Aggregating Local Descriptors for Epigraphs Recognition

Size: px
Start display at page:

Download "Aggregating Local Descriptors for Epigraphs Recognition"

Transcription

1 Aggegating Local Descitos fo Eigahs Recognition Giusee Amato, Fabizio Falchi, Fausto Rabitti, Lucia Vadicamo ISTI-CNR, via G. Mouzzi, 5624 Pisa (Italy) s: Abstact. In this ae, we conside the task of ecognizing eigahs in images such as hotos taken using mobile devices. Given a set of 7,55 hotos elated to 4,56 eigahs, we used a k-neaestneighbo aoach in ode to efom the ecognition. The contibution of this wok is in evaluating state-ofthe-at visual object ecognition techniques in this secific context. The exeimental esults conducted show that Vecto of Locally Aggegated Descitos obtained aggegating SIFT descitos is the best choice fo this task. Keywods: Eigahs Recognition, Object Recognition, Content-Base Image Retieval, Bag-of-Featues, VLAD. Intoduction Because of lage availability of digital cameas, esecially embedded in smathones and tablets, thee is a gowing demand of infomation etieval systems able to seach by using images as quey. The basic idea is to allow the use to make a hoto of the object she is inteested on in ode to eceive elated infomation. To his goal, the objects in the hoto ae comae with the ones stoed in a eositoy in ode to ecognize it and give back elevant metadata, links, etc. We conducted the wok eoted in this ae in the context of the Euoeana netwok of Ancient Geek and Latin Eigahy (EAGLE) CIP-Best Pactice Netwok. In EAGLE, an object is essentially an ancient eigah. In this wok, we focus on seaching fo the most simila eigahs with esect to the one eesented in a hoto made by the use. The dataset we used consists of 7,55 hotos elated to 4,56 eigahs that wee made available, within the EAGLE oject, by Saienza Univesity of Rome. This functionality will be integated on the flagshi mobile alication, to enable touists to undestand inscitions they find on location. The alication will allow a visito of a site whee one of the stoed eigahy is visible (museum, steet, achaeological site, inted eoduction, etc.) to take a ictue with a mobile hone, send the ictue to the cental eositoy and eceive back the eniched infomation associated with that ictue. Moeove, cowd-soucing functionalities will be develoed in the alication to acceleate the adotion of the EAGLE and eigahic content. Fo achieving the task of identifying objects in an image o video sequence, usually efeed to as object ecognition, eseach conducted in both Comute Vision and Multimedia Infomation Retieval fields has focused on local featues that ovide a Digital Pesentation and Pesevation of Cultual and Scientific Heitage, Vol. 4, 24, ISSN: 34-46

2 eesentation that allows matching local stuctues between images. Fist, distinctive key oints ae selected in each image. Second, a descition of the selected egions is given. Diect local featues matching has been oved to be vey effective in ecognizing the same objects in two hotos. Howeve, to achieve scalability (i.e., to be able to seach in lage datasets) aggegation techniques ae necessay in ode to summaize the infomation eoted fo each key oint. Taditionally, object ecognition has been successfully alied to consume oducts, buildings, monuments and landmaks. Howeve, we did not find any secific exeiments conducted on ancient eigahs. Moeove, state-of-the-at techniques ae vey effective on small sets (tens) of objects while aoximate techniques ae alied when millions of objects have to be ecognized. The numbe of hotos of eigahs exected in the context of the EAGLE oject is in the middle of these two extemes. Thus, it is vey inteesting to undestand what is the best technique fo ecognizing eigahs. In the following, we eot the esults obtained testing vaious state-of-the at techniques in ode to effectively ecognizing eigahs in hotos given a medium-lage scale set (tens of thousands) of known eigahs. The est of this ae is stuctued as follows. In Section 2, we discuss elated wok. In Section 3, we give infomation about the tested aoaches. Then, we secify the exeimental settings and discuss the obtained esults in Section 4. Finally, in Section 5, we eot conclusions and discuss futue wok. 2 Related Wok In the last few yeas, eseach on object ecognition has focused on local featues [8], []. Following this aoach, an image is eesented by descibing the visual content of tyically thousands of egions of inteest automatic selected. To achieve best effectiveness, images ae comaed by matching thei local featues and seaching fo a geometic tansfomation that can associate the egions of both images. In the last few yeas, the oblem of ecognizing cultual heitage elated objects, in aticula landmaks, has eceived gowing attention by the eseach community. As an examle, Google esented its aoach to building a web-scale landmak ecognition engine [3]. The oblem of landmak ecognition is tyically addessed by leveaging on techniques of automatic classification, as fo instances knn Classification [4], alied to image featues. Between 27 and 2, the VISITO Tuscany, (VIsual Suot to Inteactive TOuism in Tuscany) oject, has focused on technologies able to offe an inteactive and customized advanced tou guide sevice to visit the cities of at in Tuscany. This oject has investigated cultual heitage object ecognition, (such as monuments, landmaks, etc.) develoing a mobile alication and elated eseach aes such as [2]. Howeve, eigahs ecognition was out of the scoe of the VISITO Tuscany oject. htt:// 5

