IMPROVED JPEG DECOMPRESSION OF DOCUMENT IMAGES BASED ON IMAGE SEGMENTATION. Tak-Shing Wong, Charles A. Bouman, and Ilya Pollak

Size: px
Start display at page:

Download "IMPROVED JPEG DECOMPRESSION OF DOCUMENT IMAGES BASED ON IMAGE SEGMENTATION. Tak-Shing Wong, Charles A. Bouman, and Ilya Pollak"

Transcription

1 IMPROVED DECOMPRESSION OF DOCUMENT IMAGES BASED ON IMAGE SEGMENTATION Tak-Shing Wong, Charle A. Bouman, and Ilya Pollak School of Electrical and Computer Engineering Purdue Univerity ABSTRACT We propoe a decompreion algorithm for document baed on egmentation. Our egmentation algorithm claifie each block of the into one of three clae: text, background, or picture. We develop a different decoding trategy for each cla and apply thi trategy to every block in thi cla. Our experiment demontrate that thi approach can improve the quality of decompreion. 1. INTRODUCTION The algorithm i till one of the mot prevalent compreion algorithm today. Further, a large number of have been compreed with the algorithm and archived in variou document and databae. Thee conideration motivate reearch on improving the quality of decompreion. We how that, for the cla of document, the quality of the decompreed can be improved ignificantly by applying different decompreion trategie over region with different characteritic. We propoe partitioning a document into text, background, and picture block, and uing different decompreion algorithm for each cla. Section 2 i a review of the baeline tandard. Section 3 i an overview of our algorithm. Section 4 decribe the egmentation algorithm we propoe for claifying each block. Section 5 introduce the three decompreion trategie ued for the three different clae. 2. REVIEW OF THE BASELINE STANDARD With compreion [1], an achromatic i partitioned into 8 8 block. Each block then undergoe the forward dicrete coine tranform (FDCT), quantization, and entropy encoding (Fig. 1). Let X be a column vector of the 64 pixel of the 8 8 block. The FDCT i an orthogonal tranform given by Y = BX, where B i a matrix whoe column are pairwie orthogonal. The firt element Y,0 of Y i called the DC coefficient of the block; the other element are the AC coefficient. The next tage ue a quantizationhmatrix, to perform element-wie uniform quantization Ỹ,i = Y,i i i, where [ ] i the integer-round operator. Finally the quantized DCT coefficient are packed into the bit-tream by entropy encoding. For a color, thi operation equence i applied to each color channel independently. Uually the i firt converted to the YC b C r color pace before compreion. Baeline decompreion revere the operation of compreion. The compreed i firt entropy decoded. Since entropy encoding i a lole operation, the quantized DCT Thi work wa upported in part by a grant from the Xerox Foundation. input X compreed compreed FDCT Entropy Decoding Y Ỹ uantization De quantization Ỹ Entropy Encoding Baeline Compreion Ỹ IDCT Baeline Decompreion Fig. 1. Baeline compreion and decompreion Image block egmentation Background block decoding Text/graphic block decoding Picture block decoding Combine reult Fig. 2. Overview of the cheme compreed Decompreed X decompreed coefficient Ỹ are retrieved perfectly. De-quantization then etimate the original true DCT coefficient (Y,i) byỹ,i = iỹ,i. Finally, invere dicrete coine tranform (IDCT) produce the decoded block a X = B t Ỹ. 3. OVERVIEW OF THE PROPOSED ALGORITHM Fig. 2 how a block diagram of our cheme. The input i firt egmented uing the 8 8 block of the into three clae: (i) text block, (ii) background block, and (iii) picture block. Thee three clae repreent region with very different characteritic. Accordingly, different technique are employed for decoding block of each cla. For a color, the decoding tage i applied to each color channel independently. Text block cover the region of text and graphic which uually contain many harp edge. The pixel in text block are typically either the background or the foreground. Thee block uffer mot everely from the ringing artifact of becaue they have a high level of high frequency content. Background block include the background of the document and mooth region of natural. They have almot no high frequencie. The low variation of the intenity in thee region make them very uceptible to the blocking artifact of X/07/$ IEEE 488 SSP 2007

2 NO Feature vector of all block block Compute AC energy E E <T AC? Compute 2 D feature vector [D,1,D,2] SMAP egmentation YES Text/picture block Picture cluter Text block S txt Picture block S pic Feature vector of text/picture block K mean clutering GMM fit for picture block Background block S bkg Fig. 3. Block-baed Segmentation Text cluter GMM fit for text block The remaining picture block conit of non-mooth region of natural. They uffer from both the ringing and blocking artifact. However, a noted in [2], ignificant high-frequency content in highly textured block make the artifact le noticeable. 4. BLOCK-BASED SEGMENTATION To exploit the unique characteritic of each cla of block, a blockbaed unupervied egmentation algorithm (Fig. 3) i firt applied to the input to partition the et of block S into the three clae: S bkg for background block; S txt for text block; S pic for picture block. Firt, background block are differentiated from the other block by computing the AC energy in each block and threholding. Next, a 2-D feature vector i calculated for each block uing the luminance channel. The feature vector of text and picture block are clutered into two group by k-mean clutering [3]. To make ue of patial regularity, each group of feature vector i fit with a Gauian mixture model (GMM) whoe order i determined by the minimum decription length criterion [4]. Finally, the two model for text and picture block are ued to egment the feature vector by the SMAP egmentation algorithm [5]. The reult of the SMAP algorithm i combined with the background block detected in the AC energy threholding tage to produce the final egmentation map Background Block Detection For an achromatic, the AC energy of the block i defined a E = P 63 i=1 Ỹ,i 2 where Ỹ,i i the etimate of the i-th DCT coefficient of the block, produced by decompreion. E i compared with a mall poitive threhold T AC. IfE <T AC, the block i claified a a background block. For a color, the AC energy i calculated for each of the three color channel of the block. The maximum i defined a E and compared with T AC Feature Vector The claification of the non-background block intotext and picture block i baed on a 2-D feature vector computed from the luminance channel. It wa reported that the code length of text block after entropy encoding tend to be longer than non-text block due to the higher level of high frequency content in thee block [6],[7]. Thu the firt feature i baed on the encoding length and computed a [7]! 63 f(ỹ,0 Ỹ 1,0)+ X f(ỹ,i) D,1 = 1 (1) 64 i=1 j log2 ( x )+4 if x > 1 where f(x) = 0 otherwie The econd feature meaure how cloe a block i to being a twocolor block. For each block, we perform a two-color projection a follow with the luminance channel decoded by conventional. The luminance value from a window centered at the block are firt clutered into two group by k-mean clutering, with mean denoted by θ,1 and θ,2. The two-color projection i formed by clipping each luminance of each pixel to the mean of the cluter to which the luminance value belong. Thi two-color projection i the bet repreentation of the block with two color in the MMSE ene. The l 2 ditance between the luminance of the block and it two-color projection i then a meaure of how cloely the block reemble a two-color block. We normalize thi projection error by the quare of the difference of the two etimated mean, θ,1 θ,2 2, o that a high contrat block ha a higher chance to be claified a a text block. The econd feature i then calculated a 1 X63 D,2 = θ,1 θ X,i X,i 2 (2),2 2 where X,i i the etimate of the i-th pixel of the block, produced by decompreion, and X,i i the value of the i-th pixel of the two-color projection. If θ,1 = θ,2, wedefine D,2 = Text and Picture Block Claification In the lat tage, the feature vector of non-background block are clutered into two group by k-mean clutering. For the norm ued in k-mean clutering, the contribution of the two feature are differently weighted. Specifically, for a vector D t = [D 1,D 2], the norm i calculated by D = p D1 2 + γd2 2. In our experiment, we et γ =15. The reulting cluter with a higher mean of D,1 and a lower mean of D,2 repreent the cla of text block, and the other cluter repreent the picture block. Each cluter i fit with a Gauian mixture model. The two model are then employed within the SMAP egmentation algorithm [5] in order to claify each non-background block a either a text block or a picture block. 5. IMAGE BLOCK DECODING decoding can be poed a an invere problem. Becaue quantization i a many-to-one tranform, many poible block X can produce the ame de-quantized DCT coefficient Ỹ at the decoder. The non-uniquene of the olution of recontructing the original block X baed on the de-quantized DCT coefficient Ỹ make thi problem ill-poed. We regularize the problem by auming a prior model for the original block. We then decode each block by computing the maximum a poteriori (MAP) etimate. 489

