Hierarchical Image Retrieval by Multi-Feature Fusion

Size: px
Start display at page:

Download "Hierarchical Image Retrieval by Multi-Feature Fusion"

Transcription

1 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v Artcle Herarchcal Image Retreval by Mult- Fuson Xaojun Lu, Jaojuan Wang,Yngq Hou, Me Yang, Q Wang* and Xangde Zhang * College of Scences, Northeastern Unversty, Shenyang 089, Chna; luxaojun@mal.neu.edu.cn(x.l.), 60023@stu.neu.edu.cn(J.W.), @63.com (Y.H.), @qq.com (M.Y.), * Correspondence: wangqmath@mal.neu.edu.cn (Q.W.), zhangxangde@mal.neu.edu.cn (X.Z.); Tel.: Abstract: Amng at the problems that are poor generalzaton performance, low retreval accuracy and large tme consumpton of exstng content-based mage retreval system, the herarchcal mage retreval method based on mult feature fuson s proposed n ths paper. The retreval accuracy rates on Corel5, Ubeach and Holdays are 68.23(Top ), 3.73(N-S) and 88.20(mAp), respectvely. The expermental results show that the method proposed n ths paper can effectvely mprove the defcency of sngle feature retreval and save tme sgnfcantly n the premse of a small amount of loss of accuracy. eywords: Herarchcal search; Image retreval; Mult-feature fuson. Introducton Large-scale mage retreval based on vsual nformaton s a key technology n many researches. In order to acheve the purpose of retrevng mages, we need to descrbe the content of the mage at frst. There are a lot of methods to represent the mage content, such as color, texture, and so on [] [2]. From the perspectve of mage representaton, the features of mage content can be dvded nto two categores, local feature [3] and global feature [4]. For the occluded mage retreval task, the local feature has a better performance. However, t may return the mage whch s rrelevant to the query due to local feature only focus on part of the mage. The global feature depcts the overall feature dstrbuton of the mage. But t has a large amount of redundant nformaton, and retreval tasks tend to return mages that look smlarly but do not correlate. Therefore, gven a specfc query, the retreval system usng a sngle feature s often dffcult to meet the needs of users. To solve ths problem, the retreval system ntegratng multple features has attracted more and more attenton due to the characterstcs of multple complementary features. Mult-feature fuson can effectvely mprove the performance of the retreval system has been confrmed [5], and there are many researches on mult-feature fuson has been done[6] [7]. At present, the popular method can be dvded nto two categores: one s fuson on the feature level; the other s rank aggregaton. The former merges a varety of features nto a new feature for retreval whle the latter fuses retreval results of dfferent features n the rank stage. The retreval system based on two fuson methods can mprove the retreval precson. Ronald Fagn et al. [8] used a number of ndependent "voters" to sort the database mages based on ther smlarty to the query, and then combned the rankngs wth some effcent fuson algorthms. Orol Ramos et al. [9] used a non-bayesan probablstc framework to solve the problem of classfer combnaton, and got two lnear combnaton rules to mnmze the msclassfcaton rate under certan constrants. Shaotng Zhang et al. [0] proposed a graph-based query fuson method, whch re-ordered multple retreval sets by chan analyss on the fuson graph. On the one hand, compared wth the sngle feature retreval, mage retreval based on 207 by the author(s). Dstrbuted under a Creatve Commons CC BY lcense.