3 3 Tested Aoaches In this Section, we eot infomation about the tested aoaches. Fist we discuss the SIFT that we selected as local featue. Then we biefly discuss the Bag-of- Featues (BoF) and Vecto of Locally Aggegated Descitos (VLAD) that make use of the local featues (SIFT in ou case) in ode to achieve high efficiency and effectiveness via aggegation of the infomation they contain. 3. SIFT The Scale Invaiant Featue Tansfomation (SIFT) [7] ae extacted fom key oints selected using diffeence of Gaussians alied in scale sace to a seies of smoothed esamled images. The descition of the egion aound this selected oints ely on histogam of gadients. SIFT ae not only the most imotant and cited local featues eve defined, but they ae still almost unbeaten in tems of effectiveness. Recently, binay local featues have been oosed in ode to imove efficiency of diect local featues matching. In ou exeiments we achieve scalability aggegating featues and thus we ae moe inteested in effective eesentation of the images than efficient comaison of the featues themselves. 3.2 Bag-of-Featues The Bag-of-Featues (BoF) was initially oosed in [] and has been studied in many othe aes. The goal of the BoF aoach is to substitute each local descito of an image with visual wods obtained fom a edefined vocabulay in ode to aly taditional text etieval techniques to CBIR. The fist ste is selecting some visual wods ceating a vocabulay. The visual vocabulay is tyically built clusteing, using k-means, local descitos of the dataset and selecting the centoids. The second ste assigns each local descito to the identifie of the neaest wod in the vocabulay. At the end of the ocess, each image is descibed as a set of visual wods. The etieval hase is then efomed using text etieval techniques consideing a quey image as disjunctive text-quey. Tyically, the cosine similaity measue in conjunction with a tem weighting scheme (e.g., TF- IDF [9]) is adoted fo evaluating the similaity between any two images. As mentioned in [2], a fundamental diffeence between an image quey (e.g. 5 visual tems) and a text quey (e.g. 3 tems) is lagely ignoed in existing index design. Efficiency and memoy constaints have been ecently addessed by aggegating local descitos into a fixed-size vecto eesentation that descibe the whole image. In aticula, Fishe Vecto (FV) and VLAD have shown bette efomance than BoF. In this wok we will focus on VLAD which has been oved to be a simlified non-obabilistic vesion of FV [6]. Desite its simlicity, VLAD efomance is comaable to that of FV. 5

4 3.3 Vecto of Locally Aggegated Desctios (VLAD) The VLAD eesentation was oosed in [5]. As fo BoF, a codebook {μ,, μ k } is fist leaned using a cluste algoithm (e.g., k-means). Each local descito x t in each image is then associated to its neaest visual wod NN(x t ) in the codebook. Fo each codewod, the diffeences x t μ i of the vectos x t assigned to μ i ae accumulated: v i = x t μ i x t : NN(x t )=μ i. () The VLAD eesentation is the concatenation of the accumulated vectos, i.e. V = [v T v K T ]. Powe-law and L 2 nomalization ae usually alied and fo comaing two VLAD descition L 2 Euclidean distance has been oved to be effective. VLAD descitions have a high dimensionality. Pincial Comonent Analysis has been oosed to have a moe comact eesentation. 4 Exeiments 4. Dataset Being atne of the EAGLE oject, we had the ootunity to access a dataset of 7,55 hotos elated to 4,56 eigahs made available to us by Saienza Univesity of Roma. Fo ou exeiments, we also needed a gound tuth, i.e., hotos in which we want to automatically ecognize the eigah togethe with the actual eigah eesented in the image. We constucted this gound tuth selecting 7 hotos fom the whole dataset and emoving them fom the knowledge base. In othe wods, we emoved these quey hotos fom the ones that ae given to the comute in ode to undestand the visual content of each eigah. This was only ossible fo the eigahs that had moe than one hoto. We also caefully selected queies that could eesent the vaious tyes of eigahs. In Fig. we eot 5 quey examles togethe with the othe images fo the same object available in the dataset. 52

5 Fig.. Examles of quey images and associated images of the same eigah 4.2 Quality Measues In ode to ecognize the actual object in a quey image, we basically efom a visual similaity seach between all the images in the dataset. Thus, the main goal is to have one image of the same eigah as fist esult. Wheneve this is not the case, it is inteesting to undestand at which osition in the esult list the most visually simila hoto of the same object aeas. In fact, while in this ae we ae focusing on techniques able to scale u to the size of the dataset, taditional comute vision techniques could be alied on the esults obtained in ode to achieve bette effectiveness. Given this consideations, we decided to eot the obability of finding an image of the same object between the fist esults. Fo =, also equals the accuacy of a classifie that ecognizes the quey eigah as the most simila that have been found. Fo each technique, we eot the obability of finding an image containing the same eigah given as quey between the fist esults vaying between and. Results ae eoted with the values on a logaithmic scale. A moe common measue of effectiveness is mean-aveage Pecision (map). In this case, not only the fist elevant image but all the images associated with the quey ae consideed. This measue eveal how good is the aoach in eoting the elated images in the to ositions of the esult list. 53

