of 36 andom views fo each object. his set contains objects which appea almost identical to each othe (Figue 5). AXON 2 is able to disciminate them, even tough they diffe in shape only. his demonstates that the textue featues intoduced hee indiectly incopoate shape infomation. he whole pocess fom image acquisition to classification esult takes no moe than two seconds on a standad Pentium II PC (350 MHz). Fo the object set of 202 objects, taining times ange fom 20 minutes to two hous pe RBF netwok, depending on the dimensionality of the featue vectos. he esults pesented hee wee obtained using the following configuation: Ô Featues: Chomaticity Histogam, Colo Histogam, Multiesolutional Enegy, Multiesolutional OGD, Multiesolutional OC OGD, and high pass filte vecto, as descibed in Section 2. Ô Classifie: six RBF netwoks, each with six units pe class, pototypes estimated by OLVQ3. Ô Combination by aveaging the classifie esults. able 2 shows the -best ecognition ates fo these thee sets. he COIL database is an easily solvable poblem. he 42 plush objects ae hade to ecognize due to Figue 5. wo objects fom the test set. he two images on uppe side ae scanned fom the manufactue s catalog. he two images on the bottom side show one actual test image fo each object as taken fom AXON 2 s camea. Both test images ae classified coectly. able 2. Recognition ates fo thee diffeent test object sets. Object Set Views 1-Best 2-Best 3-Best COIL % 42 plush obj % 99.9% 100% 202 plush obj % 93.8% 95.8% stong similaities among some of them. he esult with 202 objects suggests that viewe-centeed, featue-based object ecognition is a pomising appoach fo object sets containing seveal hunded objects, even though thee is oom fo impovements. able 3 shows the ecognition esults fo 202 objects, using the best combinations of featues. It can be seen, that always the best esults ae obtained by using both colo and textue featues. he most poweful single featue type is the Colo Histogam, despite its simplicity. he high pass filte featue impoves the esults only maginally. 5. Conclusions and futue wok We have pesented an object ecognition system which is able to yield good ecognition esults fo object sets up to 200 objects despite the pesence of patial occlusion. We have demonstated that the combination of appopiately designed featues with neual classifies can cope with vey difficult ecognition poblems. Even though the pesented appoach explicitly excludes shape featues, the system is capable of identifying objects which diffe almost only in

6 able 3. Best ecognition ates fo diffeent featue combinations. # Featues Ch. Hist. Col. Hist. MR OC OGD MR OGD MR Enegy high pass 1-Best 2-Best 3-Best 1 Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ Õ shape. he successo of the pesented system will ovecome some of its limitations. hese ae on the one hand, its estiction to scenes showing only one object, on the othe hand, its inability to handle object sets containing moe than seveal hunded objects. he expeience fom woking with the system pesented hee makes us confident that the new impoved system will be able to disciminate object sets of seveal thousand objects. 6. Acknowledgements his poject has been suppoted by the Heinz-Nixdof Foundation. We would like to thank Dik Kumbiegel, Ingo Elsen, and Pete Walte, who wee ou co-wokes on this poject and who deseve a geat pat of the cedit fo the poject s success. We also would like to thank ou students who helped implementing the system. Refeences [1] I. Elsen, K.-F. Kaiss, D. Kumbiegel, P. Walte, and J. Wickel. Visual Infomation Retieval fo 3D Poduct Identification. Künstliche Intelligenz, Jan. 1999, 1/99: [2] A. R. Pope. Model-based object ecognition a suvey of ecent eseach. echnical epot 94 04, Univesity of Bitish Columbia, [3] B. W. Mel. SEEMORE: Combining Colo, Shape, and extue Histogamming in a Neually Inspied Appoach to Visual Object Recognition. Neual Computation, May 1997, 9(4): the 4th Intenational Confeence on Compute Vision. IEEE Compute Society, 1993, [6] G. D. Finlayson, B. Schiele, and J. L. Cowley. Compehensive colou image nomalization. In ECCV, 1998, [7] M. Swain and D. H. Ballad. Colo indexing. Intenational Jounal of Compute Vision, 1991, 7(1): [8] M. S. Dew, J. W. Wei, and Z. Li. Illuminationinvaiant colo object ecognition via compessed chomaticity histogams of nomalized images. echnical Repot CMP-R 97-09, Simon Fase Univesity School of Compute Science, [9] J. R. Smith. Integated Spatial and Featue Image Systems: Retieval, Analysis and Compession. PhD thesis, Columbia Univesity, [10] L. Itti, C. Koch, and E. Niebu. A model of saliencybased visual attention fo apid scene analysis. IEEE ansactions on Patten Analysis and Machine Intelligence, Novembe 1998, 20(11): [11] A. Jain and G. Healey. A multiscale epesentation including opponent colo featues fo textue ecognition. IEEE ansaction on Image Pocessing, Janua 1998, 7(1): [12] B. S. Manjunath and W. Y. Ma. extue featues fo bowsing and etieval of image data. IEEE ansactions on Patten Analysis and Machine Intelligence, August 1996, 18(8): [13] W.. Feeman and E. H. Adelson. he design and use of steeable filtes. IEEE ansactions on Patten Analysis and Machine Intelligence, Septembe 1991, 13(9): [14]. Kohonen. Self-Oganizing Maps. Spinge-Velag Belin Heidelbeg New Yok, [15] K. ume and J. Ghosh. heoetical Foundations of Linea and Ode Statistics Combines fo Neual Patten Classifies. echnical Repot R , Cente fo Imaging Science, John Hopkins Univesity, Baltimoe, USA, [16] S. A. Nene, S. K. Naya, and H. Muase. Columbia Object Image Libay (COIL-100). echnical Repot CUCS , Columbia Univesity, [4] B. Schiele and A. Pentland. Pobabilistic object ecognition and localization. In Poc. Int. Conf. on Compute Vision ICCV 99, 1999, [5] G. D. Finlayson, M. Dew, and B. Funt. Diagonal tansfoms suffice fo colo constancy. In Poc. of

