Traffic Sign Classification Using Ring Partitioned Method
|
|
- Anissa Johns
- 6 years ago
- Views:
Transcription
1 Traffic Sign Classification Using Ring Partitioned Metod Aryuanto Soetedjo and Koici Yamada Laboratory for Management and Information Systems Science, Nagaoa University of Tecnology 603- Kamitomioamaci, Nagaoa Niigata JAPAN a_aryu@alice.nagaoaut.ac.jp, yamada@js.nagaoaut.ac.jp Abstract---Traffic sign recognition usually consists of two parts : detection and classification. In tis paper we describe te classification stage using ring partitioned metod. In tis metod, first te RGB image is converted into gray scale image using color tresolding and istogram specification tecnique. Tis gray scale image, called as specified gray scale image is invariant to te illumination canges. Ten te image is classified using ring partitioned metod. Te image is divided by several concentric areas lie rings. In every ring te istogram is used as an image descriptor. Te matcing process is done by computing te istogram distances for all rings of te images by introducing te weigts for every ring. Te metod doesn t need a lot of samples of sign images for training process, alternatively only te standard sign images are used as te reference images. Te experimental results sow te effectiveness of te metod in te matcing of occluded, rotated, and illumination problems of traffic sign images. I. INTRODUCTION Road traffic signs are important parts in te transportation system. An intelligent system for recognizing traffic signs would be benefical in several ways. It could assist te driver before passing tem; a warning system sends signal to te driver as response to te traffic sign, for example te speed limitation.te driver assist system will mae te driving more safer and easier. In te autonomous veicle, in wic no uman driver in te car, te traffic sign recognition become essensial part in order to guide te veicle wen running in te street. Te maintenance and inventory of tose signs are te oter benefits. By recognition of te status or condition about te traffic signs, te maintenance and inventory can be effectively done using intelligent vision system rater tan manual inspection by uman. Researcers divide te traffic sign recognition system into two stages : te detection stage and te classification stage [,3,,5,7]. In te detection stage, color or sape features are used to generate te candidate regions were traffic signs migt exist in a particular image. In te classification stage, te candidate signs obtained before are classified. Tere are no standard algoritm in te bot stages. Neural networs are used in te classification stage in [3,,7 ]. In [5], te tecnique of Marov model is used to classify te sign. Te matcing pursuit filter is used for classification in []. Te above-mentioned metods consider te rotated, scaled, illumination canges of te traffic sign images. In [3,,7] te rotated, scaled, and imperfect signs were used as training patterns of artificial neural networs. Tose signs were also used as training patterns of matcing pursuit filter in []. In [], since matcing pursuit filter uses wavelet to extract te features, te recognition is invariant to te illumination canges. In [7], te ligting, weater, and sadows influences are reduced by converting te original image to te new image using a pre-selected formula. After conversion, only pixels wose original red components dominate te oter two components (green and blue) can ave a nonzero red component in te new image. Te ue and saturation components were taing into account in order to avoid ligting condition [3]. In [], te template based correlation wic was immune to te ligting canges was used. Te partial occluded images were considered in [3,7]. In [3] te occlusion affects te ue values, producing te ig variance. In te classification stage, te occlusion information obtained in te first analysis is used to decide te matcing image given by neural networ. If te networ discovers tat tere is no learnt pattern similar enoug to te sign presented, te system verifies if tere is occlusion. In te case of occlusion, te maximum value is taen altoug it was low. On te oter and, if te networ response is ig but tere is an occlusion, it means tat te occlusion is small and affects only te sign border. In [7], te partial occluded signs were used in te training process by added Gausssian noise and sift te perfect signs to te left, rigt, up and down. In te previous wors, artificial neural networ or matcing filter are used to classify te traffic signs. Tose metods need a lot of samples of signs in te training process to acieve a good recognition of te disturbed signs. Concerning wit te occlusion problem, in [7] te result depends on te sample patterns used in te training process. Meanwile, in [3] te occlusion information obtained in te previous analysis is used to justify te classification of te neural networ. Here, we propose a new simple metod to cope wit te problems of occlusion, rotation, and illumination canges of traffic sign images. Te metod doesn t need a lot of samples of sign images for training process, alternatively only te standard sign images are used as te reference images. In order to cope wit te occlusion problem, we tae into account te degree of occlusion in te matcing process. It will increase te robustness of te matcing regarding to te occlusion. Te metod divides a rectangular image into several rings, called as ring partitioned metod. An additional metod is introduced in te pre-processing stage to convert te RGB image into a gray scale image wic is invariant to illumination canges, called as specified gray scale image. Te istogram distance matcing is used to matc te target
2 images wit te reference images. Since, te istogram is used, te rotation invariant properties is acquired. Comparing wit te oters, te proposed metod is simple, fast, and robust. Te outline of te paper is as follow. In section, we introduce te pre-processing stage to convert RGB image into a specified gray scale image. Te ring partitioned matcing metod is explained in section 3. Section contains te experimental results. Conclusion is covered in section 5. II. PRE-PROCESSING Te traffic sign images often come wit unperfect images. Te signs are partially occluded by objects or interfered by sadows, rotation, oblique (image tat taen from side view) and affected by illumination canges. Since te illumination canges influence te istogram of image, a pre-processing stage is needed to convert a RGB image into a gray scale image wic is invariant to te illumination canges and sadows. Here, we propose a simple metod based on color tresolding to convert a RGB image into a gray scale image. Fig. illustrates te proposed metod. in some columns of te image will be missed. If it occurs, te pixels along tose columns are replaced wit te blac color. To examine te columns of image wic is occluded, te projection tecnique [5] of red color is employed. Te sign mages come wit te different sizes. Te specified gray scale image obtained before is resized to te size 