Comparing Image Representations for Training a Convolutional Neural Network to Classify Gender

Size: px
Start display at page:

Download "Comparing Image Representations for Training a Convolutional Neural Network to Classify Gender"

Transcription

1 2013 Frst Internatonal Conference on Artfcal Intellgence, Modellng & Smulaton Comparng Image Representatons for Tranng a Convolutonal Neural Network to Classfy Gender Choon-Boon Ng, Yong-Haur Tay, Bok-Mn Go Faculty of Engneerng and Scence Unverst Tunku Abdul Rahman Kuala Lumpur, Malaysa {ngcb,tayyh,gobm}@utar.edu.my Abstract In ths work, we evaluated the effect of dfferent mage representatons on the classfcaton performance of a convolutonal neural network. Several dfferent methods for normalzaton of the nput data were also consdered. The network was dscrmnatvely traned for the task of gender classfcaton of pedestrans. A publcly avalable dataset was used for tranng, contanng both frontal and rear vews of pedestrans. The best result was obtaned usng grayscale representaton as compared to RGB and YUV, gvng crossvaldated accuracy of 81.5 % on the dataset. The performance of the convolutonal neural network s compettve and comparable to prevous works on the same dataset. Keywords- convolutonal neural network; pedestran; gender classfcaton; nput representaton I. INTRODUCTION A convolutonal neural network s a class of neural network wth a multple-layered archtecture and usng the dea of shared flter weghts to learn task-dependent features from the tranng data provded to t. Its orgn s nspred bologcally by the herarchcal nature of the mammalan prmary vsual cortex. Other such bologcally nspred network archtectures whch have been studed n computer vson system nclude the Neocogntron [1], whch can be consdered as ts predecessor, and the HMAX model [2]. Neural networks wth several hdden layers such as convolutonal neural networks are consdered to have a deep archtecture. Compared to shallow archtectures, a deep archtecture s beleved to be able to more effcently learn complcated functons to represent the hgh-level abstractons requred for complex tasks n artfcal ntellgence [3]. Untl recently, wth the breakthrough of greedy layerwse tranng ntroduced n [4], t has been dffcult to tran such multlayered network archtectures. However, the convolutonal neural network has been a notable excepton, havng been successfully traned usng back-propagaton [5]. Convolutonal neural networks have acheved hghly compettve results n a varety of pattern recognton problems such as handwrtten dgt recognton [6], face detecton [7], traffc sgn classfcaton [8] pedestran detecton [9], object recognton [10] and acton recognton [11]. A convolutonal neural network ntegrates feature extracton and classfcaton nto a sngle framework. Dscrmnatve feature detectors are learnt automatcally by the network from the data durng the supervsed tranng stage, hence the classfer system desgn does not need to rely on dffcult hand-engneered feature extracton. We are nterested n the problem of gender classfcaton of humans usng computer vson. The majorty of prevous research reported n lterature reled on usng only the face or head regon alone to tran a gender classfer [12]. In ths work, we make use of the whole body of a person to tran a gender classfer. Ths s justfable n stuatons where the face s of nsuffcent resoluton or not vsble (e.g. rear vew of the person) The problem of nferrng gender from whole body nformaton was frst nvestgated n [13], n whch Hstogram of Orented Gradents (HOG) features [14] were used to represent patches of the human body mage. These features were used to tran a boostng type classfer. HOG features have also been used n other pror works [15][16]. Gabor flters were used to extract features n [17]. The features dmensons were reduced usng varous manfold learnng methods such as Prncpal Component Analyss and Localty Senstve Dscrmnant Analyss. The features are then fed nto a lnear Support Vector Machne classfer. The best result was obtaned n a ppelne contanng a vew classfer before the gender classfer. In ths work, we traned a convolutonal neural network for the task of pedestran gender classfcaton. When used for such vson tasks, the nput data fed nto the network are the raw pxel values of the mage. An mage can be represented n varous color models. Varous researchers have used dfferent mage representatons to tran the convolutonal neural network, such as grayscale, RGB and YUV. However, t s not clear whch s the most sutable representaton for our task. Referrng to the pror works on recognzng gender from whole body, color was used as supplementary features n [15] [16]. The remander of ths paper s organzed as follows. In secton 2, the methodology of our experment s presented. In secton 3, detals of the experments and tranng procedures are gven. The results are analyzed and compared n secton 4. Fnally, secton 5 presents the concludng remarks for our paper and future work s suggested /13 $ IEEE DOI /AIMS