3 Let T ( ) be the quantization tranform o that T (Y )=Ỹ. The invere B 1 T 1 (Ỹ) ={x R64 : T (Bx) =Ỹ} i the et of block which encode to Ỹ. The forward model i: ( pỹ X (Ỹ x) = 1 if x B 1 T 1 (Ỹ) (3) 0 otherwie If the original block X follow the prior ditribution p X (x), the MAP etimate of X i ˆX =argmax{p X Ỹ(x Ỹ)} x R 64 (4) = argmin x B 1 T 1 (Ỹ) { log p X (x)} The contraint x B 1 T 1 (Ỹ), however, i difficult to enforce. A the DCT i an orthogonal tranform, intead of ˆX,we will be olving for the MAP etimate of the DCT of ˆX a follow: Ŷ = argmin y T 1 (Ỹ) { log p X (B 1 y)} (5) It hould be noted that the contraint now become the impler pointwie contraint y i Ỹ,i i/2 for i = The ignificance of egmentation manifet itelf here in that it allow u to apply a different prior model for each cla of block. The prior model choen for each cla capture the unique characteritic of the cla and conequently produce a decoded block that i of higher quality Text Block Decoding We oberve that pixel in a text block typically concentrate around two value of the foreground and the background color, and chooe a prior model reflecting thi. Let θ,1 and θ,2 be thee two value of the text block, θ,1 <θ,2. The bimodal ditribution of the pixel i modeled a 63X p θ (x 1 )= z exp{ 1 min( x 2σ 2,i θ,1 2, x,i θ,2 2 )} (6) where θ =[θ,1,θ,2] are the parameter of the ditribution, z i a normalization contant, and σ 2 i related to the variance of each cluter. The MAP etimate of the pixel X in each block then depend on the unknown parameter vector θ. We ue the maximum likelihood etimate of θ intead of it true value, to yield the following MAP etimate of X : ˆX = argmin x B 1 T 1 n 1 X63 o min( x,i θ,1 2, x,i θ,2 2 ) θ 2σ 2 (Ỹ) min (7) Intuitively, we eek a et of value for an block x B 1 T 1 (Ỹ) and two color θ,1 and θ,2 o that each pixel value i a cloe a poible to one of the two color. While thi decoding cheme produce reaonably good reult, we have found that the etimation of the two color i not accurate for ome text block. Thi happen when, for example, the majority of the pixel are in the foreground, leaving very few background pixel for a good etimate of the background color. To improve the accuracy, we oberve that two neighboring text block generally have imilar foreground and background color. Alo, a background block next to a text block i uually a continuation of the background of the text. Baed on thee obervation, we apply patial regularization to the etimation of the two color of each text block. Specifically, we let ρ( ) be a monotone function, and penalize ρ( θ r,1 θ,1 )+ρ( θ r,2 θ,2 ) (8) for every pair {r, } of neighbor text block. (We ue neighborhood of ize eight.) We ue the convention that θ,1 <θ,2 for all block, o that our penalty enforce imilarity between the lighter color of two neighbor block, and between the darker color of two neighbor block. The pecific function ρ that we ue in our experiment i given by ρ(x) =min(x 2,T 2 ) where the threhold i T = 40. Thi threhold i impoed on the growth of the penalty function in order to account for the poibility that two neighboring text region may come from different piece of text and therefore may have very different background and/or foreground color. In thi ituation, we would like to avoid impoing an exceive penalty. In addition, for every pair of neighboring block {r, } uch that i a text block and r i a background block, we aume that it i very likely that the background color of the two block are imilar. We etimate the color of the background block r from it etimated DC coefficient a μ r = Ỹr,0/8, and penalize the difference between thi color and the one of the two color of the text block that i cloer to μ r,byuing the following penalty: min(ρ(μ r θ,1),ρ(μ r θ,2)). (9) Combining the two new penalty term of Eq. (8,9) with the cot function of Eq. (7) and recating the cot in the DCT tranform domain, we obtain the following global cot function: 63X C = X min( (B t y ) i θ,1 2, (B t y ) i θ,2 2 ) +w X ˆρ(θr,1 θ,1)+ρ(θ r,2 θ,2) r, +w X r, min(ρ(μ r θ,1),ρ(μ r θ,2)) (10) where w>0i a weight factor, (B t y ) i i the i-th element of the intenity vector x = B t y for the block,and the firt ummation i performed over all text block ; the econd ummation i performed over all the pair of neighboring text block and r; the third ummation i performed over all the pair of neighboring block and r uch that i a text block and r i a background block. Thi cot mut be minimized with repect to all the DCT coefficient y of all the text block and all the parameter θ of all the text block. Thi minimization problem i non-convex and large, and for thee reaon i difficult to olve. To compute an approximate olution, we ue the iterative coordinate decent (ICD) algorithm. The ICD algorithm allow u to ue a local cot function for which it i eaier to contruct the optimization update. Our experiment how that the reulting two-color decoding trategy decode the text block at high quality. The ICD algorithm ucceively optimize the cot function with repect to one variable at a time. In our cheme, one full iteration of the ICD algorithm conit of two phae. In the firt phae, called block update, the cot function i optimized with repect to the DCT coefficient of all the text block. In the econd phae, called parameter update, the cot function i optimized with repect to the 490