70x70 pixels. Te resizing image is used to speed up te matcing process. Beside tat, by resizing te image into a square image, an oblique image wic te circular red sign appears as an ellipse can be converted into a circle. III. RING PARTITIONED MATCHING A. Basic idea Histogram is a simple descriptor for an image. Since, te istogram only counts te number of pixels for eac gray scale (or color) value, te spatial information is lost. Some different images will ave te same istograms, suc as sown in fig.. Suppose tere are two blac and wite images A and B, if we divide te images into four areas as sown in fig. 3, te above problem can be reduced. Figure. Two different images wit te same istogram. Figure. Pre-processing. Te traffic signs used in te experiment are limited to te circular sign wit red color border (include te no entry sign). Using tis limitation, tere are tree inds of colors in te signs, i.e. red, blue, and wite. Te color tresolding ten applied to te RGB image to extract te red, blue, wite, and oter colors of te image. After color tresolding, te gray scale image for eac color can be obtained. Since an image is segmented by four color tresolding, tere are four gray scale images corresponding to eac color. Using istogram specification tecnique [], we can create an image wit a particular istogram. By selecting te different gray scale ranges in eac image, finally we can combine four images into an image were gray scales range for red, blue, wite, and oter colors(blac color) are definitely separated. Anoter problem in te real sign image is te occlusion. Te sign often occluded by oter object suc as trees, poles, leaves, etc. Most of te occlusions are caused by trees and poles. Te trees appear wit te brown or green color. In tis case, te color tresolding is able to classify te trees as blac object properly, liewise te dar or blac poles. However, te wite or gray poles can not be solved by te color tresolding. Since tey are similar to te wite color, tey will ave te same appearance wit te wite color of te sign. Here, we assume tat te wite pole wic occlude te sign image is straigt pole and cover te image sign from upper side until lower side of te circle. By tis assumption, te wite pole can be separated from wite color in te sign bacground using te information from te red circle of te image. Sortly, if a pole occludes te sign image, te red color Image A Image B Figure 3. Image partition. Let A and B are istograms of image A and B, respectively. A, A, A3, A, B 3, B are istograms of sub-images A, A, A 3, A 3, respectively. Suppose tat, A (0) = p ; A () = m p ; A (0) = q ; A () = m q; A3 (0) = r ; A3 () = m r ; A (0) = s ; A () = m s; B (0) = q ; B () = m q ; B (0) = p ; B () = m p; B3 (0) = r ; B3 () = m r ; B (0) = s ; B () = m s; () were, * (0) = number of blac pixels, * () = number of wite pixels, m = total pixels in a sub - image, p, q, r, s : positive integers. Ten, te istograms of image A and B are : ( 0) = p + q + r + s; ( ) = m ( p + q + r + s) () A A (0) = q + p + r + s; () = m ( q + p + r + s) B B Te distance between A and B ( dist _ AB ) can be computed using te Euclidean s distance. In general, te distance between A and B in Euclidean space n R is given by n dist i i i = In te case of (), tis distance is given by _ AB = A B = ( A B ). (3)
3 dist _ AB = ( (0) (0)) + ( () Ai Bi Ai i = i = In te case of (), tis distance is given by Bi ()). () dist _ AB = ( (0) (0)) + ( () ()). (5) A B A B If we compute te distance between A and B ( dist _ AB ) using sub-images suc as in (), ten from () and () we obtain dist _ AB = ( p q) + ( p q) (6) = ( p q). Meanwile, if we compute te istogram distance between A and B ( dist _ AB ) using (), ten from () and (5) we obtain dist _ AB = 0. (7) From (6) and (7) (wit assumption tat p q ), we obtain tat two different images wit te same istograms can be recognized as two different images if we mae partition of te image and compute te istogram distance based on te partition. Te above idea can be extended to solve te occlusion problem. In te fig., image A is an image witout occlusion, image B is te anoter image, and image C is image A wit occlusion. Human can intepret easily tat image C is more similar to A tan B. Here, we will sow tat te partition metod sows better performance tan non-partition metod to interpret te occluded image C. Image A Image B Image C (a) (b) (c) Figure. Image partition in te occluded image. Let images in fig. be partitioned into four areas in te same way as fig. 3. By inspecting te pictures, we ave A (0) = p ; A (0) = q ; A3 (0) = r ; A (0) = s; B (0) = p ; B (0) = q ; B3 (0) = r ; B ( 0) = s β ; C (0) = p ; C ( 0) = q α ; C3 ( 0) = r ; C ( 0) = s α ; (8) were β = α ; p, q, r, s, α, β : positive integers. Te distance between image A and C (dist_ac_p) and distance between image B and C (dist_bc_p) for partitioned images are obtained as dist _ AC _ P = α, (9) dist _ BC _ P = α 5. (0) If te distances are calculated for non-partitioned images, te results are given as dist _ AC _ NP = α, () dist _ BC _ NP = α. () From (9)-(), if we compute te distance of image witout partition, te distance between image A and C is te same as te distance between B and C, or image C (image A wit a partial occlusion) can be recognized as image A or as image B. However, using partition metod, te distance between image A and C is lower tan te distance between image B and C, tus image C is more similar to image A tan to image B, similarly to te uman interpretation. Let us observe te general case to mae te partitioning effects clear. We can discuss it wit two partitions witout any loss of generality. Let A and B are te reference images and C is a target image. Te target image C is an occluded image of A. Suppose tat (0) = p; (0) = q; A A (0) = p + β ; (0) = q + β ; (3) B B (0) = p γ ; (0) = q γ ; C C were 0 p m; 0 q n; p β m p; q β n q; 0 < γ p; 0 < γ q; m: number of pixels of te st partition; n: number of pixels of te nd partition. Te distances between A and C and between B and C for partitioned images are obtained as dist _ AC _ P = ( γ + γ ), () dist _ BC _ P = ( β + γ ) + ( β + γ ). (5) For non-partitioned images, te distances are obtained as dist _ AC _ NP = ( γ + γ ), (6) dist _ BC _ NP = ( β + β + γ + γ ). (7) In te partitioned case, C is more similar to A tan B, if te next condition is satisfied. ( β + γ ) + ( β + γ ) > γ + γ. (8) In te non-partitioned case, te condition is ( β + β + γ + γ ) > ( γ + ) γ. (9) or ( β + γ ) + ( β + γ ) > γ + γ + γ γ (0) ( β + γ )( β + γ ). If β and β are positive, te inequalities in (8) and (9) are always satisfied. It means tat by using partitioned or non-partitioned metod, we can classify image C as te original image A properly. In te case were β and β are negative or ave te different signs, tere is no guarantee tat (8) and (9) are satisfied. However, (8) is always satisfied wen (9) olds, if te next condition is satisfied. γ γ ( β + γ )( β + γ ) > 0. () Namely if te above inequality olds, te partitioned case sows te better performance tan te non-partitioned case.