2 II. METHODOLOGY A. Convolutonal Neural Network We use a dscrmnatvely-traned convolutonal neural network, nspred by the work of LeCun [6] for our experment. A convolutonal neural network typcally ncludes two dfferent knds of layers correspondng to the smple and complex cells n the vsual cortex model of [18], performng local feature detecton and output poolng respectvely. Each layer s comprsed of a stack of so-called feature maps, whch are obtaned by performng convoluton or subsamplng on the prevous layer s feature maps. The archtecture of a convolutonal neural network attempts to acheve some degree of scale, translaton and deformaton nvarance n ts structure through the use of local receptve felds, shared weghts and spatal subsamplng. Fg. 1 shows the archtecture of the convolutonal neural network that we have used n our experments. The network contans of a total of 7 layers, ncludng the nput and output layers. The hdden layers consst of 2 stages of convoluton and subsamplng layers, labeled as C and S respectvely and a fully-connected layer of perceptron unts labeled as F. The feature maps n the frst convoluton layer C1 are obtaned as a result of the convoluton operaton of a set of flters wth the nput unts, whch are then passed through a squashng actvaton functon. Let W,j be the flter of sze f g whch connects the -th feature map from the prevous layer I to the j-th feature map C j and b j the correspondng tranable bas. The feature map s obtaned as follows, C j ( W, j S I b Here denotes the convoluton operaton and S denotes the set of all or selected feature maps from the prevous layer. We use the hyperbolc tangent functon as the squashng actvaton functon whch ntroduces nonlneartes, as gven by the followng equaton, x x e e ( x) tanh( x) (2) x x e e Gven that the sze of a feature map s hw, convoluton wth flter of sze f g wll produce an output feature map wth a sze of (h f + 1) (w g + 1),.e. convoluton s only performed on vald pxel values as opposed to full convoluton whch requres zero-paddng pxels of the mage borders. The same set of weghts for the flter s shared by each unt of a feature map. The values of these weghts are parameters whch are obtaned by supervsed tranng. Hence, weght sharng greatly reduces the number of tranable parameters n order to help reduce overfttng. Overfttng s a problem n whch a traned classfer performs well on the tranng data but does poorly when presented wth new data whch are not from the tranng set. Such a classfer s sad to have poor generalzaton ablty. j ) (1) Next, the feature maps of the subsamplng layer S2 s obtaned by downsamplng each feature map n layer C1. We use the maxmum-poolng operaton [2], n whch the largest value n a local regon s taken. Specfcally, each feature map n layer C1 s parttoned nto non-overlappng pp subregons. The largest value s output from each sub-regon to become the unts that form the feature map of the S2 layer. Every feature map s thus downsampled n ts sze by a factor of p to form the next layer S2. Fg. 2 llustrates an example of the maxmum-poolng operaton. The 66 mage, after applyng 22 maxmum-poolng, s reduced to 33. In the 22 sub-regon shown, the largest value s taken for the subsampled output. Smlarly, the convoluton layer C3 s obtaned followed by subsamplng layer S4. In our work, each feature map n layer C3 s connected to all the feature maps from layer S2. Layer F5 s a layer of perceptrons smlar to the hdden layer of a neural network, each unt fully connected to all the unts n the feature maps of layer S4. The unts n layer F5 also use hyperbolc tangent actvaton functon. Fnally, the output layer unts are connected to all the unts of layer F5. Softmax actvaton unts are used n the output layer to form a lnear classfer. The unt apples a softmax functon on the nput x, gven the weghts W m and bas b m of the m-th unt. Gven n s the number of unts, the output of the m-th unt s gven by the followng equaton, y x m ( ) n exp( W x b 1 exp( W x b ) Each softmax actvaton unt s assocated wth a class label and the output value gves the probablty of the class gven the nput, weghts and bas. The largest output probablty gves the predcted class. Fg. 1. Archtecture of convolutonal neural network. Fg. 2. Example llustratng maxmum poolng operaton. m m ) (3) 25

3 B. Image Representatatons The nput data whch are fed nto the network are the raw pxel values of the mage. An mage can be represented n varous dfferent color models. In our experments we consder three dfferent representatons grayscale, RGB, and YUV. Grayscale representaton contans only ntensty or lumnance nformaton. A grayscale mage, assumng pxel depth of 8 bts, contans pxel values from 0 to 255, presentng a sngle channel as the nput to the neural network. In the RGB model, an mage conssts of three component mages, one for each of the prmary colors red, green or blue. Thus, there are three channels to the nput of the neural network. The YUV model separates a color mage nto lumnance (Y) and chromnance (U and V) components. YUV s actually a famly of color spaces, but generally, n computer vdeo YUV usually refers to the Y C b C r color space [19] where C b and C r are the color dfference values obtaned by subtractng luma Y from blue and red components respectvely In our experments, we use Y C b C r to represent the YUV model. The relatonshp between Y C b C r and RGB color models are gven by the followng equatons, Y = R G B (4) C b = B Y (5) C r = R Y (6) C. Dataset The MIT Pedestran dataset [20] was used n the experments. The dataset conssts of 924 color mages of people. Each mage s 64x128 pxels, contanng the whole body of a person algned to the center of the mage. We used the ground truth for gender label as provded by [13], consstng of 600 males and 288 females. There are vews of both front (420 mages) and rear (468 mages) of a person. The mages were then cropped to 54x108 by removng equally the border pxels and then reszed down to 40x80. Generally, from our ntal work, we obtaned better results usng cropped mages compared to the orgnal. Examples of the cropped mages from the dataset are shown n Fgure 3. Fg. 3. Example of cropped mages from the MIT Pedestran dataset [20] Three dfferent mage representatons as mentoned n the prevous secton were used n the experments. The orgnal mages n RGB are converted to grayscale and YUV. Addtonally, we appled three dfferent normalzaton methods. The value of the pxels n each channel of an mage are scaled down to the range [0,1] or [-1,1], whch we refer to as Type I and II respectvely. Specfcally, for every pxel x n an mage, Type I: x x / 255 (7) Type II: x ( x )/127.5 (8) In the thrd method, Type III, the pxel values are normalzed to zero mean and unt standard devaton. nput data whch are fed nto the network are the raw pxel values of the mage. Gven an mage, the mean and standard devaton 2 of the pxels are calculated. The followng s then appled to every pxel x, Type III: x ( x )/ 2 (9) III. EXPERIMENTS We used the followng parameters for the archtecture of the convolutonal neural network. The sze of the nput mage s 40x80. The number of nput channels s one for grayscale mage and three for color (RGB and YUV) mages. Flters of sze 5x5 are used to obtan 10 feature maps of convoluton layer C1, resultng n feature maps of sze 36x76. Maxmum-poolng operaton s appled to 2x2 nonoverlappng patches of each feature map, resultng n subsamplng layer S2 wth feature maps of sze 18x38. Layer C3 contans 20 features maps wth the sze 14x34 produced from 5x5 flters, each connected to all the feature maps n layer S2. Layer S4 contans feature maps of sze 7x17 obtaned after 2x2 maxmum- poolng operaton. Layer F5 contans 25 neuron unts whch are fully connected to layer S4. The output layer has two softmax actvaton unts for bnary classfcaton. All unts use hyperbolc tangent actvaton. The total number of tranable weghts ncludng bases s 64,857 when usng grayscale mage as nput and 69,857 when usng color mages. For the supervsed tranng of the network, the weghts were ntalzed randomly from a unform dstrbuton n the range of [-(6/f), (6/f)]. Followng the suggeston n [21], f equals the number of nput connectons plus the number of output connectons. The weghts were updated by backpropagaton usng mn-batch stochastc gradent descent on the tranng dataset wth randomzed order. A small batch sze of 4 was chosen, and we tred varous learnng rates unformly dstrbuted between 0.01 and 0.2. Fve-fold cross-valdaton method was appled as follows. The dataset was dvded nto fve folds, wth one fold held out as the valdaton set whle the rest formed the tranng set. After each epoch, the accuracy of the classfer on the valdaton set was checked. The hghest valdaton set accuracy after 500 epochs was taken as the best result and the mean of the valdaton results was taken as the overall accuracy. 26