Detection and Recognition of Alert Traffic Signs

Detection and Recognition of Alert Traffic Signs Detection and Recognition of Alet Taffic Signs Chia-Hsiung Chen, Macus Chen, and Tianshi Gao 1 Stanfod Univesity Stanfod, CA 9305 {echchen, macuscc, tianshig}@stanfod.edu Abstact Taffic signs povide dives

More information

Journal of World s Electrical Engineering and Technology J. World. Elect. Eng. Tech. 1(1): 12-16, 2012

Journal of World s Electrical Engineering and Technology J. World. Elect. Eng. Tech. 1(1): 12-16, 2012 2011, Scienceline Publication www.science-line.com Jounal of Wold s Electical Engineeing and Technology J. Wold. Elect. Eng. Tech. 1(1): 12-16, 2012 JWEET An Efficient Algoithm fo Lip Segmentation in Colo

More information

Controlled Information Maximization for SOM Knowledge Induced Learning

Controlled Information Maximization for SOM Knowledge Induced Learning 3 Int'l Conf. Atificial Intelligence ICAI'5 Contolled Infomation Maximization fo SOM Knowledge Induced Leaning Ryotao Kamimua IT Education Cente and Gaduate School of Science and Technology, Tokai Univeisity

More information

An Unsupervised Segmentation Framework For Texture Image Queries

An Unsupervised Segmentation Framework For Texture Image Queries An Unsupevised Segmentation Famewok Fo Textue Image Queies Shu-Ching Chen Distibuted Multimedia Infomation System Laboatoy School of Compute Science Floida Intenational Univesity Miami, FL 33199, USA chens@cs.fiu.edu

More information

A Two-stage and Parameter-free Binarization Method for Degraded Document Images

A Two-stage and Parameter-free Binarization Method for Degraded Document Images A Two-stage and Paamete-fee Binaization Method fo Degaded Document Images Yung-Hsiang Chiu 1, Kuo-Liang Chung 1, Yong-Huai Huang 2, Wei-Ning Yang 3, Chi-Huang Liao 4 1 Depatment of Compute Science and

More information

An Extension to the Local Binary Patterns for Image Retrieval

An Extension to the Local Binary Patterns for Image Retrieval , pp.81-85 http://x.oi.og/10.14257/astl.2014.45.16 An Extension to the Local Binay Pattens fo Image Retieval Zhize Wu, Yu Xia, Shouhong Wan School of Compute Science an Technology, Univesity of Science

More information

Topic -3 Image Enhancement

Topic -3 Image Enhancement Topic -3 Image Enhancement (Pat 1) DIP: Details Digital Image Pocessing Digital Image Chaacteistics Spatial Spectal Gay-level Histogam DFT DCT Pe-Pocessing Enhancement Restoation Point Pocessing Masking

More information

MULTI-TEMPORAL AND MULTI-SENSOR IMAGE MATCHING BASED ON LOCAL FREQUENCY INFORMATION

MULTI-TEMPORAL AND MULTI-SENSOR IMAGE MATCHING BASED ON LOCAL FREQUENCY INFORMATION Intenational Achives of the Photogammety Remote Sensing and Spatial Infomation Sciences Volume XXXIX-B3 2012 XXII ISPRS Congess 25 August 01 Septembe 2012 Melboune Austalia MULTI-TEMPORAL AND MULTI-SENSOR

More information

Image Enhancement in the Spatial Domain. Spatial Domain

Image Enhancement in the Spatial Domain. Spatial Domain 8-- Spatial Domain Image Enhancement in the Spatial Domain What is spatial domain The space whee all pixels fom an image In spatial domain we can epesent an image by f( whee x and y ae coodinates along

More information

A Novel Automatic White Balance Method For Digital Still Cameras

A Novel Automatic White Balance Method For Digital Still Cameras A Novel Automatic White Balance Method Fo Digital Still Cameas Ching-Chih Weng 1, Home Chen 1,2, and Chiou-Shann Fuh 3 Depatment of Electical Engineeing, 2 3 Gaduate Institute of Communication Engineeing

More information

IP Network Design by Modified Branch Exchange Method

IP Network Design by Modified Branch Exchange Method Received: June 7, 207 98 IP Netwok Design by Modified Banch Method Kaiat Jaoenat Natchamol Sichumoenattana 2* Faculty of Engineeing at Kamphaeng Saen, Kasetsat Univesity, Thailand 2 Faculty of Management

More information

(a, b) x y r. For this problem, is a point in the - coordinate plane and is a positive number.

(a, b) x y r. For this problem, is a point in the - coordinate plane and is a positive number. Illustative G-C Simila cicles Alignments to Content Standads: G-C.A. Task (a, b) x y Fo this poblem, is a point in the - coodinate plane and is a positive numbe. a. Using a tanslation and a dilation, show

More information

Color Correction Using 3D Multiview Geometry

Color Correction Using 3D Multiview Geometry Colo Coection Using 3D Multiview Geomety Dong-Won Shin and Yo-Sung Ho Gwangju Institute of Science and Technology (GIST) 13 Cheomdan-gwagio, Buk-ku, Gwangju 500-71, Republic of Koea ABSTRACT Recently,

More information

Effects of Model Complexity on Generalization Performance of Convolutional Neural Networks

Effects of Model Complexity on Generalization Performance of Convolutional Neural Networks Effects of Model Complexity on Genealization Pefomance of Convolutional Neual Netwoks Tae-Jun Kim 1, Dongsu Zhang 2, and Joon Shik Kim 3 1 Seoul National Univesity, Seoul 151-742, Koea, E-mail: tjkim@bi.snu.ac.k

More information

Spiral Recognition Methodology and Its Application for Recognition of Chinese Bank Checks

Spiral Recognition Methodology and Its Application for Recognition of Chinese Bank Checks Spial Recognition Methodology and Its Application fo Recognition of Chinese Bank Checks Hanshen Tang 1, Emmanuel Augustin 2, Ching Y. Suen 1, Olivie Baet 2, Mohamed Cheiet 3 1 Cente fo Patten Recognition

More information

A Neural Network Model for Storing and Retrieving 2D Images of Rotated 3D Object Using Principal Components