4 parameter θ of all the text block. The DCT coefficient y of all the text block are initialized with the de-quantized coefficient Ỹ. The parameter θ of a text block i initialized with the mean of the two cluter reulting from k-mean clutering of X. The ICD iteration i repeated until the change in the cot function between two ucceive iteration i maller than a predetermined threhold. During the block update tage, the cot function i optimized with repect to one DCT coefficient at a time. The update of the k-th DCT coefficient of the text block i guided by y,k. Depending on the ign of y,k, y,k i adjuted until y,k =0or y,k reache either one of the end point of the contraint interval [Ỹ,k i/2, Ỹ,k + i/2]. During the parameter update tage, the cot function i optimized with repect to one parameter at a time. Similar to the block update tage, θ,1 i adjuted according to the ign of θ,1 =0. θ,1 until 5.2. Background Block and Picture Block Decoding Background block include both the background of the document and mooth region in natural. For thee block, mot of the information i captured by the DC coefficient. To mooth out the dicontinuitie acro block boundarie, a continuou Gauian Markov random field (GMRF) [8], [9] i choen a the prior model for the DC coefficient of the background block. AGMRFipecified completely by the neighborhood ytem and the prediction filter. In our application, we chooe a 8-point neighborhood ytem, and a patial invariant prediction filter h with coefficient given by h,r =1/6if r and are vertical or horizontal neighbor block, and h,r =1/12 if r and are diagonal neighbor block. The MAP etimate of the background block i again computed uing the ICD algorithm. For picture block, we ue the conventional decoding. Tranaction on Circuit and Sytem for Video Technology, vol. 9, no. 3, pp , Apr [3] J. Mcueen, Some method for claification and analyi of multivariate obervation, Proceeding of the Fifth Berkeley Sympoium on Mathematical Statitic and Probability, pp , [4] C. A. Bouman, Cluter: An unupervied algorithm for modeling Gauian mixture, Available from bouman, April [5] Charle A. Bouman and Michael Shapiro, A multicale random field model for Bayeian egmentation, IEEE Tranaction on Image Proceing, vol. 3, no. 2, pp , Mar [6] Ricardo L. de ueiroz, Proceing -compreed and document, IEEE Tranaction on Image Proceing, vol. 7, no. 12, pp , Dec [7] K. Kontantinide and D. Tretter, A variable quantization method for compound document, IEEE Tranaction on Image Proceing, vol. 9, no. 7, pp , Jul [8] Julian Beag, On the tatitical analyi of dirty picture, Journal of the Royal Statitical Society, vol. 48, no. 3, pp , [9] Julian Beag, Spatial interaction and the tatitical analyi of lattice ytem, Journal of the Royal Statitical Society, vol. 36, no. 2, pp , RESULTS AND CONCLUSION Fig. 5(a) how an converted from PDF at 300dpi. Fig. 4(b) and 6(b) how enlarged verion of the two rectangular region marked in Fig. 5(a). The reult of applying the block-baed egmentation algorithm of Section 4 to the in Fig. 5(a) i hown in Fig. 5(b). The original of Fig. 5(a) wa compreed by baeline with default quantization matrix multiplied by 3, and decompreed by baeline. The ame compreed wa alo decompreed by our propoed cheme. Fig. 6 how the reult of both baeline and our cheme for a mall text region. Ringing artifact are effectively eliminated by our propoed algorithm. Note that our formulation allow the foreground and background color to be locally adaptive, and there i no obviou color hift in the decoded. Fig. 4 how the reult for a mall region containing both background block and picture block. In the lowly varying blue ky, contouring and blocking artifact are ignificantly reduced by our algorithm. (a) Segmentation (b) Original 7. REFERENCES [1] Gregory K. Wallace, The till picture compreion tandard, Commun. ACM, vol. 34, no. 4, pp , [2] Thoma Meier, King N. Ngan, and Gregory Crebbin, Reduction of blocking artifact in and video coding, IEEE (c) Baeline (d) Propoed algorithm Fig. 4. Background block and picture block decoding. 491

5 (a) Segmentation (a) Original (b) Original (c) Baeline (b) Block-baed egmentation Fig. 5. Block-baed egmentation. (a) Original. (b) Segmentation. White: Background block; Red: Text block; Blue: Picture block. (d) Text block decoding Fig. 6. Text block decoding reult. 492

3D SMAP Algorithm. April 11, 2012

3D SMAP Algorithm. April 11, 2012 3D SMAP Algorithm April 11, 2012 Baed on the original SMAP paper [1]. Thi report extend the tructure of MSRF into 3D. The prior ditribution i modified to atify the MRF property. In addition, an iterative

More information

Markov Random Fields in Image Segmentation

Markov Random Fields in Image Segmentation Preented at SSIP 2011, Szeged, Hungary Markov Random Field in Image Segmentation Zoltan Kato Image Proceing & Computer Graphic Dept. Univerity of Szeged Hungary Zoltan Kato: Markov Random Field in Image

More information

Trainable Context Model for Multiscale Segmentation

Trainable Context Model for Multiscale Segmentation Trainable Context Model for Multicale Segmentation Hui Cheng and Charle A. Bouman School of Electrical and Computer Engineering Purdue Univerity Wet Lafayette, IN 47907-1285 {hui, bouman}@ ecn.purdue.edu

More information

Performance of a Robust Filter-based Approach for Contour Detection in Wireless Sensor Networks

Performance of a Robust Filter-based Approach for Contour Detection in Wireless Sensor Networks Performance of a Robut Filter-baed Approach for Contour Detection in Wirele Senor Network Hadi Alati, William A. Armtrong, Jr., and Ai Naipuri Department of Electrical and Computer Engineering The Univerity

More information

Universität Augsburg. Institut für Informatik. Approximating Optimal Visual Sensor Placement. E. Hörster, R. Lienhart.

Universität Augsburg. Institut für Informatik. Approximating Optimal Visual Sensor Placement. E. Hörster, R. Lienhart. Univerität Augburg à ÊÇÅÍÆ ËÀǼ Approximating Optimal Viual Senor Placement E. Hörter, R. Lienhart Report 2006-01 Januar 2006 Intitut für Informatik D-86135 Augburg Copyright c E. Hörter, R. Lienhart Intitut

More information

Analyzing Hydra Historical Statistics Part 2

Analyzing Hydra Historical Statistics Part 2 Analyzing Hydra Hitorical Statitic Part Fabio Maimo Ottaviani EPV Technologie White paper 5 hnode HSM Hitorical Record The hnode i the hierarchical data torage management node and ha to perform all the

More information

MAT 155: Describing, Exploring, and Comparing Data Page 1 of NotesCh2-3.doc

MAT 155: Describing, Exploring, and Comparing Data Page 1 of NotesCh2-3.doc MAT 155: Decribing, Exploring, and Comparing Data Page 1 of 8 001-oteCh-3.doc ote for Chapter Summarizing and Graphing Data Chapter 3 Decribing, Exploring, and Comparing Data Frequency Ditribution, Graphic

More information

Gray-level histogram. Intensity (grey-level) transformation, or mapping. Use of intensity transformations:

Gray-level histogram. Intensity (grey-level) transformation, or mapping. Use of intensity transformations: Faculty of Informatic Eötvö Loránd Univerity Budapet, Hungary Lecture : Intenity Tranformation Image enhancement by point proceing Spatial domain and frequency domain method Baic Algorithm for Digital

More information

Focused Video Estimation from Defocused Video Sequences

Focused Video Estimation from Defocused Video Sequences Focued Video Etimation from Defocued Video Sequence Junlan Yang a, Dan Schonfeld a and Magdi Mohamed b a Multimedia Communication Lab, ECE Dept., Univerity of Illinoi, Chicago, IL b Phyical Realization

More information

/06/$ IEEE 364

/06/$ IEEE 364 006 IEEE International ympoium on ignal Proceing and Information Technology oie Variance Etimation In ignal Proceing David Makovoz IPAC, California Intitute of Technology, MC-0, Paadena, CA, 95 davidm@ipac.caltech.edu;