4 IV. RESULTS AND DISCUSSIONS Gven that three dfferent mage representatons were evaluated, each normalzed by three dfferent methods, as mentoned n the prevous secton, ths resulted n nne dfferent combnatons of datasets to tran the convolutonal neural network. The results obtaned are shown n Table 1, where the lowest average error from fve-fold cross valdaton s shown n each case. Overall, t can be observed that grayscale produces the lowest error, regardless of the type of normalzaton used. RGB s second best, gvng better results compared to YUV. In Fg. 4, the average error at varous learnng rates s plotted for the dfferent mage representatons wth Type I normalzaton. The average error ncreases by a large amount when the learnng rate becomes too large, and slghtly when too small. TABLE I. EXPERIMENTAL RESULTS Normalzaton Average error (%) method grayscale RGB YUV Unnormalzed Type I Type II Type III Fg. 4. Plot of average error for Type-I normalzaton Fg. 5. Plot of average error for grayscale representaton TABLE II. COMPARISON WITH OTHER METHODS Method Average accuracy std. dev. (%) Cao et al. [13] Collns et al. [15] a Guo et al. [17] Our classfer Usng frontal vew mages only It can also be observed that the best normalzaton method s Type I, n whch the data was scaled to the range [0,1]. It gves the lowest error for all three dfferent mage representatons. Indeed, some form of normalzaton seems essental for achevng good performance, as otherwse the average error s sgnfcantly hgher. In Fg. 5, the average error at varous learnng rates s plotted for grayscale representaton wth dfferent normalzaton methods. For dfferent learnng rates, Type II and III do not have such a large varance of error compared to Type I. Ths was also observed for the RGB and YUV representatons (charts not shown here). In any case, the lowest error s found at dfferent learnng rates. In terms of computaton tme, grayscale representaton, havng only one mage channel, has the advantage of slghtly faster tranng tme compared to RGB or YUV wth three mage channels. In our mplementaton usng Python wth the Theano mathematcal computaton lbrary [22] runnng on GPU, t takes an average of 4.8 mnutes for a sngle run of 500 epochs when usng grayscale compared to 5.1 mnutes usng RGB or YUV, approxmately 6% faster. The lowest error overall s 18.53% wth normalzaton to [0,1] and grayscale representaton. Ths corresponds to an accuracy of %. As shown n Table 2, the performance of the convolutonal neural network s compettve and comparable to other methods appled on the same dataset. To the best of our understandng, these methods also reported the average accuracy from fve-fold cross valdaton. V. CONCLUSION In ths paper, we have compared three dfferent mage representatons for tranng a convolutonal neural network to classfy the gender of pedestrans. Three dfferent normalzaton methods were used for preprocessng. From the experments, grayscale representaton produced the best result, followed by RGB and YUV. Usng grayscale mages normalzed to the range of [0,1] resulted n average gender classfcaton accuracy of 81.5% on the MIT pedestran dataset, whch s compettve wth prevous works on the same dataset. In the future, we plan to attempt more challengng datasets of people n varous poses and artculatons. ACKNOWLEDGMENT C.B.N. gratefully acknowledges the support obtaned from UTAR Research Fund and UTAR Staff Scholarshp Scheme. 27