A Neural Network Model for Storing and Retrieving 2D Images of Rotated 3D Object Using Principal Components A Neual Netwok Model fo Stong and Reteving 2D Images of Rotated 3D Object Using Pncipal Components Tsukasa AMANO, Shuichi KUROGI, Ayako EGUCHI, Takeshi NISHIDA, Yasuhio FUCHIKAWA Depatment of Contol Engineeng,

More information

Cardiac C-Arm CT. SNR Enhancement by Combining Multiple Retrospectively Motion Corrected FDK-Like Reconstructions

Cardiac C-Arm CT. SNR Enhancement by Combining Multiple Retrospectively Motion Corrected FDK-Like Reconstructions Cadiac C-Am CT SNR Enhancement by Combining Multiple Retospectively Motion Coected FDK-Like Reconstuctions M. Pümme 1, L. Wigstöm 2,3, R. Fahig 2, G. Lauitsch 4, J. Honegge 1 1 Institute of Patten Recognition,

More information

A modal estimation based multitype sensor placement method

A modal estimation based multitype sensor placement method A modal estimation based multitype senso placement method *Xue-Yang Pei 1), Ting-Hua Yi 2) and Hong-Nan Li 3) 1),)2),3) School of Civil Engineeing, Dalian Univesity of Technology, Dalian 116023, China;

More information

Optical Flow for Large Motion Using Gradient Technique

Optical Flow for Large Motion Using Gradient Technique SERBIAN JOURNAL OF ELECTRICAL ENGINEERING Vol. 3, No. 1, June 2006, 103-113 Optical Flow fo Lage Motion Using Gadient Technique Md. Moshaof Hossain Sake 1, Kamal Bechkoum 2, K.K. Islam 1 Abstact: In this

More information

Segmentation of Casting Defects in X-Ray Images Based on Fractal Dimension

Segmentation of Casting Defects in X-Ray Images Based on Fractal Dimension 17th Wold Confeence on Nondestuctive Testing, 25-28 Oct 2008, Shanghai, China Segmentation of Casting Defects in X-Ray Images Based on Factal Dimension Jue WANG 1, Xiaoqin HOU 2, Yufang CAI 3 ICT Reseach

More information

Frequency Domain Approach for Face Recognition Using Optical Vanderlugt Filters

Frequency Domain Approach for Face Recognition Using Optical Vanderlugt Filters Optics and Photonics Jounal, 016, 6, 94-100 Published Online August 016 in SciRes. http://www.scip.og/jounal/opj http://dx.doi.og/10.436/opj.016.68b016 Fequency Domain Appoach fo Face Recognition Using

More information

Illumination methods for optical wear detection

Illumination methods for optical wear detection Illumination methods fo optical wea detection 1 J. Zhang, 2 P.P.L.Regtien 1 VIMEC Applied Vision Technology, Coy 43, 5653 LC Eindhoven, The Nethelands Email: jianbo.zhang@gmail.com 2 Faculty Electical

More information

Extract Object Boundaries in Noisy Images using Level Set. Final Report

Extract Object Boundaries in Noisy Images using Level Set. Final Report Extact Object Boundaies in Noisy Images using Level Set by: Quming Zhou Final Repot Submitted to Pofesso Bian Evans EE381K Multidimensional Digital Signal Pocessing May 10, 003 Abstact Finding object contous

More information

Towards Adaptive Information Merging Using Selected XML Fragments

Towards Adaptive Information Merging Using Selected XML Fragments Towads Adaptive Infomation Meging Using Selected XML Fagments Ho-Lam Lau and Wilfed Ng Depatment of Compute Science and Engineeing, The Hong Kong Univesity of Science and Technology, Hong Kong {lauhl,

More information

Prof. Feng Liu. Fall /17/2016

Prof. Feng Liu. Fall /17/2016 Pof. Feng Liu Fall 26 http://www.cs.pdx.edu/~fliu/couses/cs447/ /7/26 Last time Compositing NPR 3D Gaphics Toolkits Tansfomations 2 Today 3D Tansfomations The Viewing Pipeline Mid-tem: in class, Nov. 2

More information

Voting-Based Grouping and Interpretation of Visual Motion

Voting-Based Grouping and Interpretation of Visual Motion Voting-Based Gouping and Intepetation of Visual Motion Micea Nicolescu Depatment of Compute Science Univesity of Nevada, Reno Reno, NV 89557 micea@cs.un.edu Géad Medioni Integated Media Systems Cente Univesity

More information

Positioning of a robot based on binocular vision for hand / foot fusion Long Han

Positioning of a robot based on binocular vision for hand / foot fusion Long Han 2nd Intenational Confeence on Advances in Mechanical Engineeing and Industial Infomatics (AMEII 26) Positioning of a obot based on binocula vision fo hand / foot fusion Long Han Compute Science and Technology,

More information

Image Registration among UAV Image Sequence and Google Satellite Image Under Quality Mismatch

Image Registration among UAV Image Sequence and Google Satellite Image Under Quality Mismatch 0 th Intenational Confeence on ITS Telecommunications Image Registation among UAV Image Sequence and Google Satellite Image Unde Quality Mismatch Shih-Ming Huang and Ching-Chun Huang Depatment of Electical

More information

ANALYTIC PERFORMANCE MODELS FOR SINGLE CLASS AND MULTIPLE CLASS MULTITHREADED SOFTWARE SERVERS

ANALYTIC PERFORMANCE MODELS FOR SINGLE CLASS AND MULTIPLE CLASS MULTITHREADED SOFTWARE SERVERS ANALYTIC PERFORMANCE MODELS FOR SINGLE CLASS AND MULTIPLE CLASS MULTITHREADED SOFTWARE SERVERS Daniel A Menascé Mohamed N Bennani Dept of Compute Science Oacle, Inc Geoge Mason Univesity 1211 SW Fifth

More information

Obstacle Avoidance of Autonomous Mobile Robot using Stereo Vision Sensor

Obstacle Avoidance of Autonomous Mobile Robot using Stereo Vision Sensor Obstacle Avoidance of Autonomous Mobile Robot using Steeo Vision Senso Masako Kumano Akihisa Ohya Shin ichi Yuta Intelligent Robot Laboatoy Univesity of Tsukuba, Ibaaki, 35-8573 Japan E-mail: {masako,