More information

A PROBABILISTIC NOTION OF CAMERA GEOMETRY: CALIBRATED VS. UNCALIBRATED

A PROBABILISTIC NOTION OF CAMERA GEOMETRY: CALIBRATED VS. UNCALIBRATED A PROBABILISTIC NOTION OF CAMERA GEOMETRY: CALIBRATED VS. UNCALIBRATED Jutin Domke and Yianni Aloimono Computational Viion Laboratory, Center for Automation Reearch Univerity of Maryland College Park,

More information

Introduction to PET Image Reconstruction. Tomographic Imaging. Projection Imaging. PET Image Reconstruction 11/6/07

Introduction to PET Image Reconstruction. Tomographic Imaging. Projection Imaging. PET Image Reconstruction 11/6/07 Introduction to PET Image Recontruction Adam Aleio Nuclear Medicine Lecture Imaging Reearch Laboratory Diviion of Nuclear Medicine Univerity of Wahington Fall 2007 http://dept.wahington.edu/nucmed/irl/education.html

More information

Planning of scooping position and approach path for loading operation by wheel loader

Planning of scooping position and approach path for loading operation by wheel loader 22 nd International Sympoium on Automation and Robotic in Contruction ISARC 25 - September 11-14, 25, Ferrara (Italy) 1 Planning of cooping poition and approach path for loading operation by wheel loader

More information

Tak-Shing Wong, Charles A. Bouman, Ilya Pollak, and Zhigang Fan. Abstract

Tak-Shing Wong, Charles A. Bouman, Ilya Pollak, and Zhigang Fan. Abstract A Document Image Model and Estimation 1 Algorithm for Optimized JPEG Decompression Tak-Shing Wong, Charles A. Bouman, Ilya Pollak, and Zhigang Fan Abstract The JPEG standard is one of the most prevalent

More information

Motion Control (wheeled robots)

Motion Control (wheeled robots) 3 Motion Control (wheeled robot) Requirement for Motion Control Kinematic / dynamic model of the robot Model of the interaction between the wheel and the ground Definition of required motion -> peed control,

More information

Multi-Target Tracking In Clutter

Multi-Target Tracking In Clutter Multi-Target Tracking In Clutter John N. Sander-Reed, Mary Jo Duncan, W.B. Boucher, W. Michael Dimmler, Shawn O Keefe ABSTRACT A high frame rate (0 Hz), multi-target, video tracker ha been developed and

More information

A TOPSIS based Method for Gene Selection for Cancer Classification

A TOPSIS based Method for Gene Selection for Cancer Classification Volume 67 No17, April 2013 A TOPSIS baed Method for Gene Selection for Cancer Claification IMAbd-El Fattah,WIKhedr, KMSallam, 1 Department of Statitic, 3 Department of Deciion upport, 2 Department of information

More information

Performance Evaluation of an Advanced Local Search Evolutionary Algorithm

Performance Evaluation of an Advanced Local Search Evolutionary Algorithm Anne Auger and Nikolau Hanen Performance Evaluation of an Advanced Local Search Evolutionary Algorithm Proceeding of the IEEE Congre on Evolutionary Computation, CEC 2005 c IEEE Performance Evaluation

More information

Bayesian segmentation for damage image using MRF prior

Bayesian segmentation for damage image using MRF prior Bayeian egmentation for damage image uing MRF prior G. Li 1, F.G. Yuan 1, R. Haftka and N. H. Kim 1 Department of Mechanical and Aeropace Engineering, North arolina State Univerity, Raleigh, N, 7695-791,

More information

A User-Attention Based Focus Detection Framework and Its Applications

A User-Attention Based Focus Detection Framework and Its Applications A Uer-Attention Baed Focu Detection Framework and It Application Chia-Chiang Ho, Wen-Huang Cheng, Ting-Jian Pan, Ja-Ling Wu Communication and Multimedia Laboratory, Department of Computer Science and Information

More information

A reduced reference image quality metric based on feature fusion and neural networks

A reduced reference image quality metric based on feature fusion and neural networks Univerity of Wollongong Reearch Online Faculty of Engineering and Information Science - Paper: Part A Faculty of Engineering and Information Science 2011 A reduced reference image quality metric baed on

More information

Hassan Ghaziri AUB, OSB Beirut, Lebanon Key words Competitive self-organizing maps, Meta-heuristics, Vehicle routing problem,

Hassan Ghaziri AUB, OSB Beirut, Lebanon Key words Competitive self-organizing maps, Meta-heuristics, Vehicle routing problem, COMPETITIVE PROBABIISTIC SEF-ORGANIZING MAPS FOR ROUTING PROBEMS Haan Ghaziri AUB, OSB Beirut, ebanon ghaziri@aub.edu.lb Abtract In thi paper, we have applied the concept of the elf-organizing map (SOM)

More information

Building a Compact On-line MRF Recognizer for Large Character Set using Structured Dictionary Representation and Vector Quantization Technique

Building a Compact On-line MRF Recognizer for Large Character Set using Structured Dictionary Representation and Vector Quantization Technique 202 International Conference on Frontier in Handwriting Recognition Building a Compact On-line MRF Recognizer for Large Character Set uing Structured Dictionary Repreentation and Vector Quantization Technique

More information

DAROS: Distributed User-Server Assignment And Replication For Online Social Networking Applications

DAROS: Distributed User-Server Assignment And Replication For Online Social Networking Applications DAROS: Ditributed Uer-Server Aignment And Replication For Online Social Networking Application Thuan Duong-Ba School of EECS Oregon State Univerity Corvalli, OR 97330, USA Email: duongba@eec.oregontate.edu

More information

Research on Star Image Noise Filtering Based on Diffusion Model of Regularization Influence Function

Research on Star Image Noise Filtering Based on Diffusion Model of Regularization Influence Function 016 Sith International Conference on Intrumentation & Meaurement Computer Communication and Control Reearch on Star Image Noie Filtering Baed on Diffuion Model of Regularization Influence Function SunJianming

More information

Compressed Sensing Image Processing Based on Stagewise Orthogonal Matching Pursuit

Compressed Sensing Image Processing Based on Stagewise Orthogonal Matching Pursuit Senor & randucer, Vol. 8, Iue 0, October 204, pp. 34-40 Senor & randucer 204 by IFSA Publihing, S. L. http://www.enorportal.com Compreed Sening Image Proceing Baed on Stagewie Orthogonal Matching Puruit

More information

Laboratory Exercise 6

Laboratory Exercise 6 Laboratory Exercie 6 Adder, Subtractor, and Multiplier The purpoe of thi exercie i to examine arithmetic circuit that add, ubtract, and multiply number. Each type of circuit will be implemented in two

More information

Image authentication and tamper detection using fragile watermarking in spatial domain

Image authentication and tamper detection using fragile watermarking in spatial domain International Journal of Advanced Reearch in Computer Engineering & Technology (IJARCET) Volume 6, Iue 7, July 2017, ISSN: 2278 1323 Image authentication and tamper detection uing fragile watermarking

More information

Multiclass Road Sign Detection using Multiplicative Kernel

Multiclass Road Sign Detection using Multiplicative Kernel Proceeding of the Croatian Computer Viion Workhop, Year 1 Multicla Road Sign Detection uing Multiplicative Kernel Valentina Zadrija Mireo d. d. Zagreb, Croatia valentina.zadrija@mireo.hr Siniša Šegvić

More information

Texture-Constrained Active Shape Models

Texture-Constrained Active Shape Models 107 Texture-Contrained Active Shape Model Shuicheng Yan, Ce Liu Stan Z. Li Hongjiang Zhang Heung-Yeung Shum Qianheng Cheng Microoft Reearch Aia, Beijing Sigma Center, Beijing 100080, China Dept. of Info.