6 4.3 Exeimental Setu We extacted SIFT fom images using the OenCV libay 2. The BoF and VLAD aoaches have been imlemented by the NeMIS gou of the ISTI-CNR in Java as at of ou Visual Infomation Retieval libay ublically available on GitHub 3. Given an image, thousands of local featues ae extacted. In ou case, we obtained an aveage of 59 SIFT e image. Howeve, the fact that some of them efe to bigge egions than othes allows to select a subset of local featues that ae in incile moe elevant [2]. Thus, in the exeiments we also tied to educe the numbe of local featues selecting only the most imotant ones u to about 25 local featues e image. In the following, we efe to this second aoach as educed-keyoints. We tested all the aoaches both on the whole extacted local featues and on the ones obtained filteing by egion size. 4.4 Results Bag-of-Featues 4k 2k k 5k k 4k 2k k 5k k Fig. 2. BoF, cos TF-IDF Fig. 3. BoF, cos TF-IDF, educed-keyoints Results in Fig. 2 have been obtained using the BoF featue aoach using the cosine TF-IDF similaity measue, vaying the size of the vocabulay between k and 4k. As exected, the lage the vocabulay the bette the esults. Howeve, diffeences between 2k o 4k ae only maginal. Thus, we did not tested lage vocabulaies. In Fig. 3 we eot the same tye of esults consideing only the most imotant local featues. Results show that the SIFT eduction is useful. 2 htt://oencv.og/ 3 htts://github.com/ffalchi/it.cn.isti.vi 54

7 VLAD Fig. 4. VLAD Fig. 5. VLAD, educed-keyoints In Fig. 4 and Fig. 5, we eot the esults obtained by the VLAD aoach vaying the size k of the codebook between 32 and 256 using all the extacted SIFT and the educed ones esectively. It is inteesting to see that the SIFT selection is useful when k is highe and is smalle. In the othe cases it can even decease the quality of the esults. Consideing that we ae moe inteested on small, which means having elevant images on the vey fist ositions, the oveall best esults ae the ones obtained fo k = 256 and educing numbe of local featues. d=6384 (no PCA) d=248 d=52 d=64 d=6384 (no PCA) d=248 d=52 d= Fig. 6. VLAD-PCA, k=28 Fig. 7. VLAD-PCA, k=28, educed-k We also tied to aly PCA on the VLAD vectos, esecially fo tying to educe the comlexity of the comaison. We then selected k = 28 and alied PCA in ode to educe the dimensionality of the vecto (6,384). Results ae eoted fo VLAD vecto educed to 248, 52 and 64 dimensions. Unfotunately, the dimensionality eduction significantly educes the quality of the esults. 55

8 Comaison In Table, we summaize the esults obtained, odeed with esect to the map quality measue. Only the best aoaches ae shown. In the fist column, we eot a bief text about the aoach. In the second column, the aveage numbe of SIFT consideed is shown (i.e., 235 when local featues eduction was alied and,59 othewise). The thid column eots the numbe of wods used. While the wods have been selected both fo BoF and VLAD using k-means, thei use is vey diffeent. Thus, in the bytes column, we comuted the aveage size in bytes of the esulting eesentation. As quality measues, we used the obability of having at least one elated image between the fist esults fo =,, and the map. In case we use these aoaches to ecognize the quey image elying on the neaest image in the dataset, the best aoach is the VLAD fo a codebook size of 256 and selecting the 25 most elevant local featues. In this case, the accuacy obtained is.69. The moe taditional BoF-cos TF-IDF aoach obtained good esults when a lage codebook (i.e., 4k) was used (as exected). It is inteesting to note that this aoach outefoms VLAD fo =,. Given that ecent woks as [3] have shown that VLAD can be moe efficiently indexed than BoF, still VLAD is efeable. Table. Comaison of esults obtained by the oveall best aoaches odeed by map avg codebook Aoach Bytes SIFTs size = = = map VLAD , BoF / cos TF-IDF 235 4, VLAD , BoF / cos TF-IDF 235 2, VLAD , VLAD , BoF / cos TF-IDF 235, VLAD , VLAD , VLAD-PCA (d'=52) , Conclusions and Futue Wok In this wok, we tested state-of-the-at object ecognition techniques on an eigahs dataset consisting of 7,55 hotos. The best accuacy was obtained by using the VLAD aoach that has been ecently oosed fo efoming object ecognition on a lage scale. The obtained accuacy was of.69, which is good consideing the difficulties of the task and the few images available fo each eigah in the dataset. In fact, the dataset consists of 7,55 hotos elated to 4,56 eigahs. This esults in most of the ei- 56