4 Let us examine te case were () olds. From (), we get te following inequality (it is assumed tat β 0 and β 0, for convenience). β γ β γ β < if β > γ ; β > if β < γ. β + γ β + γ () In te case of β = γ, () always olds. For given occlusion, namely γ and γ, te combinations of β and β satisfying () are sown in fig. 5. As sown in fig. 5, te area were () is satisfied is larger tan te area were () is not satisfied, if te occlusion ( γ, ) is not so large compared wit (p,q) of te original γ image. Tus te partitioned case sows te better performance tan te non-partitioned case in many cases. -p -γ β n-q -γ -q m-p β : area were bot te partitioned and te nonpartitioned cases conclude C is similar to A, : area were () is satisfied (te partitioned case is better tan te non-partitioned case), : : area were () is not satisfied (te non-partitioned case is better tan te partitioned case). Figure 5. Solution areas of (). Te effect of te partitioning will increase by partitioning te image into more tan two areas. Here, we use te ring (circle) partition, were an image is partitioned into several areas lie rings. An image is partitioned into -rings wit te same space/widt as sown in fig. 6. Te ring is numbered from te innermost ring. Te outermost ring actually doesn t form a ring. Since an image is rectangular, te outermost ring is used to mae a complete partition of an image. Comparing wit square partition suc as in fig. 3, te ring partition as te following advantages : - Since an image is partitioned into ring areas, it is invariant to te rotation. - Since te traffic sign images are circular signs, te inner parts of te ring partitions ave more important information tan te outer parts. Tus using ring partition, we can adjust te weigting factors easier rater tan square partition. Figure 6. Ring partition. B. Fuzzy istogram Te istogram provides te information about te dynamic range of gray distribution in an image. Te gray value n in a digital image can be (n+) or (n-) witout any appreciable cange in te visual perception. Suc imprecisions are not considered in te normal/usual istogram. Te fuzzy istogram considers te uncertainty arising out of te imprecision of gray levels of te image. Te fuzzy istogram is expressed as [0] : f( ) = ( ' ) µ ( ' ) ' µ (3) were ( ) is usual normalized istogram. µ (') is te resemblance degree of gray level to te gray level. In te case of symmetrical triangular membersip function, µ (') is expressed : ( ') = ' µ max 0, () α α is a positive real number. A fuzzy istogram provides a smooted istogram. By smooting te istogram, suc imprecisions of image caused by te natural of image or along te digital image processing can be reduced. Since in te digital image we wor wit te discrete level, te specified istogram obtained by te istogram specification tecnique is not exactly similar to te particular istogram. Te fuzzy istogram maes te specified istogram similar to tis particular istogram. C. Matcing strategy Matcing between target image and reference images basically finds te reference image wic as te igest similarity wit te target image or te lowest distance to te target image. Te distance or similarity is computed from te te extracted features of image called image descriptor. Te istogram is used as an image descriptor. Te normalized istogram for every ring is calculated and saved as an image descriptor. After discarding te occluded part, te new normalized istogram is saved as an image descriptor. Tis istogram is used in te istogram matcing by computing te istogram distance. Te euclidean s distance is used to calculate te distance between target and reference image. Te distance between target and reference image is calculated as: TR d = α ( TH RH ) α ( TH RH ). (5) were
5 TH i = istogram of target image of ring i, RH i = istogram of reference image of ring i, α i = weigting factor of ring i. α, α, α 3,..., α are weigting factors (between 0 and ) and denote te occlusion degree in every ring, for no occlusion and 0 for occlusion at all. Te reference image wit te lowest distance is selected as matced image. IV. EXPERIMENTS Te algoritm was evaluated to recognize te occluded, illumination cange, sadowed, oblique, and rotated circular traffic signs. Te tested images are taen from real scene using digital camera Te images are taen in te daytime on te sunny and cloudy weater. Since te classification or recognition stage is investigated, te real sign image is selected in te rectangle size from te real scene image by manually using image software tools. Te tested images consist of 80 images, some of tose images are sown in fig. 7. Te algoritm is implemented using MATLAB. Te figs. 8- sow some of te specified gray scales images obtained from te target images. In every figure, te upper side sows te usual gray scale image and its istogram, te lower side sows te specified gray scale image obtained by te proposed metod and its istogram. In fig. 8 and fig. 9, two sign images wit different illumination are sown. Te istograms of te usual gray scale images sow te difference in te range of gray scale. In fig. 9, were te image is darer tan fig. 8, te gray scale for red and blue color is closer compared wit te image in fig. 8. Meanwile, te proposed specified gray scale images ave te same range of gray scales. In te fig. 0, tere are leaf sadows in te sign image. Te proposed metod can reconstruct te image properly. Fig. sows te occluded image. Te occluded object (tree) can be detected and sown as blac object. Figure 9. Low brigtness image. Figure 0. Sadowed image. Figure. Occluded image. In te matcing process, six approaces are evaluated : istogram matcing witout partition, fuzzy istogram matcing witout partition, istogram matcing wit square partition, fuzzy istogram matcing wit square partition, istogram matcing wit ring partition, and fuzzy istogram matcing wit ring partition. Table sows te experiment results. Figure 7. Some of te tested sign images. Figure 8. Hig brigtness image. Table Matcing results. Metod Matcing rate Execution time No partition Usual istogram 9. % s Fuzzy istogram 6.7 % s Square partition (3x3) Usual istogram 7.8 % 0.07 s Fuzzy istogram 8.7 % 0.37 s Ring partition (7 rings) Usual istogram 6.8 % 0. s Fuzzy istogram 93.9 % 0.0 s