More information

On combining Learning Vector Quantization and the Bayesian classifiers for natural textured images

On combining Learning Vector Quantization and the Bayesian classifiers for natural textured images On combining Learning Vector Quantization and the Bayeian claifier for natural textured image María Guiarro Dept. Ingeniería del Software e Inteligencia Artificial Facultad Informática Univeridad Complutene

More information

(A)ATSR RE-ANALYSIS FOR CLIMATE - CLOUD CLEARING METHODOLOGY

(A)ATSR RE-ANALYSIS FOR CLIMATE - CLOUD CLEARING METHODOLOGY (A)ATSR RE-ANALYSIS FOR CLIMATE - CLOUD CLEARING METHODOLOGY Chri Old, Chri Merchant Univerity of Edinburgh, The Crew Building, Wet Main Road, Edinburgh, EH9 3JN, United Kingdom Email: cold@ed.ac.uk Email:

More information

A Multi-objective Genetic Algorithm for Reliability Optimization Problem

A Multi-objective Genetic Algorithm for Reliability Optimization Problem International Journal of Performability Engineering, Vol. 5, No. 3, April 2009, pp. 227-234. RAMS Conultant Printed in India A Multi-objective Genetic Algorithm for Reliability Optimization Problem AMAR

More information

How to Select Measurement Points in Access Point Localization

How to Select Measurement Points in Access Point Localization Proceeding of the International MultiConference of Engineer and Computer Scientit 205 Vol II, IMECS 205, March 8-20, 205, Hong Kong How to Select Meaurement Point in Acce Point Localization Xiaoling Yang,

More information

A Novel Feature Line Segment Approach for Pattern Classification

A Novel Feature Line Segment Approach for Pattern Classification 12th International Conference on Information Fuion Seattle, WA, USA, July 6-9, 2009 A Novel Feature Line Segment Approach for Pattern Claification Yi Yang Intitute of Integrated Automation Xi an Jiaotong

More information

Routing Definition 4.1

Routing Definition 4.1 4 Routing So far, we have only looked at network without dealing with the iue of how to end information in them from one node to another The problem of ending information in a network i known a routing

More information

Maneuverable Relays to Improve Energy Efficiency in Sensor Networks

Maneuverable Relays to Improve Energy Efficiency in Sensor Networks Maneuverable Relay to Improve Energy Efficiency in Senor Network Stephan Eidenbenz, Luka Kroc, Jame P. Smith CCS-5, MS M997; Lo Alamo National Laboratory; Lo Alamo, NM 87545. Email: {eidenben, kroc, jpmith}@lanl.gov

More information

A METHOD OF REAL-TIME NURBS INTERPOLATION WITH CONFINED CHORD ERROR FOR CNC SYSTEMS

A METHOD OF REAL-TIME NURBS INTERPOLATION WITH CONFINED CHORD ERROR FOR CNC SYSTEMS Vietnam Journal of Science and Technology 55 (5) (017) 650-657 DOI: 10.1565/55-518/55/5/906 A METHOD OF REAL-TIME NURBS INTERPOLATION WITH CONFINED CHORD ERROR FOR CNC SYSTEMS Nguyen Huu Quang *, Banh

More information

CENTER-POINT MODEL OF DEFORMABLE SURFACE

CENTER-POINT MODEL OF DEFORMABLE SURFACE CENTER-POINT MODEL OF DEFORMABLE SURFACE Piotr M. Szczypinki Iintitute of Electronic, Technical Univerity of Lodz, Poland Abtract: Key word: Center-point model of deformable urface for egmentation of 3D

More information

Advanced Encryption Standard and Modes of Operation

Advanced Encryption Standard and Modes of Operation Advanced Encryption Standard and Mode of Operation G. Bertoni L. Breveglieri Foundation of Cryptography - AES pp. 1 / 50 AES Advanced Encryption Standard (AES) i a ymmetric cryptographic algorithm AES

More information

( ) subject to m. e (2) L are 2L+1. = s SEG SEG Las Vegas 2012 Annual Meeting Page 1

( ) subject to m. e (2) L are 2L+1. = s SEG SEG Las Vegas 2012 Annual Meeting Page 1 A new imultaneou ource eparation algorithm uing frequency-divere filtering Ying Ji*, Ed Kragh, and Phil Chritie, Schlumberger Cambridge Reearch Summary We decribe a new imultaneou ource eparation algorithm

More information

Comparison of Methods for Horizon Line Detection in Sea Images

Comparison of Methods for Horizon Line Detection in Sea Images Comparion of Method for Horizon Line Detection in Sea Image Tzvika Libe Evgeny Gerhikov and Samuel Koolapov Department of Electrical Engineering Braude Academic College of Engineering Karmiel 2982 Irael

More information

Khoirul Umam 1, Agus Zainal Arifin 2 and Dini Adni Navastara 3

Khoirul Umam 1, Agus Zainal Arifin 2 and Dini Adni Navastara 3 I J C T A, 9(-A), 016, pp 763-777 International Science Pre A Novel Strategy of Differential Evolution Algorithm Croover Operator Baed on Graylevel Cluter Similarity for Automatic Multilevel Image Threholding

More information

Lecture 14: Minimum Spanning Tree I

Lecture 14: Minimum Spanning Tree I COMPSCI 0: Deign and Analyi of Algorithm October 4, 07 Lecture 4: Minimum Spanning Tree I Lecturer: Rong Ge Scribe: Fred Zhang Overview Thi lecture we finih our dicuion of the hortet path problem and introduce

More information

Delaunay Triangulation: Incremental Construction

Delaunay Triangulation: Incremental Construction Chapter 6 Delaunay Triangulation: Incremental Contruction In the lat lecture, we have learned about the Lawon ip algorithm that compute a Delaunay triangulation of a given n-point et P R 2 with O(n 2 )

More information

On successive packing approach to multidimensional (M-D) interleaving

On successive packing approach to multidimensional (M-D) interleaving On ucceive packing approach to multidimenional (M-D) interleaving Xi Min Zhang Yun Q. hi ankar Bau Abtract We propoe an interleaving cheme for multidimenional (M-D) interleaving. To achieved by uing a

More information

E-APPLAB #1

E-APPLAB #1 E-APPLAB-93-069850 #1 Ultrafat tomography experiment etup at D-1 Beamline at CHESS The experiment etup conit of three major part: x-ray ource, injection chambe and detecto a hown chematically in Fig. EPAPS1a.