9 gahs been eesented by only one o two hotos. Howeve, we lan to imove this esults efoming e-anking of the images obtained using these scalable techniques efoming diect local featues matching. To this goal, we also eoted the obability of having a elevant images between the etieve images. The esults show that it is ossible to have a elevant image between etieved ones with obability.9 using the VLAD aoach with a codebook of size 256 and filteing the SIFT. Thus, we lan to ty binay local featues and othe techniques in ode to imove the obtained.69 accuacy u to the.9 obtainable, in theoy, by e-anking the obtained using VLAD. Bibliogahy. G. Amato, P. Bolettiei, F. Falchi and C. Gennao, "Lage Scale Image Retieval Using Vecto of Locally Aggegated Descitos," in Similaity Seach and Alications, vol. 899, N. Bisaboa, O. Pedeia and P. Zezula, Eds., Singe Belin Heidelbeg, 23, G. Amato, F. Falchi and C. Gennao, "Geometic consistency checks fo knn based image classification elying on local featues," in SISAP ': Fouth Intenational Confeence on Similaity Seach and Alications, SISAP 2, Liai Island, Italy, June 3 - July, 2, G. Amato, F. Falchi and C. Gennao, "On Reducing the Numbe of VisualWods in the Bag-of-Featues Reesentation," in VISAPP 23 - Poceedings of the Intenational Confeence on Comute Vision Theoy and Alications, S. Dudani, "The Distance-Weighted K-Neaest-Neighbou Rule," IEEE Tansactions on Systems, Man and Cybenetics, Vols. SMC-6(4), , H. Jégou, M. Douze, C. Schmid and P. Péez, "Aggegating local descitos into a comact image eesentation," in Comute Vision and Patten Recognition (CVPR), 2 IEEE Confeence on, H. Jégou, F. Peonnin, M. Douze, J. Sànchez, P. Péez and C. Schmid, "Aggegating local image descitos into comact codes," IEEE Tansactions on Patten Analysis and Machine Intelligence, Se D. G. Lowe, "Distinctive Image Featues fom Scale-Invaiant Keyoints," Intenational Jounal of Comute Vision, vol. 6, no. 2,. 9-, K. Mikolajczyk and C. Schmid, "A efomance evaluation of local descitos," Patten Analysis and Machine Intelligence, IEEE Tansactions on, vol. 27, no., , oct G. Salton and M. J. McGill, Intoduction to Moden Infomation Retieval, New Yok, NY, USA: McGaw-Hill, Inc., J. Sivic and A. Zisseman, "Video Google: A Text Retieval Aoach to Object Matching in Videos," in Poceedings of the Ninth IEEE Intenational Confeence on Comute Vision - Volume 2, Washington, DC, USA, 23.. T. Tuytelaas and K. Mikolajczyk, "Local invaiant featue detectos: a suvey," Found. Tends. Comut. Gah. Vis., vol. 3, no. 3, , X. Zhang, Z. Li, L. Zhang, W. Y. Ma and H. Y. Shum, "Efficient indexing fo lage scale visual seach," in Comute Vision, 29 IEEE 2th Intenational Confeence on,

10 3. Y. Zheng, M. Zhao, Y. Song, H. Adam, U. Buddemeie, A. Bissacco, F. Buche, T. S. Chua and H. Neven, "Tou the wold: Building a web-scale landmak ecognition," in CVPR, 29. Acknowledgement This wok was atially suoted by the Euoeana netwok of Ancient Geek and Latin Eigahy (EAGLE, gant ageement numbe: 32522) co-funded by the Euoean Commission within the ICT Policy Suot Pogamme. 58

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

Trinocular Stereo using Shortest Paths and the Ordering Constraint

Trinocular Stereo using Shortest Paths and the Ordering Constraint Tinocula Steeo using Shotest Paths and the Odeing Constaint Motilal Agawal and Lay S. Davis Deatment of Comute Science, Univesity of Mayland, College Pak, MD 20742, USA email: mla,lsd @umiacs.umd.edu Abstact

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

Information Retrieval. CS630 Representing and Accessing Digital Information. IR Basics. User Task. Basic IR Processes

Information Retrieval. CS630 Representing and Accessing Digital Information. IR Basics. User Task. Basic IR Processes CS630 Repesenting and Accessing Digital Infomation Infomation Retieval: Basics Thosten Joachims Conell Univesity Infomation Retieval Basics Retieval Models Indexing and Pepocessing Data Stuctues ~ 4 lectues

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

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

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

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

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

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

Web scale image retrieval using compact tensor aggregation of visual descriptors

Web scale image retrieval using compact tensor aggregation of visual descriptors Web scale image etieval using compact tenso aggegation of visual desciptos Romain Negel, David Picad, Philippe-Heni Gosselin To cite this vesion: Romain Negel, David Picad, Philippe-Heni Gosselin. Web