6 Te result sows te effectiveness of ring partitioned metod in te recognition of real sign images. Comparing wit te usual istogram, te fuzzy istogram sows te better performance. V. CONCLUSION In tis pa per, we propose a m etod to classify te traffic sign image. First te RGB image is pre-processed using color tresolding and istogram specification tecnique to mae a specified gray scale image. Te specified gray scale image is invariant to te illumination canges. Ten te ring partitioned metod, wic divide an image into several ring areas is used to matc te image by computing te istogram for every ring. It follows from te experiment tat te ring partitioned metod sows te best performance in te matcing of te occluded, rotated, sadowed, and illumination canges images, comparing wit non-partitioned and square partitioned metod. Te color tresolding and specified istogram tecnique wor effectively in te images wit varying illumination and sadows. Te fuzzy istogram tecnique affords te excelent result in te matcing process regarding to te te imprecision, uncertainty of te real digital image. From te experiment, te execution time is 0. second. Te execution time can be faster if te algoritm is written in C languange instead of MATLAB. Te proposed tecnique provides a simple algoritm, terefore te real time implementation can be acieved. Te problems wic may arise using tis proposed classification tecnique are segmentation of traffic sign according to te certain color and detecting te occluded object for various traffic sign images and occluded objects. Te proposed classification metod is part of a traffic sign recognition system. Te future wors will cover te detection stage of traffic signs and classification for anoter type of traffic signs, suc rectangular signs and triangular signs. Recognition in disturbing Environments, Proceedings of te Fourteent International Symposium on Metodologies for Intelligent Systems (ISMIS'03). Maebasi City, Japan, 8-3 October 003. [8] Seanina Luas, Torresen Jim. Detection of Norwegian Speed Limit Signs, Proc. of te 6t European Simulation Multiconference, Delft, NL, SCS, 00. [9] C.V. Jawaar and A.K.Ray. Fuzzy statistics of digital image, IEEE Signal Processing Letters,vol. 3(8), pp. 5-7, 996. [0] R. Zang and M. Zang. A Clustering-based approac to efficient image retrieval, Proc. Of te t IEEE Int l Conf. On Tool wit Artificial Intelligence, Wasington DC,USA, November 00. [] R.C. Gonzalez. Digital Image Processing, Addison-Wesley, 987. REFERENCES [] S.-H. Hsu, C.-L. Huang. Road sign detection and recognition using matcing pursuit metod, Image and Vision Computing, vol. 9, pp. 9-9,00. [] A. de la Escalera, L. Moreno, M.A. Salics, J. M. Armingol. Road Traffic Sign Detection and Classification, IEEE Transaction on Industrial Electronics, vol. (6), pp , 997. [3] A. de la Escalera, J.M Armingol, M. Mata. Traffic sign recognition and analysis for intelligent veicles, Image and Vision Com puting, vol., pp. 7-58, 003. [] D.M. Gavrila. Traffic sign Recognition Revisited, Proc. of te st DAGM Symposium fur Mustereennung, pp , Springer Verlag, 999. [5] J.C. Hsien and S.Y. Cen. Road Sign Detection and Recognition Using marov Model, Te t Worsop on OOTA, 003. [6] J. Miura, T. Kanda, and Y. Sirai. An Active Vision System for Real Time Traffic Sign Recognition, Proceeding 000 IEEE Int. Conf. On Intelligent Transportation Systems, pp. 5-57, Dearborn, MI, Oct [7] H.M. Yang, C.L. Liu, and S.M. Huang. Traffic Sign
INTELLIGENT MACHINE VISION SYSTEM FOR ROAD TRAFFIC SIGN RECOGNITION
INTELLIGENT MACHINE VISION SYSTEM FOR ROAD TRAFFIC SIGN RECOGNITION Aryuanto 1), Koichi Yamada 2), F. Yudi Limpraptono 3) Jurusan Teknik Elektro, Fakultas Teknologi Industri, Institut Teknologi Nasional
More informationVector Processing Contours
Vector Processing Contours Andrey Kirsanov Department of Automation and Control Processes MAMI Moscow State Tecnical University Moscow, Russia AndKirsanov@yandex.ru A.Vavilin and K-H. Jo Department of
More informationTwo Modifications of Weight Calculation of the Non-Local Means Denoising Method
Engineering, 2013, 5, 522-526 ttp://dx.doi.org/10.4236/eng.2013.510b107 Publised Online October 2013 (ttp://www.scirp.org/journal/eng) Two Modifications of Weigt Calculation of te Non-Local Means Denoising
More informationMAPI Computer Vision
MAPI Computer Vision Multiple View Geometry In tis module we intend to present several tecniques in te domain of te 3D vision Manuel Joao University of Mino Dep Industrial Electronics - Applications -
More informationwrobot k wwrobot hrobot (a) Observation area Horopter h(θ) (Virtual) horopters h(θ+ θ lim) U r U l h(θ+ θ) Base line Left camera Optical axis
Selective Acquisition of 3-D Information Enoug for Finding Passable Free Spaces Using an Active Stereo Vision System Atsusi Nisikawa, Atsusi Okubo, and Fumio Miyazaki Department of Systems and Human Science
More informationUUV DEPTH MEASUREMENT USING CAMERA IMAGES
ABCM Symposium Series in Mecatronics - Vol. 3 - pp.292-299 Copyrigt c 2008 by ABCM UUV DEPTH MEASUREMENT USING CAMERA IMAGES Rogerio Yugo Takimoto Graduate Scool of Engineering Yokoama National University
More informationMATH 5a Spring 2018 READING ASSIGNMENTS FOR CHAPTER 2
MATH 5a Spring 2018 READING ASSIGNMENTS FOR CHAPTER 2 Note: Tere will be a very sort online reading quiz (WebWork) on eac reading assignment due one our before class on its due date. Due dates can be found
More informationHaar Transform CS 430 Denbigh Starkey
Haar Transform CS Denbig Starkey. Background. Computing te transform. Restoring te original image from te transform 7. Producing te transform matrix 8 5. Using Haar for lossless compression 6. Using Haar
More informationSection 1.2 The Slope of a Tangent
Section 1.2 Te Slope of a Tangent You are familiar wit te concept of a tangent to a curve. Wat geometric interpretation can be given to a tangent to te grap of a function at a point? A tangent is te straigt
More informationPedestrian Detection Algorithm for On-board Cameras of Multi View Angles
Pedestrian Detection Algoritm for On-board Cameras of Multi View Angles S. Kamijo IEEE, K. Fujimura, and Y. Sibayama Abstract In tis paper, a general algoritm for pedestrian detection by on-board monocular