2 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 2 of mult-feature fuson ncreases the retreval tme. On the other hand, many researches on mage retreval have been carred out on large-scale datasets, whch may contan up to several mllon pctures, It s very tme-consumng to search for the mages we need from the massve mages. On ths ssue, Ja Deng et al. [] used a herarchcal relatonshp, whch s most sutable for large-scale problems. At the same tme, a new hash scheme s proposed, whch reduces the computatonal cost effcently. evn Ln [2] utlzed the hdden layer learnng bnary code to represent the potental concept of the mage, and used the herarchcal deep search, whch mproved the retreval performance to a great extent. Inspred by the above work, we present a herarchcal mage retreval system based on mult-feature fuson. Gven a query, frstly we extract the CNN feature, and pre-retreve wth the bnary CNN feature. Then, the Color, Lbp and GIST features are fused for second retreval to obtan the search results. Further, we apply our approach to several datasets, and demonstrate ts effectveness and scalablty. 2. Materals and Methods 2.. Datasets Corel-5[3] dataset contans 5,000 mages that are dvded nto 50 categores, such as beach, brd, jewelry, sunset, etc. Each category has 00 mages. Each mage s taken as query n turn, and the other 4999 mages are served as retreval lbrary. The precson of top-r mages, namely the rato of smlar mages n the returned mages, s used as the evaluaton standard of the retreval system. Ubench[4] dataset s released wth 2550 objects, and each object has 4 pctures taken from dfferent vsual angles. All the 0,200 mages were served as query mages. We use N-S score, the average number of smlar mages n Top-4, as a measurement of retreval performance. Holdays[5] dataset ncludes 49 personal holday pctures that are composed of 500 categores. The frst mage n each category s used as the query, and the remanng mages are relevant mages. map s used to evaluate the retreval performance s Color. For each mage, we compute 2,000-dm HSV hstogram. (H, S, V are 20, 0, 0). LBP. For each mage, we dvde the detecton wndow nto 6 * 6 small cells and extract 256-dmensonal lbp descrptors. GIST. Each mage s reszed to 28 * 28. We use 4 scales wth the number of orentatons (8, 8, 8, 8), and extract 52-dmensonal GIST descrptors. CNN. Here we use VGG network pre-traned on Imagenet, whch conssts of 3 convoluton layers, 5 poolng layers and 3 fully connected layers. For all the mages n the dataset, the fc6 layer (the frst fully connected layer) feature s obtaned by forward operaton. The output of ths layer contans a wealth of mage nformaton, whose dmenson s 4096 dmensons Our Method In the mage retreval based on mult-feature fuson, we manly focus on two aspects: one s how to determne the weght of each feature to mprove the retreval accuracy; the other s how to mprove the retreval effcency. Tradtonally, the weght of feature has two ways to be determned, the global weght and the

3 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 3 of adaptve weght. The former s an average value or s decded by experenced experts, whch leads the retreval system to have poor generalzaton performance and low retreval performance for dfferent retreval mages. The latter s derved from retreval feedback based on ths feature, whch s better than the global weght. However, n the sum or product fuson, the dstncton between good features and bad features s not obvous. If the weght of the bad features n the retreval work s large, t wll also reduce the retreval performance to a certan extent. The tradtonal mage retreval system based on mult-feature fuson s a sngle-layer mage retreval system. The retreval process s shown n Fgure. That s, the system receves a query mage and extracts a pluralty of features of query and mages n the mage database. Then multple features are fused to get the comprehensve smlarty measure. Fnally, the system returns relevant mages to the user accordng to the comprehensve smlarty measure. At each retreval, the retreval system needs to extract the features of all mages and calculate the smlarty between query and all mages n the mage database, whch may lead to low retreval effcency and poor adaptablty to the retreval task on the large-scale mage database. In order to overcome the shortcomngs of the tradtonal global weght method, a herarchcal mage retreval system based on mult-feature s proposed. The basc framework of the retreval system s shown n Fgure 2. Next, the herarchcal retreval and mult-feature fuson wll be descrbed n detal Herarchcal retreval As shown n Fgure 2, n the herarchcal retreval system proposed n ths paper, pre-retreval s performed based on the bnary CNN feature. Then Color, Lbp, GIST and bnary CNN features are fused to perform second search wthn the pre-search results. Pre-retreval. In the pre-retreval stage, the CNN feature of query mage and mages n mage database are extracted respectvely. The CNN feature from the mddle layer has a hgh dmenson, so t wll take a long tme to be carred out the smlarty measure by usng t. User Return Smlar Images Retreval Result Images Query Image... (feature) Smlarty... (Smlarty) Image Database... (feature) Smlarty Comprehensve Smlarty Measures Fgure. Tradtonal mage retreval system framework based on mult-feature fuson