More information

3/1/18. Overview. Program Representations. Abstract Syntax Tree (AST) Eclipse JDT. Java Model. The Tree Structure of Java Project[2]

3/1/18. Overview. Program Representations. Abstract Syntax Tree (AST) Eclipse JDT. Java Model. The Tree Structure of Java Project[2] Oveview Pogam Reesentations Abstact Syntax Tee Eclise JDT Java Model Eclise JDT AST Contol Flow Gah Pogam Deendence Gah Points-to Gah Call Gah 2 Abstact Syntax Tee (AST) Ceated by the comile at the end

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

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

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

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

A New Free-form Deformation Through the Control of Parametric Surfaces

A New Free-form Deformation Through the Control of Parametric Surfaces A New Fee-fom Defomation Though the Contol of Paametic Sufaces Jieqing Feng Lizhuang Ma and Qunsheng Peng State Key Lab. of CAD&CG, Zhejiang Univesity Hangzhou 310027, P. R. of CHINA email: jqfeng@cad.zju.edu.cn

More information

The EigenRumor Algorithm for Ranking Blogs

The EigenRumor Algorithm for Ranking Blogs he EigenRumo Algoithm fo Ranking Blogs Ko Fujimua N Cybe Solutions Laboatoies N Copoation akafumi Inoue N Cybe Solutions Laboatoies N Copoation Masayuki Sugisaki N Resonant Inc. ABSRAC he advent of easy

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

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

Defining and Implementing Dynamic Semantics of Object Oriented High Level Petri Nets

Defining and Implementing Dynamic Semantics of Object Oriented High Level Petri Nets Defining and Imlementing Dynamic Semantics of Object Oiented High Level Peti Nets Maius Bezovan Faculty of Automation Comutes and Electonics, Univesity of Caiova 1100 Caiova, Romania Abstact. This ae deals

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

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

ESTIMATION OF INTERPOLATION ERROR IN DEMS USING STATISTICAL METHODS

ESTIMATION OF INTERPOLATION ERROR IN DEMS USING STATISTICAL METHODS ESTIMATION OF INTERPOLATION ERROR IN DEMS USING STATISTICAL METHODS Robet PAQUET, Austalia Key wods: DEM, inteolation eo, ALS, hotogammety, Delaunay tiangulation SUMMARY To wok with DEMs euies a knowledge

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

Integrated Dynamic Congestion Controller Andreas Pitsillides 1, Petros Ioannou 2, Loukas Rossides 1

Integrated Dynamic Congestion Controller Andreas Pitsillides 1, Petros Ioannou 2, Loukas Rossides 1 Integated Dynamic Congestion Contolle Andeas Pitsillides 1, Petos Ioannou 2, Loukas Rossides 1 Abstact--The gowing demand of comute usage equies efficient ways of managing netwok taffic in ode to avoid

More information

SYSTEM LEVEL REUSE METRICS FOR OBJECT ORIENTED SOFTWARE : AN ALTERNATIVE APPROACH

SYSTEM LEVEL REUSE METRICS FOR OBJECT ORIENTED SOFTWARE : AN ALTERNATIVE APPROACH I J C A 7(), 202 pp. 49-53 SYSTEM LEVEL REUSE METRICS FOR OBJECT ORIENTED SOFTWARE : AN ALTERNATIVE APPROACH Sushil Goel and 2 Rajesh Vema Associate Pofesso, Depatment of Compute Science, Dyal Singh College,

More information

A Novel Genetic Algorithm for Static Task Scheduling in Distributed Systems

A Novel Genetic Algorithm for Static Task Scheduling in Distributed Systems Intenational Jounal of Comute Theoy and Engineeing, Vol., No., Ail 200 A Novel Genetic Algoithm fo Static Task Scheduling in Distibuted Systems Ami Masoud Rahmani and Mojtaba Rezvani Abstact The static

More information

A New Finite Word-length Optimization Method Design for LDPC Decoder

A New Finite Word-length Optimization Method Design for LDPC Decoder A New Finite Wod-length Optimization Method Design fo LDPC Decode Jinlei Chen, Yan Zhang and Xu Wang Key Laboatoy of Netwok Oiented Intelligent Computation Shenzhen Gaduate School, Habin Institute of Technology

More information

Interesting Target Detection and Matching Algorithm Based on Phase Information

Interesting Target Detection and Matching Algorithm Based on Phase Information 009 Intenational Confeence on Machine Leaning and Comuting IPCSIT vol.3 0 0 IACSIT Pess Singaoe Inteesting Taget Detection and Matching Algoithm Based on Phase Infomation hitao Xiao Jun u Lei Geng Jianming

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

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

Assessment of query reweighing, by rocchio method in farsi information retrieval

Assessment of query reweighing, by rocchio method in farsi information retrieval Univesity of Wollongong Reseach Online Univesity of Wollongong in Dubai - Papes Univesity of Wollongong in Dubai 2008 Assessment of quey eweighing, by occhio metho in fasi infomation etieval F Sabooi Bu-Ali