More information

AN ALGORITHM FOR RESTRICTED NORMAL FORM TO SOLVE DUAL TYPE NON-CANONICAL LINEAR FRACTIONAL PROGRAMMING PROBLEM

AN ALGORITHM FOR RESTRICTED NORMAL FORM TO SOLVE DUAL TYPE NON-CANONICAL LINEAR FRACTIONAL PROGRAMMING PROBLEM RAC Univerity Journal, Vol IV, No, 7, pp 87-9 AN ALGORITHM FOR RESTRICTED NORMAL FORM TO SOLVE DUAL TYPE NON-CANONICAL LINEAR FRACTIONAL PROGRAMMING PROLEM Mozzem Hoain Department of Mathematic Ghior Govt

More information

A note on degenerate and spectrally degenerate graphs

A note on degenerate and spectrally degenerate graphs A note on degenerate and pectrally degenerate graph Noga Alon Abtract A graph G i called pectrally d-degenerate if the larget eigenvalue of each ubgraph of it with maximum degree D i at mot dd. We prove

More information

Minimum congestion spanning trees in bipartite and random graphs

Minimum congestion spanning trees in bipartite and random graphs Minimum congetion panning tree in bipartite and random graph M.I. Otrovkii Department of Mathematic and Computer Science St. John Univerity 8000 Utopia Parkway Queen, NY 11439, USA e-mail: otrovm@tjohn.edu

More information

USING ARTIFICIAL NEURAL NETWORKS TO APPROXIMATE A DISCRETE EVENT STOCHASTIC SIMULATION MODEL

USING ARTIFICIAL NEURAL NETWORKS TO APPROXIMATE A DISCRETE EVENT STOCHASTIC SIMULATION MODEL USING ARTIFICIAL NEURAL NETWORKS TO APPROXIMATE A DISCRETE EVENT STOCHASTIC SIMULATION MODEL Robert A. Kilmer Department of Sytem Engineering Unite State Military Acaemy Wet Point, NY 1996 Alice E. Smith

More information

Distributed Packet Processing Architecture with Reconfigurable Hardware Accelerators for 100Gbps Forwarding Performance on Virtualized Edge Router

Distributed Packet Processing Architecture with Reconfigurable Hardware Accelerators for 100Gbps Forwarding Performance on Virtualized Edge Router Ditributed Packet Proceing Architecture with Reconfigurable Hardware Accelerator for 100Gbp Forwarding Performance on Virtualized Edge Router Satohi Nihiyama, Hitohi Kaneko, and Ichiro Kudo Abtract To

More information

Data Mining with Linguistic Thresholds

Data Mining with Linguistic Thresholds Int. J. Contemp. Math. Science, Vol. 7, 22, no. 35, 7-725 Data Mining with Linguitic Threhold Tzung-Pei Hong Department of Electrical Engineering National Univerity of Kaohiung Kaohiung, Taiwan, R.O.C.

More information

SLA Adaptation for Service Overlay Networks

SLA Adaptation for Service Overlay Networks SLA Adaptation for Service Overlay Network Con Tran 1, Zbigniew Dziong 1, and Michal Pióro 2 1 Department of Electrical Engineering, École de Technologie Supérieure, Univerity of Quebec, Montréal, Canada

More information

The norm Package. November 15, Title Analysis of multivariate normal datasets with missing values

The norm Package. November 15, Title Analysis of multivariate normal datasets with missing values The norm Package November 15, 2003 Verion 1.0-9 Date 2002/05/06 Title Analyi of multivariate normal dataet with miing value Author Ported to R by Alvaro A. Novo . Original by Joeph

More information

Algorithmic Discrete Mathematics 4. Exercise Sheet

Algorithmic Discrete Mathematics 4. Exercise Sheet Algorithmic Dicrete Mathematic. Exercie Sheet Department of Mathematic SS 0 PD Dr. Ulf Lorenz 0. and. May 0 Dipl.-Math. David Meffert Verion of May, 0 Groupwork Exercie G (Shortet path I) (a) Calculate

More information

A DIVISIVE HIERARCHICAL CLUSTERING- BASED METHOD FOR INDEXING IMAGE INFORMATION

A DIVISIVE HIERARCHICAL CLUSTERING- BASED METHOD FOR INDEXING IMAGE INFORMATION A DIVISIVE HIERARCHICAL CLUSTERING- BASED METHOD FOR INDEXING IMAGE INFORMATION ABSTRACT Najva Izadpanah Department of Computer Engineering, Ilamic Azad Univerity, Qazvin Branch, Qazvin, Iran In mot practical

More information

A New Approach for In-Vehicle Camera Traffic Sign Detection and Recognition.

A New Approach for In-Vehicle Camera Traffic Sign Detection and Recognition. A New Approach for In-Vehicle Camera Traffic Sign Detection and Recognition. Andrzej Ruta, Yongmin Li School of Information Sytem, Computing & Mathematic Brunel Univerity Uxbridge, Middleex UB8 3PH, United

More information

Localized Minimum Spanning Tree Based Multicast Routing with Energy-Efficient Guaranteed Delivery in Ad Hoc and Sensor Networks

Localized Minimum Spanning Tree Based Multicast Routing with Energy-Efficient Guaranteed Delivery in Ad Hoc and Sensor Networks Localized Minimum Spanning Tree Baed Multicat Routing with Energy-Efficient Guaranteed Delivery in Ad Hoc and Senor Network Hanne Frey Univerity of Paderborn D-3398 Paderborn hanne.frey@uni-paderborn.de

More information

Aalborg Universitet. Published in: Proceedings of the Working Conference on Advanced Visual Interfaces

Aalborg Universitet. Published in: Proceedings of the Working Conference on Advanced Visual Interfaces Aalborg Univeritet Software-Baed Adjutment of Mobile Autotereocopic Graphic Uing Static Parallax Barrier Paprocki, Martin Marko; Krog, Kim Srirat; Kritofferen, Morten Bak; Krau, Martin Publihed in: Proceeding

More information

See chapter 8 in the textbook. Dr Muhammad Al Salamah, Industrial Engineering, KFUPM

See chapter 8 in the textbook. Dr Muhammad Al Salamah, Industrial Engineering, KFUPM Goal programming Objective of the topic: Indentify indutrial baed ituation where two or more objective function are required. Write a multi objective function model dla a goal LP Ue weighting um and preemptive

More information

Contents. shortest paths. Notation. Shortest path problem. Applications. Algorithms and Networks 2010/2011. In the entire course:

Contents. shortest paths. Notation. Shortest path problem. Applications. Algorithms and Networks 2010/2011. In the entire course: Content Shortet path Algorithm and Network 21/211 The hortet path problem: Statement Verion Application Algorithm (for ingle ource p problem) Reminder: relaxation, Dijktra, Variant of Dijktra, Bellman-Ford,

More information

Modeling of underwater vehicle s dynamics

Modeling of underwater vehicle s dynamics Proceeding of the 11th WEA International Conference on YTEM, Agio Nikolao, Crete Iland, Greece, July 23-25, 2007 44 Modeling of underwater vehicle dynamic ANDRZEJ ZAK Department of Radiolocation and Hydrolocation

More information

Anisotropic filtering on normal field and curvature tensor field using optimal estimation theory

Anisotropic filtering on normal field and curvature tensor field using optimal estimation theory Aniotropic filtering on normal field and curvature tenor field uing optimal etimation theory Min Liu Yuhen Liu and Karthik Ramani Purdue Univerity, Wet Lafayette, Indiana, USA Email: {liu66 liu28 ramani}@purdue.edu

More information

UC Berkeley International Conference on GIScience Short Paper Proceedings

UC Berkeley International Conference on GIScience Short Paper Proceedings UC Berkeley International Conference on GIScience Short Paper Proceeding Title A novel method for probabilitic coverage etimation of enor network baed on 3D vector repreentation in complex urban environment

More information

ANALYSIS OF THE FIRST LAYER IN WEIGHTLESS NEURAL NETWORKS FOR 3_DIMENSIONAL PATTERN RECOGNITION

ANALYSIS OF THE FIRST LAYER IN WEIGHTLESS NEURAL NETWORKS FOR 3_DIMENSIONAL PATTERN RECOGNITION ANALYSIS OF THE FIRST LAYER IN WEIGHTLESS NEURAL NETWORKS FOR 3_DIMENSIONAL PATTERN RECOGNITION A. Váque-Nava * Ecuela de Ingeniería. CENTRO UNIVERSITARIO MEXICO. DIVISION DE ESTUDIOS SUPERIORES J. Figueroa

More information

Key Terms - MinMin, MaxMin, Sufferage, Task Scheduling, Standard Deviation, Load Balancing.