More informationSection 2.3: Calculating Limits using the Limit Laws
Section 2.3: Calculating Limits using te Limit Laws In previous sections, we used graps and numerics to approimate te value of a it if it eists. Te problem wit tis owever is tat it does not always give
More information4.2 The Derivative. f(x + h) f(x) lim
4.2 Te Derivative Introduction In te previous section, it was sown tat if a function f as a nonvertical tangent line at a point (x, f(x)), ten its slope is given by te it f(x + ) f(x). (*) Tis is potentially
More informationNumerical Derivatives
Lab 15 Numerical Derivatives Lab Objective: Understand and implement finite difference approximations of te derivative in single and multiple dimensions. Evaluate te accuracy of tese approximations. Ten
More informationUnsupervised Learning for Hierarchical Clustering Using Statistical Information
Unsupervised Learning for Hierarcical Clustering Using Statistical Information Masaru Okamoto, Nan Bu, and Tosio Tsuji Department of Artificial Complex System Engineering Hirosima University Kagamiyama
More informationMulti-Stack Boundary Labeling Problems
Multi-Stack Boundary Labeling Problems Micael A. Bekos 1, Micael Kaufmann 2, Katerina Potika 1 Antonios Symvonis 1 1 National Tecnical University of Atens, Scool of Applied Matematical & Pysical Sciences,
More informationSome Handwritten Signature Parameters in Biometric Recognition Process
Some Handwritten Signature Parameters in Biometric Recognition Process Piotr Porwik Institute of Informatics, Silesian Uniersity, Bdziska 39, 41- Sosnowiec, Poland porwik@us.edu.pl Tomasz Para Institute
More informationProceedings of the 8th WSEAS International Conference on Neural Networks, Vancouver, British Columbia, Canada, June 19-21,
Proceedings of te 8t WSEAS International Conference on Neural Networks, Vancouver, Britis Columbia, Canada, June 9-2, 2007 3 Neural Network Structures wit Constant Weigts to Implement Dis-Jointly Removed
More information2 The Derivative. 2.0 Introduction to Derivatives. Slopes of Tangent Lines: Graphically
2 Te Derivative Te two previous capters ave laid te foundation for te study of calculus. Tey provided a review of some material you will need and started to empasize te various ways we will view and use
More information12.2 TECHNIQUES FOR EVALUATING LIMITS
Section Tecniques for Evaluating Limits 86 TECHNIQUES FOR EVALUATING LIMITS Wat ou sould learn Use te dividing out tecnique to evaluate its of functions Use te rationalizing tecnique to evaluate its of
More informationREVERSIBLE DATA HIDING USING IMPROVED INTERPOLATION TECHNIQUE
International Researc Journal of Engineering and Tecnology (IRJET) e-issn: 2395-0056 REVERSIBLE DATA HIDING USING IMPROVED INTERPOLATION TECHNIQUE Devendra Kumar 1, Dr. Krisna Raj 2 (Professor in ECE Department
More informationA UPnP-based Decentralized Service Discovery Improved Algorithm
Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol.1, No.1, Marc 2013, pp. 21~26 ISSN: 2089-3272 21 A UPnP-based Decentralized Service Discovery Improved Algoritm Yu Si-cai*, Wu Yan-zi,
More informationCESILA: Communication Circle External Square Intersection-Based WSN Localization Algorithm
Sensors & Transducers 2013 by IFSA ttp://www.sensorsportal.com CESILA: Communication Circle External Square Intersection-Based WSN Localization Algoritm Sun Hongyu, Fang Ziyi, Qu Guannan College of Computer
More informationOur Calibrated Model has No Predictive Value: An Example from the Petroleum Industry
Our Calibrated Model as No Predictive Value: An Example from te Petroleum Industry J.N. Carter a, P.J. Ballester a, Z. Tavassoli a and P.R. King a a Department of Eart Sciences and Engineering, Imperial
More information12.2 Techniques for Evaluating Limits
335_qd /4/5 :5 PM Page 863 Section Tecniques for Evaluating Limits 863 Tecniques for Evaluating Limits Wat ou sould learn Use te dividing out tecnique to evaluate its of functions Use te rationalizing
More informationExtracting Road Signs using the Color Information
Extracting Road Signs using the Color Information Wen-Yen Wu, Tsung-Cheng Hsieh, and Ching-Sung Lai Abstract In this paper, we propose a method to extract the road signs. Firstly, the grabbed image is
More informationThe (, D) and (, N) problems in double-step digraphs with unilateral distance
Electronic Journal of Grap Teory and Applications () (), Te (, D) and (, N) problems in double-step digraps wit unilateral distance C Dalfó, MA Fiol Departament de Matemàtica Aplicada IV Universitat Politècnica
More informationDesign of PSO-based Fuzzy Classification Systems
Tamkang Journal of Science and Engineering, Vol. 9, No 1, pp. 6370 (006) 63 Design of PSO-based Fuzzy Classification Systems Cia-Cong Cen Department of Electronics Engineering, Wufeng Institute of Tecnology,
More informationMore on Functions and Their Graphs
More on Functions and Teir Graps Difference Quotient ( + ) ( ) f a f a is known as te difference quotient and is used exclusively wit functions. Te objective to keep in mind is to factor te appearing in
More informationLaser Radar based Vehicle Localization in GPS Signal Blocked Areas
International Journal of Computational Intelligence Systems, Vol. 4, No. 6 (December, 20), 00-09 Laser Radar based Veicle Localization in GPS Signal Bloced Areas Ming Yang Department of Automation, Sangai
More informationFault Localization Using Tarantula
Class 20 Fault localization (cont d) Test-data generation Exam review: Nov 3, after class to :30 Responsible for all material up troug Nov 3 (troug test-data generation) Send questions beforeand so all
More information1.4 RATIONAL EXPRESSIONS
6 CHAPTER Fundamentals.4 RATIONAL EXPRESSIONS Te Domain of an Algebraic Epression Simplifying Rational Epressions Multiplying and Dividing Rational Epressions Adding and Subtracting Rational Epressions
More informationSUPER OBLIQUE INCIDENCE INTERFEROMETER USING SWS PRISM
SUPER OBLIQUE INCIDENCE INTERFEROMETER USING SWS PRISM Yukitosi OTANI 1,2), Yasuiro MIZUTANI 2), Noriiro UMEDA 2) 1) Optical Sciences Center, University of Arizona, Arizona 85721 2) Dept, of Mec. Sys.