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

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

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

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

a Not yet implemented in current version SPARK: Research Kit Pointer Analysis Parameters Soot Pointer analysis. Objectives

a Not yet implemented in current version SPARK: Research Kit Pointer Analysis Parameters Soot Pointer analysis. Objectives SPARK: Soot Reseach Kit Ondřej Lhoták Objectives Spak is a modula toolkit fo flow-insensitive may points-to analyses fo Java, which enables expeimentation with: vaious paametes of pointe analyses which

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

MapReduce Optimizations and Algorithms 2015 Professor Sasu Tarkoma

MapReduce Optimizations and Algorithms 2015 Professor Sasu Tarkoma apreduce Optimizations and Algoithms 2015 Pofesso Sasu Takoma www.cs.helsinki.fi Optimizations Reduce tasks cannot stat befoe the whole map phase is complete Thus single slow machine can slow down the

More information

Performance Optimization in Structured Wireless Sensor Networks

Performance Optimization in Structured Wireless Sensor Networks 5 The Intenational Aab Jounal of Infomation Technology, Vol. 6, o. 5, ovembe 9 Pefomance Optimization in Stuctued Wieless Senso etwoks Amine Moussa and Hoda Maalouf Compute Science Depatment, ote Dame

More information

Simulating Life of Virtual Plants, Fishes and Butterflies

Simulating Life of Virtual Plants, Fishes and Butterflies Simulating Life of Vitual Plants, Fishes and Butteflies 1 3 Simulating Life of Vitual Plants, Fishes and Butteflies Hansudi Nose, Daniel Thalmann Comute Gahics Lab, Swiss Fedeal Institute of Technology

More information

Getting the System Sizing and Performance Test Right

Getting the System Sizing and Performance Test Right Getting the System Sizing and Pefomance Test Right Steve Hazeltine - Sema Gou Systems Ltd. (now Oacle UK Ltd.) (Email shazelti@uk.oacle.co - Telehone +44 171 830 4444) Andew Johnston - Indeendent Consultant

More information

Quality-of-Content (QoC)-Driven Rate Allocation for Video Analysis in Mobile Surveillance Networks

Quality-of-Content (QoC)-Driven Rate Allocation for Video Analysis in Mobile Surveillance Networks Quality-of-Content (QoC)-Diven Rate Allocation fo Video Analysis in Mobile Suveillance Netwoks Xiang Chen, Jenq-Neng Hwang, Kuan-Hui Lee, Ricado L. de Queioz Depatment of Electical Engineeing, Univesity

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

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

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

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

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

BUPT at TREC 2006: Spam Track

BUPT at TREC 2006: Spam Track BUPT at TREC 2006: Spam Tac Zhen Yang, Wei Xu, Bo Chen, Weian Xu, and Jun Guo PRIS Lab, School of Infomation Engineeing, Beijing Univesity of Posts and Telecommunications, 100876, Beijing, China yangzhen@pis.edu.cn

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

Linear Ensembles of Word Embedding Models

Linear Ensembles of Word Embedding Models Linea Ensembles of Wod Embedding Models Avo Muomägi Univesity of Tatu Tatu, Estonia avom@ut.ee Kaiit Sits Univesity of Tatu Tatu, Estonia kaiit.sits@ut.ee Sven Lau Univesity of Tatu Tatu, Estonia swen@math.ut.ee

More information

Effective Data Co-Reduction for Multimedia Similarity Search

Effective Data Co-Reduction for Multimedia Similarity Search Effective Data Co-Reduction fo Multimedia Similaity Seach Zi Huang Heng Tao Shen Jiajun Liu Xiaofang Zhou School of Infomation Technology and Electical Engineeing The Univesity of Queensland, QLD 472,

More information

An Assessment of the Efficiency of Close-Range Photogrammetry for Developing a Photo-Based Scanning Systeminthe Shams Tabrizi Minaret in Khoy City

An Assessment of the Efficiency of Close-Range Photogrammetry for Developing a Photo-Based Scanning Systeminthe Shams Tabrizi Minaret in Khoy City Austalian Jounal of Basic and Applied Sciences, 5(1): 80-85, 011 ISSN 1991-8178 An Assessment of the Efficiency of Close-Range Photogammety fo Developing a Photo-Based Scanning Systeminthe Shams Tabizi

More information

Conservation Law of Centrifugal Force and Mechanism of Energy Transfer Caused in Turbomachinery

Conservation Law of Centrifugal Force and Mechanism of Energy Transfer Caused in Turbomachinery Poceedings of the 4th WSEAS Intenational Confeence on luid Mechanics and Aeodynamics, Elounda, Geece, August 1-3, 006 (pp337-34) Consevation Law of Centifugal oce and Mechanism of Enegy Tansfe Caused in