Key Terms - MinMin, MaxMin, Sufferage, Task Scheduling, Standard Deviation, Load Balancing. Volume 3, Iue 11, November 2013 ISSN: 2277 128X International Journal of Advanced Reearch in Computer Science and Software Engineering Reearch Paper Available online at: www.ijarce.com Tak Aignment in

More information

Analysis of the results of analytical and simulation With the network model and dynamic priority Unchecked Buffer

Analysis of the results of analytical and simulation With the network model and dynamic priority Unchecked Buffer International Reearch Journal of Applied and Baic Science 218 Available online at www.irjab.com ISSN 2251-838X / Vol, 12 (1): 49-53 Science Explorer Publication Analyi of the reult of analytical and imulation

More information

A Practical Model for Minimizing Waiting Time in a Transit Network

A Practical Model for Minimizing Waiting Time in a Transit Network A Practical Model for Minimizing Waiting Time in a Tranit Network Leila Dianat, MASc, Department of Civil Engineering, Sharif Univerity of Technology, Tehran, Iran Youef Shafahi, Ph.D. Aociate Profeor,

More information

Application of Social Relation Graphs for Early Detection of Transient Spammers

Application of Social Relation Graphs for Early Detection of Transient Spammers Radolaw rendel and Henryk Krawczyk Application of Social Relation raph for Early Detection of Tranient Spammer RADOSLAW RENDEL and HENRYK KRAWCZYK Electronic, Telecommunication and Informatic Department

More information

Symmetric Stereo Matching for Occlusion Handling

Symmetric Stereo Matching for Occlusion Handling Symmetric Stereo Matching for Occluion Handling Jian Sun 1 Yin Li 1 Sing Bing Kang 2 Heung-Yeung Shum 1 1 Microoft Reearch Aia 2 Microoft Reearch Beijing, P.R. China Redmond, WA, USA Abtract In thi paper,

More information

A NEW APPROACH IN MEASURING OF THE ROUGHNESS FOR SURFACE CONSTITUTED WITH MACHINING PROCESS BY MATERIAL REMOVAL

A NEW APPROACH IN MEASURING OF THE ROUGHNESS FOR SURFACE CONSTITUTED WITH MACHINING PROCESS BY MATERIAL REMOVAL International Journal of Mechanical and Production Engineering Reearch and Development (IJMPERD) ISSN 49-689 Vol. 3, Iue, Mar 3, 4-5 TJPRC Pvt. Ltd. A NEW APPROACH IN MEASURING OF THE ROUGHNESS FOR SURFACE

More information

A System Dynamics Model for Transient Availability Modeling of Repairable Redundant Systems

A System Dynamics Model for Transient Availability Modeling of Repairable Redundant Systems International Journal of Performability Engineering Vol., No. 3, May 05, pp. 03-. RAMS Conultant Printed in India A Sytem Dynamic Model for Tranient Availability Modeling of Repairable Redundant Sytem

More information

3D MODELLING WITH LINEAR APPROACHES USING GEOMETRIC PRIMITIVES

3D MODELLING WITH LINEAR APPROACHES USING GEOMETRIC PRIMITIVES MAKARA, TEKNOLOGI, VOL. 9, NO., APRIL 5: 3-35 3D MODELLING WITH LINEAR APPROACHES USING GEOMETRIC PRIMITIVES Mochammad Zulianyah Informatic Engineering, Faculty of Engineering, ARS International Univerity,

More information

Drawing Lines in 2 Dimensions

Drawing Lines in 2 Dimensions Drawing Line in 2 Dimenion Drawing a traight line (or an arc) between two end point when one i limited to dicrete pixel require a bit of thought. Conider the following line uperimpoed on a 2 dimenional

More information

M 4 CD: A Robust Change Detection Method for Intelligent Visual Surveillance

M 4 CD: A Robust Change Detection Method for Intelligent Visual Surveillance 1 M 4 CD: A Robut Change Detection Method for Intelligent Viual Surveillance Kunfeng Wang, Member, IEEE, Chao Gou, and Fei-Yue Wang, Fellow, IEEE Abtract In thi paper, we propoe a robut change detection

More information

A Linear Interpolation-Based Algorithm for Path Planning and Replanning on Girds *

A Linear Interpolation-Based Algorithm for Path Planning and Replanning on Girds * Advance in Linear Algebra & Matrix Theory, 2012, 2, 20-24 http://dx.doi.org/10.4236/alamt.2012.22003 Publihed Online June 2012 (http://www.scirp.org/journal/alamt) A Linear Interpolation-Baed Algorithm

More information

Stochastic Search and Graph Techniques for MCM Path Planning Christine D. Piatko, Christopher P. Diehl, Paul McNamee, Cheryl Resch and I-Jeng Wang

Stochastic Search and Graph Techniques for MCM Path Planning Christine D. Piatko, Christopher P. Diehl, Paul McNamee, Cheryl Resch and I-Jeng Wang Stochatic Search and Graph Technique for MCM Path Planning Chritine D. Piatko, Chritopher P. Diehl, Paul McNamee, Cheryl Rech and I-Jeng Wang The John Hopkin Univerity Applied Phyic Laboratory, Laurel,

More information

Modeling the Effect of Mobile Handoffs on TCP and TFRC Throughput

Modeling the Effect of Mobile Handoffs on TCP and TFRC Throughput Modeling the Effect of Mobile Handoff on TCP and TFRC Throughput Antonio Argyriou and Vijay Madietti School of Electrical and Computer Engineering Georgia Intitute of Technology Atlanta, Georgia 3332 25,

More information

Edits in Xylia Validity Preserving Editing of XML Documents

Edits in Xylia Validity Preserving Editing of XML Documents dit in Xylia Validity Preerving diting of XML Document Pouria Shaker, Theodore S. Norvell, and Denni K. Peter Faculty of ngineering and Applied Science, Memorial Univerity of Newfoundland, St. John, NFLD,

More information

The Association of System Performance Professionals

The Association of System Performance Professionals The Aociation of Sytem Performance Profeional The Computer Meaurement Group, commonly called CMG, i a not for profit, worldwide organization of data proceing profeional committed to the meaurement and