More information

Haptic Glove. Chan-Su Lee. Abstract. This is a final report for the DIMACS grant of student-initiated project. I implemented Boundary

Haptic Glove. Chan-Su Lee. Abstract. This is a final report for the DIMACS grant of student-initiated project. I implemented Boundary Physically Accuate Haptic Rendeing of Elastic Object fo a Haptic Glove Chan-Su Lee Abstact This is a final epot fo the DIMACS gant of student-initiated poject. I implemented Bounday Element Method(BEM)

More information

HISTOGRAMS are an important statistic reflecting the

HISTOGRAMS are an important statistic reflecting the JOURNAL OF L A T E X CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 1 D 2 HistoSketch: Disciminative and Dynamic Similaity-Peseving Sketching of Steaming Histogams Dingqi Yang, Bin Li, Laua Rettig, and Philippe

More information

Improvement of First-order Takagi-Sugeno Models Using Local Uniform B-splines 1

Improvement of First-order Takagi-Sugeno Models Using Local Uniform B-splines 1 Impovement of Fist-ode Takagi-Sugeno Models Using Local Unifom B-splines Felipe Fenández, Julio Gutiéez, Gacián Tiviño and Juan Calos Cespo Dep. Tecnología Fotónica, Facultad de Infomática Univesidad Politécnica

More information

Lecture # 04. Image Enhancement in Spatial Domain

Lecture # 04. Image Enhancement in Spatial Domain Digital Image Pocessing CP-7008 Lectue # 04 Image Enhancement in Spatial Domain Fall 2011 2 domains Spatial Domain : (image plane) Techniques ae based on diect manipulation of pixels in an image Fequency

More information

Visual Servoing from Deep Neural Networks

Visual Servoing from Deep Neural Networks Visual Sevoing fom Deep Neual Netwoks Quentin Bateux 1, Eic Machand 1, Jügen Leitne 2, Fançois Chaumette 3, Pete Coke 2 Abstact We pesent a deep neual netwok-based method to pefom high-pecision, obust

More information

COLOR EDGE DETECTION IN RGB USING JOINTLY EUCLIDEAN DISTANCE AND VECTOR ANGLE

COLOR EDGE DETECTION IN RGB USING JOINTLY EUCLIDEAN DISTANCE AND VECTOR ANGLE COLOR EDGE DETECTION IN RGB USING JOINTLY EUCLIDEAN DISTANCE AND VECTOR ANGLE Slawo Wesolkowski Systems Design Engineeing Univesity of Wateloo Wateloo (Ont.), Canada, NL 3G s.wesolkowski@ieee.og Ed Jenigan

More information

A ROI Focusing Mechanism for Digital Cameras

A ROI Focusing Mechanism for Digital Cameras A ROI Focusing Mechanism fo Digital Cameas Chu-Hui Lee, Meng-Feng Lin, Chun-Ming Huang, and Chun-Wei Hsu Abstact With the development and application of digital technologies, the digital camea is moe popula

More information

Multi-azimuth Prestack Time Migration for General Anisotropic, Weakly Heterogeneous Media - Field Data Examples

Multi-azimuth Prestack Time Migration for General Anisotropic, Weakly Heterogeneous Media - Field Data Examples Multi-azimuth Pestack Time Migation fo Geneal Anisotopic, Weakly Heteogeneous Media - Field Data Examples S. Beaumont* (EOST/PGS) & W. Söllne (PGS) SUMMARY Multi-azimuth data acquisition has shown benefits

More information

Introduction to Medical Imaging. Cone-Beam CT. Introduction. Available cone-beam reconstruction methods: Our discussion:

Introduction to Medical Imaging. Cone-Beam CT. Introduction. Available cone-beam reconstruction methods: Our discussion: Intoduction Intoduction to Medical Imaging Cone-Beam CT Klaus Muelle Available cone-beam econstuction methods: exact appoximate Ou discussion: exact (now) appoximate (next) The Radon tansfom and its invese

More information

Improved Fourier-transform profilometry

Improved Fourier-transform profilometry Impoved Fouie-tansfom pofilomety Xianfu Mao, Wenjing Chen, and Xianyu Su An impoved optical geomety of the pojected-finge pofilomety technique, in which the exit pupil of the pojecting lens and the entance

More information

Transmission Lines Modeling Based on Vector Fitting Algorithm and RLC Active/Passive Filter Design

Transmission Lines Modeling Based on Vector Fitting Algorithm and RLC Active/Passive Filter Design Tansmission Lines Modeling Based on Vecto Fitting Algoithm and RLC Active/Passive Filte Design Ahmed Qasim Tuki a,*, Nashien Fazilah Mailah b, Mohammad Lutfi Othman c, Ahmad H. Saby d Cente fo Advanced

More information

Generalized Grey Target Decision Method Based on Decision Makers Indifference Attribute Value Preferences

Generalized Grey Target Decision Method Based on Decision Makers Indifference Attribute Value Preferences Ameican Jounal of ata ining and Knowledge iscovey 27; 2(4): 2-8 http://www.sciencepublishinggoup.com//admkd doi:.648/.admkd.2724.2 Genealized Gey Taget ecision ethod Based on ecision akes Indiffeence Attibute

More information

A Shape-preserving Affine Takagi-Sugeno Model Based on a Piecewise Constant Nonuniform Fuzzification Transform

A Shape-preserving Affine Takagi-Sugeno Model Based on a Piecewise Constant Nonuniform Fuzzification Transform A Shape-peseving Affine Takagi-Sugeno Model Based on a Piecewise Constant Nonunifom Fuzzification Tansfom Felipe Fenández, Julio Gutiéez, Juan Calos Cespo and Gacián Tiviño Dep. Tecnología Fotónica, Facultad

More information

Color Interpolation for Single CCD Color Camera

Color Interpolation for Single CCD Color Camera Colo Intepolation fo Single CCD Colo Camea Yi-Ming Wu, Chiou-Shann Fuh, and Jui-Pin Hsu Depatment of Compute Science and Infomation Engineeing, National Taian Univesit, Taipei, Taian Email: 88036@csie.ntu.edu.t;