More information

User Group testing report

User Group testing report Use Goup testing epot Deliveable No: D6.10 Contact No: Integated Poject No. 506723: SafetyNet Aconym: SafetyNet Title: Building the Euopean Road Safety Obsevatoy Integated Poject, Thematic Pioity 6.2 Sustainable

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

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

Scaling Location-based Services with Dynamically Composed Location Index

Scaling Location-based Services with Dynamically Composed Location Index Scaling Location-based Sevices with Dynamically Composed Location Index Bhuvan Bamba, Sangeetha Seshadi and Ling Liu Distibuted Data Intensive Systems Laboatoy (DiSL) College of Computing, Geogia Institute

More information

Collision Detection with Swept Spheres and Ellipsoids

Collision Detection with Swept Spheres and Ellipsoids Collision etection with Swet Shees and Ellisoids Joit Rouwé joit@games.lostbos.com Souce code: htt://www.thee4.demon.nl/swetellisoid/swetellisoid.zi. Intoduction Toda most games use conex olgons fo collision

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

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

Parallel processing model for XML parsing

Parallel processing model for XML parsing Recent Reseaches in Communications, Signals and nfomation Technology Paallel pocessing model fo XML pasing ADRANA GEORGEVA Fac. Applied Mathematics and nfomatics Technical Univesity of Sofia, TU-Sofia

More information

Effective Missing Data Prediction for Collaborative Filtering

Effective Missing Data Prediction for Collaborative Filtering Effective Missing Data Pediction fo Collaboative Filteing Hao Ma, Iwin King and Michael R. Lyu Dept. of Compute Science and Engineeing The Chinese Univesity of Hong Kong Shatin, N.T., Hong Kong { hma,

More information

Pledge: Signature:

Pledge: Signature: S/PM 0 Final Exam 7 May 005 Name: KEY E-mail ID: @viginia.edu Pledge: Signatue: Thee ae 80 minutes 3 hous fo this exam and 80 oints on the test; don t send too long on any one uestion! Thee is an exam

More information

Communication vs Distributed Computation: an alternative trade-off curve

Communication vs Distributed Computation: an alternative trade-off curve Communication vs Distibuted Computation: an altenative tade-off cuve Yahya H. Ezzeldin, Mohammed amoose, Chistina Fagouli Univesity of Califonia, Los Angeles, CA 90095, USA, Email: {yahya.ezzeldin, mkamoose,

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

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

Siggraph Precomputed Radiance Transfer: Theory and Practice

Siggraph Precomputed Radiance Transfer: Theory and Practice Siggah 2005 Pecomuted Radiance Tansfe: Theoy and Pactice Summay Geneal model of shading and shadowing fo eal-time endeing. Basic adiance tansfe techniques, moe advanced techniques that incooate highe-fequency

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

SCALABLE ENERGY EFFICIENT AD-HOC ON DEMAND DISTANCE VECTOR (SEE-AODV) ROUTING PROTOCOL IN WIRELESS MESH NETWORKS

SCALABLE ENERGY EFFICIENT AD-HOC ON DEMAND DISTANCE VECTOR (SEE-AODV) ROUTING PROTOCOL IN WIRELESS MESH NETWORKS SCALABL NRGY FFICINT AD-HOC ON DMAND DISTANC VCTOR (S-AODV) ROUTING PROTOCOL IN WIRLSS MSH NTWORKS Sikande Singh Reseach Schola, Depatment of Compute Science & ngineeing, Punjab ngineeing College (PC),

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

LaSaS: an Aggregated Search based Graph Matching Approach

LaSaS: an Aggregated Search based Graph Matching Approach LaSaS: an ggegated Seach based Gaph Matching ppoach Ghizlane EHBRTHI Univesité de Lyon, NRS Univesité Lyon 1, LIRIS, UMR5205, F-69622, Fance Email: ghizlane.echbathi@univ-lyon1.f Hamamache KHEDDOUI Univesité

More information

Prediction of Time Series Using RBF Neural Networks: A New Approach of Clustering

Prediction of Time Series Using RBF Neural Networks: A New Approach of Clustering 138 The Intenational Aab Jounal of Infomation Technology, Vol. 6,. 2, Apil 2009 Pediction of Time Seies Using RBF Neual Netwoks: A New Appoach of Clusteing Mohammed Awad 2, Hécto Pomaes 1, Ignacio Rojas

More information

Performance Evaluation of TCP Vegas over Optical Burst Switched Networks

Performance Evaluation of TCP Vegas over Optical Burst Switched Networks Pefomae Evaluation of TCP Vegas ove Otical Bust Switched Netwoks Basem Shihada 1, Qiong Zhang 3, Pin-Han Ho 1,, 1 School of Comute Sciee, Univesity of ateloo, ateloo, Canada Deatment of Electical and Comute