5 REFERENCES [1] K. Fukushma, Neocogntron: A self-organzng neural network model for a mechansm of pattern recognton unaffected by shft n poston, Bologcal Cybernetcs, vol. 36, no. 4, pp , [2] M. Resenhuber and T. Poggo, Herarchcal models of object recognton n cortex., Nature Neuroscence, vol. 2, no. 11, pp , Nov [3] Y. Bengo, Learnng deep archtectures for AI, n Foundatons and Trends n Machne Learnng, vol. 2, no. 1, 2009, pp [4] G. Hnton, S. Osndero, and Y. Teh, A fast learnng algorthm for deep belef nets, Neural Computaton, vol. 1554, no. 18, pp , [5] Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, and L. D. Jackel, Backpropagaton appled to handwrtten zp code recognton, Neural Computaton, vol. 1, no. 4, pp , [6] Y. LeCun, L. Bottou, Y. Bengo, and P. Haffner, Gradentbased learnng appled to document recognton, Proceedngs of the IEEE, vol. 86, no. 11, pp , [7] M. Osadchy, Y. Cun, and M. Mller, Synergstc face detecton and pose estmaton wth energy-based models, The Journal of Machne Learnng Research, vol. 8, pp , [8] D. Cresan, U. Meer, J. Masc, and J. Schmdhuber, A commttee of neural networks for traffc sgn classfcaton, n Internatonal Jont Conference on Neural Networks, 2011, pp [9] P. Sermanet and K. Kavukcuoglu, Pedestran detecton wth unsupervsed mult-stage feature learnng, n Internatonal Conference on Computer Vson and Pattern Recognton, 2013, n press. [10] A. Krzhevsky, I. Sutskever, and G. Hnton, ImageNet classfcaton wth deep convolutonal neural networks, n Advances n Neural Informaton Processng Systems, 2012, vol. 25, pp [11] S. J, W. Xu, M. Yang, and K. Yu, 3D convolutonal neural networks for human acton recognton, IEEE Transactons on Pattern Analyss and Machne Intellgence, vol. 35, no. 1, pp , [12] C. Ng, Y. Tay, and B. Go, Recognzng human gender n computer vson: a survey, n PRICAI 2012: Trends n Artfcal Intellgence, Sprnger Berln Hedelberg, 2012, pp [13] L. Cao, M. Dkmen, Y. Fu, and T. S. Huang, Gender recognton from body, n Proceedngs of the 16th ACM Internatonal Conference on Multmeda, 2008, pp [14] N. Dalal and B. Trggs, Hstograms of orented gradents for human detecton, n 2005 IEEE Computer Socety Conference on Computer Vson and Pattern Recognton, 2005, vol. 1, pp [15] M. Collns, J. Zhang, and P. Mller, Full body mage feature representatons for gender proflng, n 2009 IEEE 12th Internatonal Conference on Computer Vson Workshops, 2009, pp [16] L. Bourdev, S. Maj, and J. Malk, Descrbng people: A poselet-based approach to attrbute classfcaton, n 2011 IEEE Internatonal Conference on Computer Vson, 2011, pp [17] G. Guo, G. Mu, and Y. Fu, Gender from body: a bologcally-nspred approach wth manfold learnng, n Proceedngs of the 9th Asan Conference on Computer Vson, 2009, no. 1, pp [18] D. Hubel and T. Wesel, Receptve felds, bnocular nteracton and functonal archtecture n the cat s vsual cortex, The Journal of Physology, vol. 160, pp , [19] About YUV vdeo, [Onlne]. Avalable: [20] M. Oren, C. Papageorgou, P. Snha, E. Osuna, and T. Poggo, Pedestran detecton usng wavelet templates, n Proceedngs of IEEE Computer Socety Conference on Computer Vson and Pattern Recognton, 1997, pp [21] X. Glorot and Y. Bengo, Understandng the dffculty of tranng deep feedforward neural networks, n Proceedngs of the Internatonal Conference on Artfcal Intellgence and Statstcs, 2010, vol. 9, pp [22] J. Bergstra, et al. "Theano: a CPU and GPU math expresson compler," n Proceedngs of the Python for Scentfc Computng Conference,

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1 4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:

More information

Classifying Acoustic Transient Signals Using Artificial Intelligence

Classifying Acoustic Transient Signals Using Artificial Intelligence Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Texture Feature Extraction Inspired by Natural Vision System and HMAX Algorithm

Texture Feature Extraction Inspired by Natural Vision System and HMAX Algorithm The Journal of Mathematcs and Computer Scence Avalable onlne at http://www.tjmcs.com The Journal of Mathematcs and Computer Scence Vol. 4 No.2 (2012) 197-206 Texture Feature Extracton Inspred by Natural

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover source-mismatch

Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover source-mismatch Deep learnng s a good steganalyss tool when embeddng key s reused for dfferent mages, even f there s a cover source-msmatch Lonel PIBRE 2,3, Jérôme PASQUET 2,3, Dno IENCO 2,3, Marc CHAUMONT 1,2,3 (1) Unversty

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Research of Image Recognition Algorithm Based on Depth Learning

Research of Image Recognition Algorithm Based on Depth Learning 208 4th World Conference on Control, Electroncs and Computer Engneerng (WCCECE 208) Research of Image Recognton Algorthm Based on Depth Learnng Zhang Jan, J Xnhao Zhejang Busness College, Hangzhou, Chna,

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Machine Learning 9. week

Machine Learning 9. week Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below

More information

Detection of an Object by using Principal Component Analysis

Detection of an Object by using Principal Component Analysis Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,

More information

Face Detection with Deep Learning

Face Detection with Deep Learning Face Detecton wth Deep Learnng Yu Shen Yus122@ucsd.edu A13227146 Kuan-We Chen kuc010@ucsd.edu A99045121 Yzhou Hao y3hao@ucsd.edu A98017773 Mn Hsuan Wu mhwu@ucsd.edu A92424998 Abstract The project here