More information

Optimal Adaptive Learning for Image Retrieval

Optimal Adaptive Learning for Image Retrieval Optimal Adaptive Leaning fo Image Retieval ao Wang Dept of Compute Sci and ech singhua Univesity Beijing 00084, P. R. China Wangtao7@63.net Yong Rui Micosoft Reseach One Micosoft Way Redmond, WA 9805,

More information

Fifth Wheel Modelling and Testing

Fifth Wheel Modelling and Testing Fifth heel Modelling and Testing en Masoy Mechanical Engineeing Depatment Floida Atlantic Univesity Boca aton, FL 4 Lois Malaptias IFMA Institut Fancais De Mechanique Advancee ampus De lemont Feand Les

More information

AUTOMATED LOCATION OF ICE REGIONS IN RADARSAT SAR IMAGERY

AUTOMATED LOCATION OF ICE REGIONS IN RADARSAT SAR IMAGERY AUTOMATED LOCATION OF ICE REGIONS IN RADARSAT SAR IMAGERY Chistophe Waceman (1), William G. Pichel (2), Pablo Clement-Colón (2) (1) Geneal Dynamics Advanced Infomation Systems, P.O. Box 134008 Ann Abo

More information

A New and Efficient 2D Collision Detection Method Based on Contact Theory Xiaolong CHENG, Jun XIAO a, Ying WANG, Qinghai MIAO, Jian XUE

A New and Efficient 2D Collision Detection Method Based on Contact Theory Xiaolong CHENG, Jun XIAO a, Ying WANG, Qinghai MIAO, Jian XUE 5th Intenational Confeence on Advanced Mateials and Compute Science (ICAMCS 2016) A New and Efficient 2D Collision Detection Method Based on Contact Theoy Xiaolong CHENG, Jun XIAO a, Ying WANG, Qinghai

More information

Hand Tracking and Gesture Recognition for Human-Computer Interaction

Hand Tracking and Gesture Recognition for Human-Computer Interaction Electonic Lettes on Compute Vision and Image Analysis 5(3):96-104, 2005 Hand Tacking and Gestue Recognition fo Human-Compute Inteaction Cistina Manesa, Javie Vaona, Ramon Mas and Fancisco J. Peales Unidad

More information

Ego-Motion Estimation on Range Images using High-Order Polynomial Expansion

Ego-Motion Estimation on Range Images using High-Order Polynomial Expansion Ego-Motion Estimation on Range Images using High-Ode Polynomial Expansion Bian Okon and Josh Haguess Space and Naval Wafae Systems Cente Pacific San Diego, CA, USA {bian.okon,joshua.haguess}@navy.mil Abstact

More information

Cryptanalysis of Hwang-Chang s a Time-Stamp Protocol for Digital Watermarking

Cryptanalysis of Hwang-Chang s a Time-Stamp Protocol for Digital Watermarking Cyptanalysis of Hwang-Chang s a Time-Stamp Potocol fo Digital Watemaking *Jue-Sam Chou, Yalin Chen 2, Chung-Ju Chan 3 Depatment of Infomation Management, Nanhua Univesity Chiayi 622 Taiwan, R.O.C *: coesponding

More information

3D Hand Trajectory Segmentation by Curvatures and Hand Orientation for Classification through a Probabilistic Approach

3D Hand Trajectory Segmentation by Curvatures and Hand Orientation for Classification through a Probabilistic Approach 3D Hand Tajectoy Segmentation by Cuvatues and Hand Oientation fo Classification though a Pobabilistic Appoach Diego R. Faia and Joge Dias Abstact In this wok we pesent the segmentation and classification

More information

Point-Biserial Correlation Analysis of Fuzzy Attributes

Point-Biserial Correlation Analysis of Fuzzy Attributes Appl Math Inf Sci 6 No S pp 439S-444S (0 Applied Mathematics & Infomation Sciences An Intenational Jounal @ 0 NSP Natual Sciences Publishing o Point-iseial oelation Analysis of Fuzzy Attibutes Hao-En hueh

More information

Robust Object Detection at Regions of Interest with an Application in Ball Recognition

Robust Object Detection at Regions of Interest with an Application in Ball Recognition Robust Object Detection at Regions of Inteest with an Application in Ball Recognition Saa Miti, Simone Fintop, Kai Pevölz, Hatmut Sumann Faunhofe Institute fo Autonomous Intelligent Systems (AIS) Schloss

More information

Compact Vectors of Locally Aggregated Tensors for 3D shape retrieval

Compact Vectors of Locally Aggregated Tensors for 3D shape retrieval Euogaphics Wokshop on 3D Object Retieval (2013) S. Biasotti, I. Patikakis, U. Castellani, and T. Scheck (Editos) Compact Vectos of Locally Aggegated Tensos fo 3D shape etieval Hedi Tabia 1, David Picad

More information

Robust Object Detection at Regions of Interest with an Application in Ball Recognition

Robust Object Detection at Regions of Interest with an Application in Ball Recognition Robust Object Detection at Regions of Inteest with an Application in Ball Recognition Saa Miti, Simone Fintop, Kai Pevölz, Hatmut Sumann Faunhofe Institute fo Autonomous Intelligent Systems (AIS) Schloss

More information

Augmented Reality. Integrating Computer Graphics with Computer Vision Mihran Tuceryan. August 16, 1998 ICPR 98 1

Augmented Reality. Integrating Computer Graphics with Computer Vision Mihran Tuceryan. August 16, 1998 ICPR 98 1 Augmented Reality Integating Compute Gaphics with Compute Vision Mihan Tuceyan August 6, 998 ICPR 98 Definition XCombines eal and vitual wolds and objects XIt is inteactive and eal-time XThe inteaction

More information

EYE DIRECTION BY STEREO IMAGE PROCESSING USING CORNEAL REFLECTION ON AN IRIS

EYE DIRECTION BY STEREO IMAGE PROCESSING USING CORNEAL REFLECTION ON AN IRIS EYE DIRECTION BY STEREO IMAGE PROCESSING USING CORNEAL REFLECTION ON AN IRIS Kumiko Tsuji Fukuoka Medical technology Teikyo Univesity 4-3-14 Shin-Katsutachi-Machi Ohmuta Fukuoka 836 Japan email: c746g@wisdomcckyushu-uacjp