More informationFast Calculation of Thermodynamic Properties of Water and Steam in Process Modelling using Spline Interpolation
P R E P R N T CPWS XV Berlin, September 8, 008 Fast Calculation of Termodynamic Properties of Water and Steam in Process Modelling using Spline nterpolation Mattias Kunick a, Hans-Joacim Kretzscmar a,
More informationBounding Tree Cover Number and Positive Semidefinite Zero Forcing Number
Bounding Tree Cover Number and Positive Semidefinite Zero Forcing Number Sofia Burille Mentor: Micael Natanson September 15, 2014 Abstract Given a grap, G, wit a set of vertices, v, and edges, various
More information4.1 Tangent Lines. y 2 y 1 = y 2 y 1
41 Tangent Lines Introduction Recall tat te slope of a line tells us ow fast te line rises or falls Given distinct points (x 1, y 1 ) and (x 2, y 2 ), te slope of te line troug tese two points is cange
More information2.8 The derivative as a function
CHAPTER 2. LIMITS 56 2.8 Te derivative as a function Definition. Te derivative of f(x) istefunction f (x) defined as follows f f(x + ) f(x) (x). 0 Note: tis differs from te definition in section 2.7 in
More informationA Practical Approach of Selecting the Edge Detector Parameters to Achieve a Good Edge Map of the Gray Image
Journal of Computer Science 5 (5): 355-362, 2009 ISSN 1549-3636 2009 Science Publications A Practical Approac of Selecting te Edge Detector Parameters to Acieve a Good Edge Map of te Gray Image 1 Akram
More informationAn Interactive X-Ray Image Segmentation Technique for Bone Extraction
An Interactive X-Ray Image Segmentation Tecnique for Bone Extraction Cristina Stolojescu-Crisan and Stefan Holban Politenica University of Timisoara V. Parvan 2, 300223 Timisoara, Romania {cristina.stolojescu@etc.upt.ro
More informationClassification of Osteoporosis using Fractal Texture Features
Classification of Osteoporosis using Fractal Texture Features V.Srikant, C.Dines Kumar and A.Tobin Department of Electronics and Communication Engineering Panimalar Engineering College Cennai, Tamil Nadu,
More informationPYRAMID FILTERS BASED ON BILINEAR INTERPOLATION
PYRAMID FILTERS BASED ON BILINEAR INTERPOLATION Martin Kraus Computer Grapics and Visualization Group, Tecnisce Universität Müncen, Germany krausma@in.tum.de Magnus Strengert Visualization and Interactive
More informationNon-Interferometric Testing
NonInterferometric Testing.nb Optics 513 - James C. Wyant 1 Non-Interferometric Testing Introduction In tese notes four non-interferometric tests are described: (1) te Sack-Hartmann test, (2) te Foucault
More informationLinear Interpolating Splines
Jim Lambers MAT 772 Fall Semester 2010-11 Lecture 17 Notes Tese notes correspond to Sections 112, 11, and 114 in te text Linear Interpolating Splines We ave seen tat ig-degree polynomial interpolation
More informationImplementation of Integral based Digital Curvature Estimators in DGtal
Implementation of Integral based Digital Curvature Estimators in DGtal David Coeurjolly 1, Jacques-Olivier Lacaud 2, Jérémy Levallois 1,2 1 Université de Lyon, CNRS INSA-Lyon, LIRIS, UMR5205, F-69621,
More informationNOTES: A quick overview of 2-D geometry
NOTES: A quick overview of 2-D geometry Wat is 2-D geometry? Also called plane geometry, it s te geometry tat deals wit two dimensional sapes flat tings tat ave lengt and widt, suc as a piece of paper.
More information3.6 Directional Derivatives and the Gradient Vector
288 CHAPTER 3. FUNCTIONS OF SEVERAL VARIABLES 3.6 Directional Derivatives and te Gradient Vector 3.6.1 Functions of two Variables Directional Derivatives Let us first quickly review, one more time, te
More informationRedundancy Awareness in SQL Queries
Redundancy Awareness in QL Queries Bin ao and Antonio Badia omputer Engineering and omputer cience Department University of Louisville bin.cao,abadia @louisville.edu Abstract In tis paper, we study QL
More informationThe Euler and trapezoidal stencils to solve d d x y x = f x, y x
restart; Te Euler and trapezoidal stencils to solve d d x y x = y x Te purpose of tis workseet is to derive te tree simplest numerical stencils to solve te first order d equation y x d x = y x, and study
More informationInvestigating an automated method for the sensitivity analysis of functions
Investigating an automated metod for te sensitivity analysis of functions Sibel EKER s.eker@student.tudelft.nl Jill SLINGER j..slinger@tudelft.nl Delft University of Tecnology 2628 BX, Delft, te Neterlands
More information, 1 1, A complex fraction is a quotient of rational expressions (including their sums) that result
RT. Complex Fractions Wen working wit algebraic expressions, sometimes we come across needing to simplify expressions like tese: xx 9 xx +, xx + xx + xx, yy xx + xx + +, aa Simplifying Complex Fractions
More informationSymmetric Tree Replication Protocol for Efficient Distributed Storage System*
ymmetric Tree Replication Protocol for Efficient Distributed torage ystem* ung Cune Coi 1, Hee Yong Youn 1, and Joong up Coi 2 1 cool of Information and Communications Engineering ungkyunkwan University
More informationDensity Estimation Over Data Stream
Density Estimation Over Data Stream Aoying Zou Dept. of Computer Science, Fudan University 22 Handan Rd. Sangai, 2433, P.R. Cina ayzou@fudan.edu.cn Ziyuan Cai Dept. of Computer Science, Fudan University
More informationMinimizing Memory Access By Improving Register Usage Through High-level Transformations
Minimizing Memory Access By Improving Register Usage Troug Hig-level Transformations San Li Scool of Computer Engineering anyang Tecnological University anyang Avenue, SIGAPORE 639798 Email: p144102711@ntu.edu.sg
More informationMeasuring Length 11and Area
Measuring Lengt 11and Area 11.1 Areas of Triangles and Parallelograms 11.2 Areas of Trapezoids, Romuses, and Kites 11.3 Perimeter and Area of Similar Figures 11.4 Circumference and Arc Lengt 11.5 Areas
More informationPiecewise Polynomial Interpolation, cont d
Jim Lambers MAT 460/560 Fall Semester 2009-0 Lecture 2 Notes Tese notes correspond to Section 4 in te text Piecewise Polynomial Interpolation, cont d Constructing Cubic Splines, cont d Having determined
More informationZernike vs. Zonal Matrix Iterative Wavefront Reconstructor. Sophia I. Panagopoulou, PhD. University of Crete Medical School Dept.