More information

Histogram of Template for Pedestrian Detection

Histogram of Template for Pedestrian Detection PAPER IEICE TRANS. FUNDAMENTALS/COMMUN./ELECTRON./INF. & SYST., VOL. E85-A/B/C/D, No. xx JANUARY 20xx Hstogram of Template for Pedestran Detecton Shaopeng Tang, Non Member, Satosh Goto Fellow Summary In

More information

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Comparison Study of Textural Descriptors for Training Neural Network Classifiers

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

More information

Learning-based License Plate Detection on Edge Features

Learning-based License Plate Detection on Edge Features Learnng-based Lcense Plate Detecton on Edge Features Wng Teng Ho, Woo Hen Yap, Yong Haur Tay Computer Vson and Intellgent Systems (CVIS) Group Unverst Tunku Abdul Rahman, Malaysa wngteng_h@yahoo.com, woohen@yahoo.com,

More information

Local Quaternary Patterns and Feature Local Quaternary Patterns

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

More information

EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS

EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS P.G. Demdov Yaroslavl State Unversty Anatoly Ntn, Vladmr Khryashchev, Olga Stepanova, Igor Kostern EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS Yaroslavl, 2015 Eye

More information

Online Detection and Classification of Moving Objects Using Progressively Improving Detectors

Online Detection and Classification of Moving Objects Using Progressively Improving Detectors Onlne Detecton and Classfcaton of Movng Objects Usng Progressvely Improvng Detectors Omar Javed Saad Al Mubarak Shah Computer Vson Lab School of Computer Scence Unversty of Central Florda Orlando, FL 32816

More information

(a) Input data X n. (b) VersNet. (c) Output data Y n. (d) Supervsed data D n. Fg. 2 Illustraton of tranng for proposed CNN. 2. Related Work In segment

(a) Input data X n. (b) VersNet. (c) Output data Y n. (d) Supervsed data D n. Fg. 2 Illustraton of tranng for proposed CNN. 2. Related Work In segment 一般社団法人電子情報通信学会 THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS 信学技報 IEICE Techncal Report SANE2017-92 (2018-01) Deep Learnng for End-to-End Automatc Target Recognton from Synthetc

More information

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

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

More information

Cluster Analysis of Electrical Behavior

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

More information

Fast Feature Value Searching for Face Detection

Fast Feature Value Searching for Face Detection Vol., No. 2 Computer and Informaton Scence Fast Feature Value Searchng for Face Detecton Yunyang Yan Department of Computer Engneerng Huayn Insttute of Technology Hua an 22300, Chna E-mal: areyyyke@63.com

More information

Willie L. Scott, II IBM, Austin, Texas. Abstract

Willie L. Scott, II IBM, Austin, Texas. Abstract The o Approach to Autonomc Face Recognton Wlle L. Scott, II IBM, Austn, Texas Abstract A method of autonomc face recognton based on the bologcally plausble network of networks (o) model of nformaton processng

More information

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like: Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A

More information

Convolutional Neural Network- based Human Recognition for Vision Occupancy Sensors

Convolutional Neural Network- based Human Recognition for Vision Occupancy Sensors 10 Int'l Conf. IP, Comp. Vson, and Pattern Recognton IPCV'18 Convolutonal Neural Network- based Human Recognton for Vson Occupancy Sensors Seung Soo Lee and Manbae Km * Dept. of Computer and Communcatons

More information

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline mage Vsualzaton mage Vsualzaton mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and Analyss outlne mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Discriminative classifiers for object classification. Last time

Discriminative classifiers for object classification. Last time Dscrmnatve classfers for object classfcaton Thursday, Nov 12 Krsten Grauman UT Austn Last tme Supervsed classfcaton Loss and rsk, kbayes rule Skn color detecton example Sldng ndo detecton Classfers, boostng

More information

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM

Classification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based

More information

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION

A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION 1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

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

More information

Vectorization of Image Outlines Using Rational Spline and Genetic Algorithm

Vectorization of Image Outlines Using Rational Spline and Genetic Algorithm 01 Internatonal Conference on Image, Vson and Computng (ICIVC 01) IPCSIT vol. 50 (01) (01) IACSIT Press, Sngapore DOI: 10.776/IPCSIT.01.V50.4 Vectorzaton of Image Outlnes Usng Ratonal Splne and Genetc

More information

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

More information

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET 1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School

More information

C3 Effective features inspired from Ventral and dorsal stream of visual cortex for view independent face recognition

C3 Effective features inspired from Ventral and dorsal stream of visual cortex for view independent face recognition ACSJ Advances n Computer Scence: an nternatonal Journal, Vol. 5, ssue 1, No.19, January 016 SSN : 3-5157 C3 Effectve features nspred from Ventral and dorsal stream of vsual cortex for vew ndependent face

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face Recognition University at Buffalo CSE666 Lecture Slides Resources: Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural

More information

arxiv: v2 [cs.cv] 9 Apr 2018

arxiv: v2 [cs.cv] 9 Apr 2018 Boundary-senstve Network for Portrat Segmentaton Xanzh Du 1, Xaolong Wang 2, Dawe L 2, Jngwen Zhu 2, Serafettn Tasc 2, Cameron Uprght 2, Stephen Walsh 2, Larry Davs 1 1 Computer Vson Lab, UMIACS, Unversty

More information

Hyperspectral Image Classification Based on Local Binary Patterns and PCANet

Hyperspectral Image Classification Based on Local Binary Patterns and PCANet Hyperspectral Image Classfcaton Based on Local Bnary Patterns and PCANet Huzhen Yang a, Feng Gao a, Junyu Dong a, Yang Yang b a Ocean Unversty of Chna, Department of Computer Scence and Technology b Ocean

More information

Vol. 5, No. 3 March 2014 ISSN Journal of Emerging Trends in Computing and Information Sciences CIS Journal. All rights reserved.