More information

RANDOM IRREGULAR BLOCK-HIERARCHICAL NETWORKS: ALGORITHMS FOR COMPUTATION OF MAIN PROPERTIES

RANDOM IRREGULAR BLOCK-HIERARCHICAL NETWORKS: ALGORITHMS FOR COMPUTATION OF MAIN PROPERTIES RANDOM IRREGULAR BLOCK-HIERARCHICAL NETWORKS: ALGORITHMS FOR COMPUTATION OF MAIN PROPERTIES Svetlana Avetisyan Mikayel Samvelyan* Matun Kaapetyan Yeevan State Univesity Abstact In this pape, the class

More information

Input Layer f = 2 f = 0 f = f = 3 1,16 1,1 1,2 1,3 2, ,2 3,3 3,16. f = 1. f = Output Layer

Input Layer f = 2 f = 0 f = f = 3 1,16 1,1 1,2 1,3 2, ,2 3,3 3,16. f = 1. f = Output Layer Using the Gow-And-Pune Netwok to Solve Poblems of Lage Dimensionality B.J. Biedis and T.D. Gedeon School of Compute Science & Engineeing The Univesity of New South Wales Sydney NSW 2052 AUSTRALIA bbiedis@cse.unsw.edu.au

More information

Module 6 STILL IMAGE COMPRESSION STANDARDS

Module 6 STILL IMAGE COMPRESSION STANDARDS Module 6 STILL IMAE COMPRESSION STANDARDS Lesson 17 JPE-2000 Achitectue and Featues Instuctional Objectives At the end of this lesson, the students should be able to: 1. State the shotcomings of JPE standad.

More information

3D Periodic Human Motion Reconstruction from 2D Motion Sequences

3D Periodic Human Motion Reconstruction from 2D Motion Sequences 3D Peiodic Human Motion Reconstuction fom D Motion Sequences Zonghua Zhang and Nikolaus F. Toje BioMotionLab, Depatment of Psychology Queen s Univesity, Canada zhang, toje@psyc.queensu.ca Abstact In this

More information

Machine Learning for Automatic Classification of Web Service Interface Descriptions

Machine Learning for Automatic Classification of Web Service Interface Descriptions Machine Leaning fo Automatic Classification of Web Sevice Inteface Desciptions Amel Bennaceu 1, Valéie Issany 1, Richad Johansson 4, Alessando Moschitti 3, Romina Spalazzese 2, and Daniel Sykes 1 1 Inia,

More information

Tissue Classification Based on 3D Local Intensity Structures for Volume Rendering

Tissue Classification Based on 3D Local Intensity Structures for Volume Rendering 160 IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, VOL. 6, NO., APRIL-JUNE 000 Tissue Classification Based on 3D Local Intensity Stuctues fo Volume Rendeing Yoshinobu Sato, Membe, IEEE, Cal-Fedik

More information

Free Viewpoint Action Recognition using Motion History Volumes

Free Viewpoint Action Recognition using Motion History Volumes Fee Viewpoint Action Recognition using Motion Histoy Volumes Daniel Weinland 1, Remi Ronfad, Edmond Boye Peception-GRAVIR, INRIA Rhone-Alpes, 38334 Montbonnot Saint Matin, Fance. Abstact Action ecognition

More information

And Ph.D. Candidate of Computer Science, University of Putra Malaysia 2 Faculty of Computer Science and Information Technology,

And Ph.D. Candidate of Computer Science, University of Putra Malaysia 2 Faculty of Computer Science and Information Technology, (IJCSIS) Intenational Jounal of Compute Science and Infomation Secuity, Efficient Candidacy Reduction Fo Fequent Patten Mining M.H Nadimi-Shahaki 1, Nowati Mustapha 2, Md Nasi B Sulaiman 2, Ali B Mamat

More information

5 4 THE BERNOULLI EQUATION

5 4 THE BERNOULLI EQUATION 185 CHATER 5 the suounding ai). The fictional wok tem w fiction is often expessed as e loss to epesent the loss (convesion) of mechanical into themal. Fo the idealied case of fictionless motion, the last

More information

A Recommender System for Online Personalization in the WUM Applications

A Recommender System for Online Personalization in the WUM Applications A Recommende System fo Online Pesonalization in the WUM Applications Mehdad Jalali 1, Nowati Mustapha 2, Ali Mamat 2, Md. Nasi B Sulaiman 2 Abstact foeseeing of use futue movements and intentions based

More information

Topological Characteristic of Wireless Network

Topological Characteristic of Wireless Network Topological Chaacteistic of Wieless Netwok Its Application to Node Placement Algoithm Husnu Sane Naman 1 Outline Backgound Motivation Papes and Contibutions Fist Pape Second Pape Thid Pape Futue Woks Refeences

More information

Gravitational Shift for Beginners

Gravitational Shift for Beginners Gavitational Shift fo Beginnes This pape, which I wote in 26, fomulates the equations fo gavitational shifts fom the elativistic famewok of special elativity. Fist I deive the fomulas fo the gavitational

More information

Slotted Random Access Protocol with Dynamic Transmission Probability Control in CDMA System

Slotted Random Access Protocol with Dynamic Transmission Probability Control in CDMA System Slotted Random Access Potocol with Dynamic Tansmission Pobability Contol in CDMA System Intaek Lim 1 1 Depatment of Embedded Softwae, Busan Univesity of Foeign Studies, itlim@bufs.ac.k Abstact In packet

More information

Analysis of uniform illumination system with imperfect Lambertian LEDs

Analysis of uniform illumination system with imperfect Lambertian LEDs Optica Applicata, Vol. XLI, No. 3, 2011 Analysis of unifom illumination system with impefect Lambetian LEDs JIAJIE TAN 1, 2, KECHENG YANG 1*, MIN XIA 1, YING YANG 1 1 Wuhan National Laboatoy fo Optoelectonics,