Zernie vs. Zonal Matrix terative Wavefront Reconstructor opia. Panagopoulou PD University of Crete Medical cool Dept. of Optalmology Daniel R. Neal PD Wavefront ciences nc. 480 Central.E. Albuquerque NM
More informationTraffic sign shape classification evaluation II: FFT applied to the signature of Blobs
Traffic sign shape classification evaluation II: FFT applied to the signature of Blobs P. Gil-Jiménez, S. Lafuente-Arroyo, H. Gómez-Moreno, F. López-Ferreras and S. Maldonado-Bascón Dpto. de Teoría de
More informationObstacle Avoiding Real-Time Trajectory Generation and Control of Omnidirectional Vehicles
2009 American Control Conference Hyatt Regency Riverfront, St. Louis, MO, USA June 10-12, 2009 FrC10.6 Obstacle Avoiding Real-Time Trajectory Generation and Control of Omnidirectional Veicles Ji-wung Coi,
More informationGeo-Registration of Aerial Images using RANSAC Algorithm
NCTAESD0 GeoRegistration of Aerial Images using RANSAC Algoritm Seema.B.S Department of E&C, BIT, seemabs80@gmail.com Hemant Kumar Department of E&C, BIT, drkar00@gmail.com VPS Naidu MSDF Lab, FMCD,CSIRNAL,
More informationA Novel QC-LDPC Code with Flexible Construction and Low Error Floor
A Novel QC-LDPC Code wit Flexile Construction and Low Error Floor Hanxin WANG,2, Saoping CHEN,2,CuitaoZHU,2 and Kaiyou SU Department of Electronics and Information Engineering, Sout-Central University
More informationA Cost Model for Distributed Shared Memory. Using Competitive Update. Jai-Hoon Kim Nitin H. Vaidya. Department of Computer Science
A Cost Model for Distributed Sared Memory Using Competitive Update Jai-Hoon Kim Nitin H. Vaidya Department of Computer Science Texas A&M University College Station, Texas, 77843-3112, USA E-mail: fjkim,vaidyag@cs.tamu.edu
More informationAll truths are easy to understand once they are discovered; the point is to discover them. Galileo
Section 7. olume All truts are easy to understand once tey are discovered; te point is to discover tem. Galileo Te main topic of tis section is volume. You will specifically look at ow to find te volume
More informationMAC-CPTM Situations Project
raft o not use witout permission -P ituations Project ituation 20: rea of Plane Figures Prompt teacer in a geometry class introduces formulas for te areas of parallelograms, trapezoids, and romi. e removes
More informationLow-complexity Image-based 3D Gaming
Low-complexity Image-based 3D Gaming Ingo Bauermann and Eckeard Steinbac Institute of Communication Networks, Media Tecnology Group Tecnisce Universität Müncen Munic, Germany {ingo.bauermann,eckeard.steinbac}@tum.de
More informationAn Effective Sensor Deployment Strategy by Linear Density Control in Wireless Sensor Networks Chiming Huang and Rei-Heng Cheng
An ffective Sensor Deployment Strategy by Linear Density Control in Wireless Sensor Networks Ciming Huang and ei-heng Ceng 5 De c e mbe r0 International Journal of Advanced Information Tecnologies (IJAIT),
More informationEfficient Content-Based Indexing of Large Image Databases
Efficient Content-Based Indexing of Large Image Databases ESSAM A. EL-KWAE University of Nort Carolina at Carlotte and MANSUR R. KABUKA University of Miami Large image databases ave emerged in various
More informationChapter K. Geometric Optics. Blinn College - Physics Terry Honan
Capter K Geometric Optics Blinn College - Pysics 2426 - Terry Honan K. - Properties of Ligt Te Speed of Ligt Te speed of ligt in a vacuum is approximately c > 3.0µ0 8 mês. Because of its most fundamental
More informationHASH ALGORITHMS: A DESIGN FOR PARALLEL CALCULATIONS
HASH ALGORITHMS: A DESIGN FOR PARALLEL CALCULATIONS N.G.Bardis Researc Associate Hellenic Ministry of te Interior, Public Administration and Decentralization 8, Dragatsaniou str., Klatmonos S. 0559, Greece
More informationTraffic Signs Recognition using HP and HOG Descriptors Combined to MLP and SVM Classifiers
Traffic Signs Recognition using HP and HOG Descriptors Combined to MLP and SVM Classifiers A. Salhi, B. Minaoui, M. Fakir, H. Chakib, H. Grimech Faculty of science and Technology Sultan Moulay Slimane
More informationAlternating Direction Implicit Methods for FDTD Using the Dey-Mittra Embedded Boundary Method
Te Open Plasma Pysics Journal, 2010, 3, 29-35 29 Open Access Alternating Direction Implicit Metods for FDTD Using te Dey-Mittra Embedded Boundary Metod T.M. Austin *, J.R. Cary, D.N. Smite C. Nieter Tec-X
More informationA signature analysis based method for elliptical shape
A signature analysis based metod for elliptical sape Ivana Guarneri, Mirko Guarnera, Giuseppe Messina and Valeria Tomaselli STMicroelectronics - AST Imaging Lab, Stradale rimosole 50, Catania, Italy ABSTRACT
More informationIntroduction to Computer Graphics 5. Clipping
Introduction to Computer Grapics 5. Clipping I-Cen Lin, Assistant Professor National Ciao Tung Univ., Taiwan Textbook: E.Angel, Interactive Computer Grapics, 5 t Ed., Addison Wesley Ref:Hearn and Baker,
More informationImage Denoising based on Sparse Representation and Dual Dictionary
ISSN: 49 8958, Volume-4 Issue-4, April 015 Image Denoising based on Sparse Representation and Dual Dictionary Anees G Nat, Sreeram G, Sarafudeen K, Sreeraj M C Abstract earning-based image denoising aims
More informationOptimal In-Network Packet Aggregation Policy for Maximum Information Freshness
1 Optimal In-etwork Packet Aggregation Policy for Maimum Information Fresness Alper Sinan Akyurek, Tajana Simunic Rosing Electrical and Computer Engineering, University of California, San Diego aakyurek@ucsd.edu,
More informationOn the use of FHT, its modification for practical applications and the structure of Hough image
On te use of FHT, its modification for practical applications and te structure of Houg image M. Aliev 1,3, E.I. Ersov, D.P. Nikolaev,3 1 Federal Researc Center Computer Science and Control of Russian Academy
More informationCS 234. Module 6. October 16, CS 234 Module 6 ADT Dictionary 1 / 33
CS 234 Module 6 October 16, 2018 CS 234 Module 6 ADT Dictionary 1 / 33 Idea for an ADT Te ADT Dictionary stores pairs (key, element), were keys are distinct and elements can be any data. Notes: Tis is
More informationA New Approach to Estimate the 3-D Surface of Paper
A New Approac to Estimate te 3-D Surface of Paper Yuan Yuan Qu* Daniel Nström* and Björn Kruse* Kewords: Paper Surface 3D Sape Gloss Abstract: Tis paper describes a metod for 3-D surface estimation based
More informationComputing geodesic paths on manifolds
Proc. Natl. Acad. Sci. USA Vol. 95, pp. 8431 8435, July 1998 Applied Matematics Computing geodesic pats on manifolds R. Kimmel* and J. A. Setian Department of Matematics and Lawrence Berkeley National
More informationDeveloping an Efficient Algorithm for Image Fusion Using Dual Tree- Complex Wavelet Transform
ISSN(Online): 30-980 ISSN (Print): 30-9798 (An ISO 397: 007 Certified Organization) Vol. 3, Issue 4, April 05 Developing an Efficient Algoritm for Image Fusion Using Dual Tree- Complex Wavelet Transform
More informationAn Anchor Chain Scheme for IP Mobility Management
An Ancor Cain Sceme for IP Mobility Management Yigal Bejerano and Israel Cidon Department of Electrical Engineering Tecnion - Israel Institute of Tecnology Haifa 32000, Israel E-mail: bej@tx.tecnion.ac.il.