Vol. 5, No. 3 March 2014 ISSN Journal of Emerging Trends in Computing and Information Sciences CIS Journal. All rights reserved. Journal of Emergng Trends n Computng and Informaton Scences 009-03 CIS Journal. All rghts reserved. http://www.csjournal.org Unhealthy Detecton n Lvestock Texture Images usng Subsampled Contourlet Transform

More information

Attributed Relational Graph Based Feature Extraction of Body Poses In Indian Classical Dance Bharathanatyam

Attributed Relational Graph Based Feature Extraction of Body Poses In Indian Classical Dance Bharathanatyam Athra.Sugathan et al Int. Journal of Engneerng Research and Applcatons ISSN : 2248-9622, Vol. 4, Issue 5( Verson 7), May 2014, pp.11-17 RESEARCH ARTICLE OPEN ACCESS Attrbuted Relatonal Graph Based Feature

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

Efficient Segmentation and Classification of Remote Sensing Image Using Local Self Similarity

Efficient Segmentation and Classification of Remote Sensing Image Using Local Self Similarity ISSN(Onlne): 2320-9801 ISSN (Prnt): 2320-9798 Internatonal Journal of Innovatve Research n Computer and Communcaton Engneerng (An ISO 3297: 2007 Certfed Organzaton) Vol.2, Specal Issue 1, March 2014 Proceedngs

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

arxiv: v1 [cs.sd] 22 Dec 2017

arxiv: v1 [cs.sd] 22 Dec 2017 Musc Genre Classfcaton wth Parallelng Recurrent Convolutonal Neural Network arxv:1712.08370v1 [cs.sd] 22 Dec 2017 Ln Feng, Shenlan Lu, Janng Yao December 2017 Abstract Deep learnng has been demonstrated

More information

Understanding the difficulty of training deep feedforward neural networks

Understanding the difficulty of training deep feedforward neural networks Understandng the dffculty of tranng deep feedforward neural networks Xaver Glorot Yoshua Bengo DIRO, Unversté de Montréal, Montréal, Québec, Canada Abstract Whereas before 2006 t appears that deep multlayer

More information

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 1. SSDH: Semi-supervised Deep Hashing for Large Scale Image Retrieval

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 1. SSDH: Semi-supervised Deep Hashing for Large Scale Image Retrieval IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY SSDH: Sem-supervsed Deep Hashng for Large Scale Image Retreval Jan Zhang, and Yuxn Peng arxv:607.08477v2 [cs.cv] 8 Jun 207 Abstract Hashng

More information

Classification algorithms on the cell processor

Classification algorithms on the cell processor Rochester Insttute of Technology RIT Scholar Works Theses Thess/Dssertaton Collectons 8-1-2008 Classfcaton algorthms on the cell processor Mateusz Wyganowsk Follow ths and addtonal works at: http://scholarworks.rt.edu/theses

More information

A Bilinear Model for Sparse Coding

A Bilinear Model for Sparse Coding A Blnear Model for Sparse Codng Davd B. Grmes and Rajesh P. N. Rao Department of Computer Scence and Engneerng Unversty of Washngton Seattle, WA 98195-2350, U.S.A. grmes,rao @cs.washngton.edu Abstract

More information

Research Article A High-Order CFS Algorithm for Clustering Big Data

Research Article A High-Order CFS Algorithm for Clustering Big Data Moble Informaton Systems Volume 26, Artcle ID 435627, 8 pages http://dx.do.org/.55/26/435627 Research Artcle A Hgh-Order Algorthm for Clusterng Bg Data Fanyu Bu,,2 Zhku Chen, Peng L, Tong Tang, 3 andyngzhang

More information

Face Recognition Based on SVM and 2DPCA

Face Recognition Based on SVM and 2DPCA Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, Unversty

More information

ALEXNET FEATURE EXTRACTION AND MULTI-KERNEL LEARNING FOR OBJECT- ORIENTED CLASSIFICATION

ALEXNET FEATURE EXTRACTION AND MULTI-KERNEL LEARNING FOR OBJECT- ORIENTED CLASSIFICATION ALEXNET FEATURE EXTRACTION AND MULTI-KERNEL LEARNING FOR OBJECT- ORIENTED CLASSIFICATION Lng Dng 1, Hongy L 2, *, Changmao Hu 2, We Zhang 2, Shumn Wang 1 1 Insttute of Earthquake Forecastng, Chna Earthquake

More information

Learning a Class-Specific Dictionary for Facial Expression Recognition

Learning a Class-Specific Dictionary for Facial Expression Recognition BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 4 Sofa 016 Prnt ISSN: 1311-970; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-016-0067 Learnng a Class-Specfc Dctonary for