More information

ANN Models for Coplanar Strip Line Analysis and Synthesis

ANN Models for Coplanar Strip Line Analysis and Synthesis 200 IJCSNS Intenational Jounal of Compute Science and Netwok Secuity, VOL.8 No.10, Octobe 2008 Models fo Coplana Stip Line Analysis and J.Lakshmi Naayana D.K.Si Rama Kishna D.L.Patap Reddy Chalapathi Institute

More information

Clustering Interval-valued Data Using an Overlapped Interval Divergence

Clustering Interval-valued Data Using an Overlapped Interval Divergence Poc. of the 8th Austalasian Data Mining Confeence (AusDM'9) Clusteing Inteval-valued Data Using an Ovelapped Inteval Divegence Yongli Ren Yu-Hsn Liu Jia Rong Robet Dew School of Infomation Engineeing,

More information

17/5/2009. Introduction

17/5/2009. Introduction 7/5/9 Steeo Imaging Intoduction Eample of Human Vision Peception of Depth fom Left and ight eye images Diffeence in elative position of object in left and ight eyes. Depth infomation in the views?? 7/5/9

More information

ADDING REALISM TO SOURCE CHARACTERIZATION USING A GENETIC ALGORITHM

ADDING REALISM TO SOURCE CHARACTERIZATION USING A GENETIC ALGORITHM ADDING REALISM TO SOURCE CHARACTERIZATION USING A GENETIC ALGORITHM Luna M. Rodiguez*, Sue Ellen Haupt, and Geoge S. Young Depatment of Meteoology and Applied Reseach Laboatoy The Pennsylvania State Univesity,

More information

Mono Vision Based Construction of Elevation Maps in Indoor Environments

Mono Vision Based Construction of Elevation Maps in Indoor Environments 8th WSEAS Intenational onfeence on SIGNAL PROESSING, OMPUTATIONAL GEOMETRY and ARTIFIIAL VISION (ISGAV 08) Rhodes, Geece, August 0-, 008 Mono Vision Based onstuction of Elevation Maps in Indoo Envionments

More information

Persistence-based Interest Point Detection for 3D Deformable Surface

Persistence-based Interest Point Detection for 3D Deformable Surface Pesistence-based Inteest Point Detection fo 3D Defomable Suface Xupeng Wang 1,3, Fedous Sohel 2, Mohammed Bennamoun 3, Yulan Guo 4,3 and Hang Lei 1 1 School of Infomation and Softwae Engineeing, Univesity

More information

THE SOLID IMAGE: a new concept and its applications

THE SOLID IMAGE: a new concept and its applications THE SOLID IMAGE: a new concept and its applications Leando Bonaz ( # ), Segio Dequal ( # ) ( # ) Politecnico di Toino - Dipatimento di Geoisose e Teitoio C.so Duca degli Abuzzi, 4 119 Toino Tel. +39.11.564.7687

More information

Concomitants of Upper Record Statistics for Bivariate Pseudo Weibull Distribution

Concomitants of Upper Record Statistics for Bivariate Pseudo Weibull Distribution Available at http://pvamuedu/aam Appl Appl Math ISSN: 93-9466 Vol 5, Issue (Decembe ), pp 8 9 (Peviously, Vol 5, Issue, pp 379 388) Applications and Applied Mathematics: An Intenational Jounal (AAM) Concomitants

More information

FACE VECTORS OF FLAG COMPLEXES

FACE VECTORS OF FLAG COMPLEXES FACE VECTORS OF FLAG COMPLEXES ANDY FROHMADER Abstact. A conjectue of Kalai and Eckhoff that the face vecto of an abitay flag complex is also the face vecto of some paticula balanced complex is veified.

More information

Lecture 3: Rendering Equation

Lecture 3: Rendering Equation Lectue 3: Rendeing Equation CS 660, Sping 009 Kavita Bala Compute Science Conell Univesity Radiomety Radiomety: measuement of light enegy Defines elation between Powe Enegy Radiance Radiosity 1 Hemispheical

More information

A Two-level Pose Estimation Framework Using Majority Voting of Gabor Wavelets and Bunch Graph Analysis

A Two-level Pose Estimation Framework Using Majority Voting of Gabor Wavelets and Bunch Graph Analysis Technical Repot pepaed fo the Pointing'04: Visual Obsevation of Diectic Gestues, ICPR Woshop, May, 004. A Two-level Pose Estimation Famewo Using Maoity Voting of Gabo Wavelets and Bunch Gaph Analysis Junwen

More information

Tier-Based Underwater Acoustic Routing for Applications with Reliability and Delay Constraints

Tier-Based Underwater Acoustic Routing for Applications with Reliability and Delay Constraints Tie-Based Undewate Acoustic Routing fo Applications with Reliability and Delay Constaints Li-Chung Kuo Depatment of Electical Engineeing State Univesity of New Yok at Buffalo Buffalo, New Yok 14260 Email:

More information

vaiation than the fome. Howeve, these methods also beak down as shadowing becomes vey signicant. As we will see, the pesented algoithm based on the il

vaiation than the fome. Howeve, these methods also beak down as shadowing becomes vey signicant. As we will see, the pesented algoithm based on the il IEEE Conf. on Compute Vision and Patten Recognition, 1998. To appea. Illumination Cones fo Recognition Unde Vaiable Lighting: Faces Athinodoos S. Geoghiades David J. Kiegman Pete N. Belhumeu Cente fo Computational

More information

A Memory Efficient Array Architecture for Real-Time Motion Estimation

A Memory Efficient Array Architecture for Real-Time Motion Estimation A Memoy Efficient Aay Achitectue fo Real-Time Motion Estimation Vasily G. Moshnyaga and Keikichi Tamau Depatment of Electonics & Communication, Kyoto Univesity Sakyo-ku, Yoshida-Honmachi, Kyoto 66-1, JAPAN

More information

Development and Analysis of a Real-Time Human Motion Tracking System

Development and Analysis of a Real-Time Human Motion Tracking System Development and Analysis of a Real-Time Human Motion Tacking System Jason P. Luck 1,2 Chistian Debunne 1 William Hoff 1 Qiang He 1 Daniel E. Small 2 1 Coloado School of Mines 2 Sandia National Labs Engineeing