More information

Experimental and numerical simulation of the flow over a spillway

Experimental and numerical simulation of the flow over a spillway Euopean Wate 57: 253-260, 2017. 2017 E.W. Publications Expeimental and numeical simulation of the flow ove a spillway A. Seafeim *, L. Avgeis, V. Hissanthou and K. Bellos Depatment of Civil Engineeing,

More information

Quality Aware Privacy Protection for Location-based Services

Quality Aware Privacy Protection for Location-based Services In Poceedings of the th Intenational Confeence on Database Systems fo Advanced Applications (DASFAA 007), Bangkok, Thailand, Apil 9-, 007. Quality Awae Pivacy Potection fo Location-based Sevices Zhen Xiao,,

More information

Feature Extraction Technique for Neural Network Based Pattern Recognition

Feature Extraction Technique for Neural Network Based Pattern Recognition Ashoka H.N et al. / Intenational Jounal on Compute Science and Engineeing (IJCSE) Featue Extaction Technique fo Neual Netwok Based Patten Recognition Ashoka H.N. Depatment of Electical and Electonics Engineeing,

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

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

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

BLUEPRINTS. Getting Ready. The Activity. Overview. Introducing

BLUEPRINTS. Getting Ready. The Activity. Overview. Introducing BLUEPRINTS GEOMETRY LOGIC Satial visualization Tansfomational eomety Gettin Reay What You ll Nee Cuisenaie Ros, 1 set e ai 1-centimete i ae, ae 99 Tansaent tae an/o lue sticks 3" x 5" cas onto which 1-centimete

More information

arxiv: v2 [cs.ro] 11 Apr 2018

arxiv: v2 [cs.ro] 11 Apr 2018 c 28 IEEE. Pesonal use of this mateial is pemitted. Pemission fom IEEE must be obtained fo all othe uses, in any cuent o futue media, including epinting / epublishing this mateial fo advetising o pomotional

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

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

E.g., movie recommendation

E.g., movie recommendation Recommende Systems Road Map Intodction Content-based ecommendation Collaboative filteing based ecommendation K-neaest neighbo Association les Matix factoization 2 Intodction Recommende systems ae widely

More information

On Error Estimation in Runge-Kutta Methods

On Error Estimation in Runge-Kutta Methods Leonado Jounal of Sciences ISSN 1583-0233 Issue 18, Januay-June 2011 p. 1-10 On Eo Estimation in Runge-Kutta Methods Ochoche ABRAHAM 1,*, Gbolahan BOLARIN 2 1 Depatment of Infomation Technology, 2 Depatment

More information

An Optimised Density Based Clustering Algorithm

An Optimised Density Based Clustering Algorithm Intenational Jounal of Compute Applications (0975 8887) Volume 6 No.9, Septembe 010 An Optimised Density Based Clusteing Algoithm J. Hencil Pete Depatment of Compute Science St. Xavie s College, Palayamkottai,

More information

Link Prediction in Heterogeneous Networks Based on Tensor Factorization

Link Prediction in Heterogeneous Networks Based on Tensor Factorization Send Odes fo Repints to epints@benthamscience.ae 36 The Open Cybenetics & Systemics Jounal, 204, 8, 36-32 Open Access Link Pediction in Heteogeneous Netwoks Based on Tenso Factoization Piao Yong,2*, Li

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

INFORMATION DISSEMINATION DELAY IN VEHICLE-TO-VEHICLE COMMUNICATION NETWORKS IN A TRAFFIC STREAM

INFORMATION DISSEMINATION DELAY IN VEHICLE-TO-VEHICLE COMMUNICATION NETWORKS IN A TRAFFIC STREAM INFORMATION DISSEMINATION DELAY IN VEHICLE-TO-VEHICLE COMMUNICATION NETWORKS IN A TRAFFIC STREAM LiLi Du Depatment of Civil, Achitectual, and Envionmental Engineeing Illinois Institute of Technology 3300

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

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

(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

Congestion Control for Differentiated-Services using Non-linear Control Theory

Congestion Control for Differentiated-Services using Non-linear Control Theory Congestion Contol fo Diffeentiated-Sevices using Non-linea Contol Theoy Andeas Pitsillides 1, Petos Ioannou 2, Loukas Rossides 1 Deatment of Comute Science, Univesity of Cyus, andeas.itsillides@ucy.ac.cy,

More information

DUe to the recent developments of gigantic social networks

DUe to the recent developments of gigantic social networks Exploing Communities in Lage Pofiled Gaphs Yankai Chen, Yixiang Fang, Reynold Cheng Membe, IEEE, Yun Li, Xiaojun Chen, Jie Zhang 1 Abstact Given a gaph G and a vetex q G, the community seach (CS) poblem

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

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

Multidimensional Testing

Multidimensional Testing Multidimensional Testing QA appoach fo Stoage netwoking Yohay Lasi Visuality Systems 1 Intoduction Who I am Yohay Lasi, QA Manage at Visuality Systems Visuality Systems the leading commecial povide of

More information