More informationNotes: Dimensional Analysis / Conversions
Wat is a unit system? A unit system is a metod of taking a measurement. Simple as tat. We ave units for distance, time, temperature, pressure, energy, mass, and many more. Wy is it important to ave a standard?
More informationθ R = θ 0 (1) -The refraction law says that: the direction of refracted ray (angle θ 1 from vertical) is (2)
LIGHT (Basic information) - Considering te ligt of a projector in a smoky room, one gets to geometrical optics model of ligt as a set of tiny particles tat travel along straigt lines called "optical rays.
More informationPerformance analysis of robust road sign identification
IOP Conference Series: Materials Science and Engineering OPEN ACCESS Performance analysis of robust road sign identification To cite this article: Nursabillilah M Ali et al 2013 IOP Conf. Ser.: Mater.
More informationTHANK YOU FOR YOUR PURCHASE!
THANK YOU FOR YOUR PURCHASE! Te resources included in tis purcase were designed and created by me. I ope tat you find tis resource elpful in your classroom. Please feel free to contact me wit any questions
More informationTexture Segmentation by Windowed Projection
Texture Segmentation by Windowed Projection 1, 2 Fan-Chen Tseng, 2 Ching-Chi Hsu, 2 Chiou-Shann Fuh 1 Department of Electronic Engineering National I-Lan Institute of Technology e-mail : fctseng@ccmail.ilantech.edu.tw
More informationUtilizing Call Admission Control to Derive Optimal Pricing of Multiple Service Classes in Wireless Cellular Networks
Utilizing Call Admission Control to Derive Optimal Pricing of Multiple Service Classes in Wireless Cellular Networks Okan Yilmaz and Ing-Ray Cen Computer Science Department Virginia Tec {oyilmaz, ircen}@vt.edu
More informationAn Algorithm for Loopless Deflection in Photonic Packet-Switched Networks
An Algoritm for Loopless Deflection in Potonic Packet-Switced Networks Jason P. Jue Center for Advanced Telecommunications Systems and Services Te University of Texas at Dallas Ricardson, TX 75083-0688
More informationCoarticulation: An Approach for Generating Concurrent Plans in Markov Decision Processes
Coarticulation: An Approac for Generating Concurrent Plans in Markov Decision Processes Kasayar Roanimanes kas@cs.umass.edu Sridar Maadevan maadeva@cs.umass.edu Department of Computer Science, University
More information12.2 Investigate Surface Area
Investigating g Geometry ACTIVITY Use before Lesson 12.2 12.2 Investigate Surface Area MATERIALS grap paper scissors tape Q U E S T I O N How can you find te surface area of a polyedron? A net is a pattern
More informationAutomatic Image Registration
Kalpa Publications in Engineering Volume 1, 2017, Pages 402 411 ICRISET2017. International Conference on Researc and Innovations in Science, Engineering &Tecnology. Selected Papers in Engineering Automatic
More informationLocal features and image matching May 8 th, 2018
Local features and image matcing May 8 t, 2018 Yong Jae Lee UC Davis Last time RANSAC for robust fitting Lines, translation Image mosaics Fitting a 2D transformation Homograpy 2 Today Mosaics recap: How
More informationCubic smoothing spline
Cubic smooting spline Menu: QCExpert Regression Cubic spline e module Cubic Spline is used to fit any functional regression curve troug data wit one independent variable x and one dependent random variable
More informationCE 221 Data Structures and Algorithms
CE Data Structures and Algoritms Capter 4: Trees (AVL Trees) Text: Read Weiss, 4.4 Izmir University of Economics AVL Trees An AVL (Adelson-Velskii and Landis) tree is a binary searc tree wit a balance
More informationFeature-Based Steganalysis for JPEG Images and its Implications for Future Design of Steganographic Schemes
Feature-Based Steganalysis for JPEG Images and its Implications for Future Design of Steganograpic Scemes Jessica Fridric Dept. of Electrical Engineering, SUNY Bingamton, Bingamton, NY 3902-6000, USA fridric@bingamton.edu
More informationSoftware Fault Prediction using Machine Learning Algorithm Pooja Garg 1 Mr. Bhushan Dua 2
IJSRD - International Journal for Scientific Researc & Development Vol. 3, Issue 04, 2015 ISSN (online): 2321-0613 Software Fault Prediction using Macine Learning Algoritm Pooja Garg 1 Mr. Busan Dua 2
More informationImage Registration via Particle Movement
Image Registration via Particle Movement Zao Yi and Justin Wan Abstract Toug fluid model offers a good approac to nonrigid registration wit large deformations, it suffers from te blurring artifacts introduced
More informationFast Global Kernel Density Mode Seeking with application to Localisation and Tracking
Fast Global Kernel Density Mode Seeking wit application to Localisation and Tracking Cunua Sen, Micael J. Brooks, Anton van den Hengel Scool of Computer Science, University of Adelaide, SA 55, Australia
More informationFUZZY INVENTORY MODEL WITH SINGLE ITEM UNDER TIME DEPENDENT DEMAND AND HOLDING COST.
Int J of Intelligent omputing and pplied Sciences 5 FZZY INVENORY MODEL WIH SINGLE IEM NDER IME DEPENDEN DEMND ND HOLDING OS S Barik SKPaikray S Misra K Misra orresponding autor: odma@driemsacin bstract:
More informationA Bidirectional Subsethood Based Similarity Measure for Fuzzy Sets
A Bidirectional Subsetood Based Similarity Measure for Fuzzy Sets Saily Kabir Cristian Wagner Timoty C. Havens and Derek T. Anderson Intelligent Modelling and Analysis (IMA) Group and Lab for Uncertainty
More informationThe navigability variable is binary either a cell is navigable or not. Thus, we can invert the entire reasoning by substituting x i for x i : (4)
A Multi-Resolution Pyramid for Outdoor Robot Terrain Perception Micael Montemerlo and Sebastian Trun AI Lab, Stanford University 353 Serra Mall Stanford, CA 94305-9010 {mmde,trun}@stanford.edu Abstract
More informationUNSUPERVISED HIERARCHICAL IMAGE SEGMENTATION BASED ON THE TS-MRF MODEL AND FAST MEAN-SHIFT CLUSTERING
UNSUPERVISED HIERARCHICAL IMAGE SEGMENTATION BASED ON THE TS-MRF MODEL AND FAST MEAN-SHIFT CLUSTERING Raffaele Gaetano, Giuseppe Scarpa, Giovanni Poggi, and Josiane Zerubia Dip. Ing. Elettronica e Telecomunicazioni,
More information