More information

= dv 3V (r + a 1) 3 r 3 f(r) = 1. = ( (r + r 2

= dv 3V (r + a 1) 3 r 3 f(r) = 1. = ( (r + r 2 Random Waypoint Model in n-dimensional Space Esa Hyytiä and Joma Vitamo Netwoking Laboatoy, Helsinki Univesity of Technology, Finland Abstact The andom waypoint model (RWP) is one of the most widely used

More information

A Texture Feature Extraction Based On Two Fractal Dimensions for Content_based Image Retrieval

A Texture Feature Extraction Based On Two Fractal Dimensions for Content_based Image Retrieval 9 Wold Congess on Compute Science and nfomation Engineeing A Textue Featue Extaction Based On To Factal Dimensions fo Content_based mage Retieval Zhao Hai-ying Xu Zheng-guang Penghong (. College of Maths-physics

More information

3D Reconstruction from 360 x 360 Mosaics 1

3D Reconstruction from 360 x 360 Mosaics 1 CENTER FOR MACHINE PERCEPTION 3D Reconstuction fom 36 x 36 Mosaics CZECH TECHNICAL UNIVERSITY {bakstein, pajdla}@cmp.felk.cvut.cz REPRINT Hynek Bakstein and Tomáš Pajdla, 3D Reconstuction fom 36 x 36 Mosaics,

More information

Title. Author(s)NOMURA, K.; MOROOKA, S. Issue Date Doc URL. Type. Note. File Information

Title. Author(s)NOMURA, K.; MOROOKA, S. Issue Date Doc URL. Type. Note. File Information Title CALCULATION FORMULA FOR A MAXIMUM BENDING MOMENT AND THE TRIANGULAR SLAB WITH CONSIDERING EFFECT OF SUPPO UNIFORM LOAD Autho(s)NOMURA, K.; MOROOKA, S. Issue Date 2013-09-11 Doc URL http://hdl.handle.net/2115/54220

More information

Evaluation of Second-order Visual Features for Land-Use Classification

Evaluation of Second-order Visual Features for Land-Use Classification Evaluation of Second-ode Visual Featues fo Land-Use Classification Romain Negel, David Picad and Philippe-Heni Gosselin ETIS/ENSEA - UCP - CNRS, F95014 Cegy, Fance Email: omain.negel,picad,gosselin@ensea.f

More information

INDEXATION OF WEB PAGES BASED ON THEIR VISUAL RENDERING

INDEXATION OF WEB PAGES BASED ON THEIR VISUAL RENDERING INDEXATION OF WEB PAGES BASED ON THEIR VISUAL RENDERING Emmanuel Buno Univesité du Sud Toulon-Va / LSIS CNRS BP 20132, F-83957 La Gade buno@univ-tln.f Nicolas Faessel LSIS CNRS Domaine Univesitaie de Saint-Jéôme

More information

High performance CUDA based CNN image processor

High performance CUDA based CNN image processor High pefomance UDA based NN image pocesso GEORGE VALENTIN STOIA, RADU DOGARU, ELENA RISTINA STOIA Depatment of Applied Electonics and Infomation Engineeing Univesity Politehnica of Buchaest -3, Iuliu Maniu

More information

DEVELOPMENT OF A PROCEDURE FOR VERTICAL STRUCTURE ANALYSIS AND 3D-SINGLE TREE EXTRACTION WITHIN FORESTS BASED ON LIDAR POINT CLOUD

DEVELOPMENT OF A PROCEDURE FOR VERTICAL STRUCTURE ANALYSIS AND 3D-SINGLE TREE EXTRACTION WITHIN FORESTS BASED ON LIDAR POINT CLOUD IAPRS Volume XXXVI, Pat 3 / W52, 27 DEVELOPMENT OF A PROCEDURE FOR VERTICAL STRUCTURE ANALYSIS AND 3D-SINGLE TREE EXTRACTION WITHIN FORESTS BASED ON LIDAR POINT CLOUD Yunsheng Wang a,*, Holge Weinacke

More information

Shortest Paths for a Two-Robot Rendez-Vous

Shortest Paths for a Two-Robot Rendez-Vous Shotest Paths fo a Two-Robot Rendez-Vous Eik L Wyntes Joseph S B Mitchell y Abstact In this pape, we conside an optimal motion planning poblem fo a pai of point obots in a plana envionment with polygonal

More information

arxiv: v2 [physics.soc-ph] 30 Nov 2016

arxiv: v2 [physics.soc-ph] 30 Nov 2016 Tanspotation dynamics on coupled netwoks with limited bandwidth Ming Li 1,*, Mao-Bin Hu 1, and Bing-Hong Wang 2, axiv:1607.05382v2 [physics.soc-ph] 30 Nov 2016 1 School of Engineeing Science, Univesity

More information

A Mathematical Implementation of a Global Human Walking Model with Real-Time Kinematic Personification by Boulic, Thalmann and Thalmann.

A Mathematical Implementation of a Global Human Walking Model with Real-Time Kinematic Personification by Boulic, Thalmann and Thalmann. A Mathematical Implementation of a Global Human Walking Model with Real-Time Kinematic Pesonification by Boulic, Thalmann and Thalmann. Mashall Badley National Cente fo Physical Acoustics Univesity of

More information

A Minutiae-based Fingerprint Matching Algorithm Using Phase Correlation

A Minutiae-based Fingerprint Matching Algorithm Using Phase Correlation A Minutiae-based Fingepint Matching Algoithm Using Phase Coelation Autho Chen, Weiping, Gao, Yongsheng Published 2007 Confeence Title Digital Image Computing: Techniques and Applications DOI https://doi.og/10.1109/dicta.2007.4426801

More information

IP Multicast Simulation in OPNET

IP Multicast Simulation in OPNET IP Multicast Simulation in OPNET Xin Wang, Chien-Ming Yu, Henning Schulzinne Paul A. Stipe Columbia Univesity Reutes Depatment of Compute Science 88 Pakway Dive South New Yok, New Yok Hauppuage, New Yok

More information