More information

Tracking High Speed Skater by Using Multiple Model

Tracking High Speed Skater by Using Multiple Model Vol. 2, No. 26 Tracing High Speed Sater by Uing Multiple Model Guojun Liu & Xianglong Tang School of Computer Science & Engineering Harbin Intitute of Technology Harbin 5000, China E-mail: hitliu@hit.edu.cn

More information

Improved Inference in Bayesian Segmentation Using Monte Carlo Sampling: Application to Hippocampal Subfield Volumetry

Improved Inference in Bayesian Segmentation Using Monte Carlo Sampling: Application to Hippocampal Subfield Volumetry Improved Inference in Bayeian Segmentation Uing Monte Carlo Sampling: Application to Hippocampal Subfield Volumetry Juan Eugenio Igleia a, Mert Rory Sabuncu a, Koen Van Leemput a,b,c, for the Alzheimer

More information

(12) Patent Application Publication (10) Pub. No.: US 2011/ A1

(12) Patent Application Publication (10) Pub. No.: US 2011/ A1 (19) United State US 2011 0316690A1 (12) Patent Application Publication (10) Pub. No.: US 2011/0316690 A1 Siegman (43) Pub. Date: Dec. 29, 2011 (54) SYSTEMAND METHOD FOR IDENTIFYING ELECTRICAL EQUIPMENT

More information

Security Analysis of the Efficient Chaos Pseudo-random Number Generator Applied to Video Encryption

Security Analysis of the Efficient Chaos Pseudo-random Number Generator Applied to Video Encryption Journal of Electronic Teting (2018) 34:709 715 http://doi.org/10.1007/10836-018-5767-0 Security Analyi of the Efficient Chao Peudo-random Number Generator Applied to Video Encryption Dragan Lambić 1 Alekandar

More information

EVen though remarkable progress has been made in

EVen though remarkable progress has been made in JOURNAL OF L A T E X CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 1 Dene Semantic 3D Recontruction Chritian Häne, Chritopher Zach, Andrea Cohen, and Marc Pollefey, Fellow, IEEE Abtract Both image egmentation

More information

A New Approach for In-Vehicle Camera Traffic Sign Detection and Recognition

A New Approach for In-Vehicle Camera Traffic Sign Detection and Recognition MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.merl.com A New Approach for In-Vehicle Camera Traffic Sign Detection and Recognition Andrzej Ruta, Fatih Porikli, Yongmin Li, Shintaro Watanabe, Hirohi

More information

Set-based Approach for Lossless Graph Summarization using Locality Sensitive Hashing

Set-based Approach for Lossless Graph Summarization using Locality Sensitive Hashing Set-baed Approach for Lole Graph Summarization uing Locality Senitive Hahing Kifayat Ullah Khan Supervior: Young-Koo Lee Expected Graduation Date: Fall 0 Deptartment of Computer Engineering Kyung Hee Univerity

More information

Audio-Visual Voice Command Recognition in Noisy Conditions

Audio-Visual Voice Command Recognition in Noisy Conditions Audio-Viual Voice Command Recognition in Noiy Condition Joef Chaloupka, Jan Nouza, Jindrich Zdanky Laboratory of Computer Speech Proceing, Intitute of Information Technology and Electronic, Technical Univerity

More information

CSE 250B Assignment 4 Report

CSE 250B Assignment 4 Report CSE 250B Aignment 4 Report March 24, 2012 Yuncong Chen yuncong@c.ucd.edu Pengfei Chen pec008@ucd.edu Yang Liu yal060@c.ucd.edu Abtract In thi project, we implemented the recurive autoencoder (RAE) a decribed

More information

x y z Design variable positions A

x y z Design variable positions A COMMUNICATIONS IN NUMERICAL METHODS IN ENGINEERING Commun. Numer. Meth. Engng 2001 00:1{7 [Verion: 2000/03/22 v1.0] A tabilied peudo-hell approach for urface parametriation in CFD deign problem O. Soto,R.Lohner

More information

Karen L. Collins. Wesleyan University. Middletown, CT and. Mark Hovey MIT. Cambridge, MA Abstract

Karen L. Collins. Wesleyan University. Middletown, CT and. Mark Hovey MIT. Cambridge, MA Abstract Mot Graph are Edge-Cordial Karen L. Collin Dept. of Mathematic Weleyan Univerity Middletown, CT 6457 and Mark Hovey Dept. of Mathematic MIT Cambridge, MA 239 Abtract We extend the definition of edge-cordial

More information

A Hybrid Deployable Dynamic Traffic Assignment Framework for Robust Online Route Guidance

A Hybrid Deployable Dynamic Traffic Assignment Framework for Robust Online Route Guidance A Hybrid Deployable Dynamic Traffic Aignment Framework for Robut Online Route Guidance Sriniva Peeta School of Civil Engineering, Purdue Univerity Chao Zhou Sabre, Inc. Sriniva Peeta School of Civil Engineering

More information

Variable Resolution Discretization in the Joint Space

Variable Resolution Discretization in the Joint Space Variable Reolution Dicretization in the Joint Space Chritopher K. Monon, David Wingate, and Kevin D. Seppi {c,wingated,keppi}@c.byu.edu Computer Science, Brigham Young Univerity Todd S. Peteron peterto@uvc.edu

More information

Quadrilaterals. Learning Objectives. Pre-Activity

Quadrilaterals. Learning Objectives. Pre-Activity Section 3.4 Pre-Activity Preparation Quadrilateral Intereting geometric hape and pattern are all around u when we tart looking for them. Examine a row of fencing or the tiling deign at the wimming pool.

More information

Chapter 13 Non Sampling Errors

Chapter 13 Non Sampling Errors Chapter 13 Non Sampling Error It i a general aumption in the ampling theory that the true value of each unit in the population can be obtained and tabulated without any error. In practice, thi aumption

More information

Segmentation of ultrasound images multiresolution 2D and 3D algorithm based on global and local statistics

Segmentation of ultrasound images multiresolution 2D and 3D algorithm based on global and local statistics Pattern Recognition Letter 24 (2003) 779 790 www.elevier.com/locate/patrec Segmentation of ultraound image multireolution 2D and 3D algorithm baed on global and local tatitic Djamal Boukerroui a,b, *,

More information

Target Oriented High Resolution SAR Image Formation via Semantic Information Guided Regularizations

Target Oriented High Resolution SAR Image Formation via Semantic Information Guided Regularizations 1 Target Oriented High Reolution SAR Image Formation via Semantic Information Guided Regularization Biao Hou, Member, IEEE, Zaidao Wen, Licheng Jiao, Senior Member, IEEE, and Qian Wu arxiv:1704.07082v1

More information

Service and Network Management Interworking in Future Wireless Systems

Service and Network Management Interworking in Future Wireless Systems Service and Network Management Interworking in Future Wirele Sytem V. Tountopoulo V. Stavroulaki P. Demeticha N. Mitrou and M. Theologou National Technical Univerity of Athen Department of Electrical Engineering

More information

Shortest Path Routing in Arbitrary Networks

Shortest Path Routing in Arbitrary Networks Journal of Algorithm, Vol 31(1), 1999 Shortet Path Routing in Arbitrary Network Friedhelm Meyer auf der Heide and Berthold Vöcking Department of Mathematic and Computer Science and Heinz Nixdorf Intitute,

More information