More information

Robust Inlier Feature Tracking Method for Multiple Pedestrian Tracking

Robust Inlier Feature Tracking Method for Multiple Pedestrian Tracking 2011 Internatonal Conference on Informaton and Intellgent Computng IPCSIT vol.18 (2011) (2011) IACSIT Press, Sngapore Robust Inler Feature Trackng Method for Multple Pedestran Trackng Young-Chul Lm a*

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

Image Matching Algorithm based on Feature-point and DAISY Descriptor

Image Matching Algorithm based on Feature-point and DAISY Descriptor JOURNAL OF MULTIMEDIA, VOL. 9, NO. 6, JUNE 2014 829 Image Matchng Algorthm based on Feature-pont and DAISY Descrptor L L School of Busness, Schuan Agrcultural Unversty, Schuan Dujanyan 611830, Chna Abstract

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

Mining Image Features in an Automatic Two- Dimensional Shape Recognition System

Mining Image Features in an Automatic Two- Dimensional Shape Recognition System Internatonal Journal of Appled Mathematcs and Computer Scences Volume 2 Number 1 Mnng Image Features n an Automatc Two- Dmensonal Shape Recognton System R. A. Salam, M.A. Rodrgues Abstract The number of

More information

On Modeling Variations For Face Authentication

On Modeling Variations For Face Authentication On Modelng Varatons For Face Authentcaton Xaomng Lu Tsuhan Chen B.V.K. Vjaya Kumar Department of Electrcal and Computer Engneerng, Carnege Mellon Unversty Abstract In ths paper, we present a scheme for

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information

A fast algorithm for color image segmentation

A fast algorithm for color image segmentation Unersty of Wollongong Research Onlne Faculty of Informatcs - Papers (Arche) Faculty of Engneerng and Informaton Scences 006 A fast algorthm for color mage segmentaton L. Dong Unersty of Wollongong, lju@uow.edu.au

More information

Enhanced Face Detection Technique Based on Color Correction Approach and SMQT Features

Enhanced Face Detection Technique Based on Color Correction Approach and SMQT Features Journal of Software Engneerng and Applcatons, 2013, 6, 519-525 http://dx.do.org/10.4236/jsea.2013.610062 Publshed Onlne October 2013 (http://www.scrp.org/journal/jsea) 519 Enhanced Face Detecton Technque

More information

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between

More information

Scale Selective Extended Local Binary Pattern For Texture Classification

Scale Selective Extended Local Binary Pattern For Texture Classification Scale Selectve Extended Local Bnary Pattern For Texture Classfcaton Yutng Hu, Zhlng Long, and Ghassan AlRegb Multmeda & Sensors Lab (MSL) Georga Insttute of Technology 03/09/017 Outlne Texture Representaton

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

Detection of hand grasping an object from complex background based on machine learning co-occurrence of local image feature

Detection of hand grasping an object from complex background based on machine learning co-occurrence of local image feature Detecton of hand graspng an object from complex background based on machne learnng co-occurrence of local mage feature Shnya Moroka, Yasuhro Hramoto, Nobutaka Shmada, Tadash Matsuo, Yoshak Shra Rtsumekan

More information

Classifier Swarms for Human Detection in Infrared Imagery

Classifier Swarms for Human Detection in Infrared Imagery Classfer Swarms for Human Detecton n Infrared Imagery Yur Owechko, Swarup Medasan, and Narayan Srnvasa HRL Laboratores, LLC 3011 Malbu Canyon Road, Malbu, CA 90265 {owechko, smedasan, nsrnvasa}@hrl.com

More information

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines (IJCSIS) Internatonal Journal of Computer Scence and Informaton Securty, Herarchcal Web Page Classfcaton Based on a Topc Model and Neghborng Pages Integraton Wongkot Srura Phayung Meesad Choochart Haruechayasak

More information

Gender Classification using Interlaced Derivative Patterns

Gender Classification using Interlaced Derivative Patterns Gender Classfcaton usng Interlaced Dervatve Patterns Author Shobernejad, Ameneh, Gao, Yongsheng Publshed 2 Conference Ttle Proceedngs of the 2th Internatonal Conference on Pattern Recognton (ICPR 2) DOI

More information

PCA Based Gait Segmentation

PCA Based Gait Segmentation Honggu L, Cupng Sh & Xngguo L PCA Based Gat Segmentaton PCA Based Gat Segmentaton Honggu L, Cupng Sh, and Xngguo L 2 Electronc Department, Physcs College, Yangzhou Unversty, 225002 Yangzhou, Chna 2 Department

More information

Hierarchical Image Retrieval by Multi-Feature Fusion

Hierarchical Image Retrieval by Multi-Feature Fusion Preprnts (www.preprnts.org) NOT PEER-REVIEWED Posted: 26 Aprl 207 do:0.20944/preprnts20704.074.v Artcle Herarchcal Image Retreval by Mult- Fuson Xaojun Lu, Jaojuan Wang,Yngq Hou, Me Yang, Q Wang* and Xangde

More information

An Ensemble Learning algorithm for Blind Signal Separation Problem

An Ensemble Learning algorithm for Blind Signal Separation Problem An Ensemble Learnng algorthm for Blnd Sgnal Separaton Problem Yan L 1 and Peng Wen 1 Department of Mathematcs and Computng, Faculty of Engneerng and Surveyng The Unversty of Southern Queensland, Queensland,

More information

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines

A Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines A Modfed Medan Flter for the Removal of Impulse Nose Based on the Support Vector Machnes H. GOMEZ-MORENO, S. MALDONADO-BASCON, F. LOPEZ-FERRERAS, M. UTRILLA- MANSO AND P. GIL-JIMENEZ Departamento de Teoría

More information

An Efficient Face Detection Method Using Adaboost and Facial Parts

An Efficient Face Detection Method Using Adaboost and Facial Parts An Effcent Face Detecton Method Usng Adaboost and Facal Parts Yasaman Heydarzadeh, Abolfazl Torogh Haghghat Computer, IT and Electronc department Azad Unversty of Qazvn Tehran, Iran heydarzadeh@ qau.ac.r,

More information

Capturing Global and Local Dynamics for Human Action Recognition

Capturing Global and Local Dynamics for Human Action Recognition 2014 22nd Internatonal Conference on Pattern Recognton Capturng Global and Local Dynamcs for Human Acton Recognton Sq Ne Department of Electrcal, Computer and System Engneerng Rensselaer Polytechnc Insttute

More information

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson

More information

A Gradient Difference based Technique for Video Text Detection

A Gradient Difference based Technique for Video Text Detection A Gradent Dfference based Technque for Vdeo Text Detecton Palaahnakote Shvakumara, Trung Quy Phan and Chew Lm Tan School of Computng, Natonal Unversty of Sngapore {shva, phanquyt, tancl }@comp.nus.edu.sg

More information

Face Recognition Based on Neuro-Fuzzy System

Face Recognition Based on Neuro-Fuzzy System IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.9 No.4, Aprl 2009 39 Face Recognton Based on Neuro-Fuzzy System Nna aher Makhsoos, Reza Ebrahmpour and Alreza Hajany Department of

More information

Pictures at an Exhibition

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

More information

Feature Reduction and Selection

Feature Reduction and Selection Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components

More information

High Five: Recognising human interactions in TV shows

High Five: Recognising human interactions in TV shows PATRON-PEREZ ET AL.: RECOGNISING INTERACTIONS IN TV SHOWS 1 Hgh Fve: Recognsng human nteractons n TV shows Alonso Patron-Perez alonso@robots.ox.ac.uk Marcn Marszalek marcn@robots.ox.ac.uk Andrew Zsserman

More information

A Probabilistic Approach to Detect Urban Regions from Remotely Sensed Images Based on Combination of Local Features

A Probabilistic Approach to Detect Urban Regions from Remotely Sensed Images Based on Combination of Local Features A Probablstc Approach to Detect Urban Regons from Remotely Sensed Images Based on Combnaton of Local Features Berl Sırmaçek German Aerospace Center (DLR) Remote Sensng Technology Insttute Weßlng, 82234,

More information

Font Recognition in Natural Images via Transfer Learning

Font Recognition in Natural Images via Transfer Learning Font Recognton n Natural Images va Transfer Learnng Yzh Wang, Zhouhu Lan, Yngmn Tang, and Janguo Xao Insttute of Computer Scence and Technology, Pekng Unversty Abstract. Font recognton s an mportant and

More information

A Gradient Difference based Technique for Video Text Detection

A Gradient Difference based Technique for Video Text Detection 2009 10th Internatonal Conference on Document Analyss and Recognton A Gradent Dfference based Technque for Vdeo Text Detecton Palaahnakote Shvakumara, Trung Quy Phan and Chew Lm Tan School of Computng,

More information

CS246: Mining Massive Datasets Jure Leskovec, Stanford University

CS246: Mining Massive Datasets Jure Leskovec, Stanford University CS46: Mnng Massve Datasets Jure Leskovec, Stanford Unversty http://cs46.stanford.edu /19/013 Jure Leskovec, Stanford CS46: Mnng Massve Datasets, http://cs46.stanford.edu Perceptron: y = sgn( x Ho to fnd

More information

WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING.

WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING. WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING Tao Ma 1, Yuexan Zou 1 *, Zhqang Xang 1, Le L 1 and Y L 1 ADSPLAB/ELIP, School of ECE, Pekng Unversty, Shenzhen 518055, Chna

More information

Switching Convolutional Neural Network for Crowd Counting

Switching Convolutional Neural Network for Crowd Counting Swtchng Convolutonal Neural Network for Crowd Countng Deepak Babu Sam Shv Surya R. Venkatesh Babu Indan Insttute of Scence Bangalore, INDIA 560012 bsdeepak@grads.cds.sc.ac.n, shv.surya314@gmal.com, venky@cds.sc.ac.n

More information

Brushlet Features for Texture Image Retrieval

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

More information

Medical X-ray Image Classification Using Gabor-Based CS-Local Binary Patterns

Medical X-ray Image Classification Using Gabor-Based CS-Local Binary Patterns Medcal X-ray Image Classfcaton Usng Gabor-Based CS-Local Bnary Patterns Fatemeh Ghofran, Mohammad Sadegh Helfroush, Habbollah Danyal, Kamran Kazem Abstract As ntensty of medcal x-ray mages vares consderably

More information

Fast Sparse Gaussian Processes Learning for Man-Made Structure Classification

Fast Sparse Gaussian Processes Learning for Man-Made Structure Classification Fast Sparse Gaussan Processes Learnng for Man-Made Structure Classfcaton Hang Zhou Insttute for Vson Systems Engneerng, Dept Elec. & Comp. Syst. Eng. PO Box 35, Monash Unversty, Clayton, VIC 3800, Australa

More information