4 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 4 of User Return smlar mages Retreval Result Gst Images Image Database Query Image Color Color Smlarty Bnarzaton CNN Bnarzaton CNN Lbp Smlarty Gst CNN Smlarty Pre-search Results Color Lbp Lbp Smlarty Gst Smlarty Comprehensve Smlarty Measures Fgure 2. The proposed herarchcal retreval system framework. The red box represents the pre-retreval process and the blue box represents the secondary retreval process. Accordng to the characterstcs of Hammng dstance operaton, ths paper bnarzes 4096-dmensonal CNN feature as follows: For each bt of feature, we output bnary codes F by: j j F ave( F ) 0 0 F = j n j 0 F ave( F ) 0 0 {,2,..., } () Here, ave( F ) s the mean of feature F 0 0, n s the feature dmenson Then, n the lght of feature F, the smlarty score between query and mages n mage database s calculated. We output top-n mages as the result of pre-retreval accordng to the smlarty score. Secondary-retreval. In the secondary-retreval stage, we adjust the smlarty measure based on CNN accordng to the smlarty score vector and the retreval results n the pre-retreval stage. Next we extract lbp, color, GIST features of query mage and mages n the pre-retreval results, and compute smlarty measure based on each feature. Then we obtan retreval performance of each feature by searchng smlar mages. We determne the weght of each feature n accordance wth t, and calculate the comprehensve smlarty measure. Fnally, the retreval results are obtaned accordng to the comprehensve smlarty measure. The herarchcal retreval system proposed n ths paper mproves the retreval effcency n two aspects. Frstly, we bnarze the hgh-dmensonal CNN feature and calculate the Hammng dstance to mprove the retreval effcency. Secondly, we fuse multple features to make accurate search wthn the pre-retreval results, whch reduces the retreval range and mproves the retreval effcency to a certan extent Mult-feature fuson Specfcally, features are fused, q s query mage, p s a database mage. The proposed fuson method s as follows.

5 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 5 of We normalze the features: F( p) = ( p (), p (2),..., p ( n)) ( {,2,..., } ) n p () j j= F( q) = ( q (), q (2),..., q ( n)) ( {,2,..., } ) n q () j j= (2) (3) Here, p () j s the j-th component of the -th feature of the database mage p. q () j s the j-th component of the -th feature of the query mage q. n s the dmenson of feature. We calculate the dstance between q and p and normalze t: n j= { } d ( k) = d ( q, p ) = q ( j) p ( j) k,2,..., m, {,2,..., } (4) k k D ( q) = ( d (), d (2),..., d ( m)) ( {,2,..., } ) m d ( k) k= (5) Here, D ( q) s the smlarty vector between the query mage q and the all database mages, whch s calculated based on -th feature. m s the total number of mages n the mage database. We calculate the comprehensve measure by fusng multple features: ~ () () q q = = + sm( q) w D ( q)+ w D ( q ) (6) = Here w, () q () w, q are the weght of a good feature, the weght of a bad feature, and the number of good features respectvely, when the query mage s q. Good features and bad features. ac ( q) s retreval performance, whch s obtaned by D ( q ), Good features and bad features are defned as follows: ac ( q) = ac _ mean = f ac ( q) > ac _ mean else feature { good _ feature} feature { bad _ feature} (7) Weght Determnaton. In order to make better use of good feature n gettng the comprehensve measure, after gettng the weghts, the paper carres on a power calculaton to the good weght, whch can ncrease the dfference of the good feature and the bad feature. The weghts of features are shown as follows:

6 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 6 of f feature { good _ feature} else ~ q W W q = exp( ac ( q) ) ac ( q) = = = ac ( q) ac ( q) (8) 3. Results 3.. The effectveness of feature fuson In ths paper, we conduct retreval experments based on maxmum fuson, multplcaton fuson [6] and sum fuson. And we experment wth adaptve weght and average global weght. The sum fuson s what we use n ths paper. The maxmum fuson and multplcaton fuson are shown as follows. The maxmum fuson: ~ sm( q) = arg max{max{ W =,2,..., },max{ W = +, + 2,..., }} (9) D ( q) The multplcaton fuson: q q ~ () () q q = = + sm( q) = w D ( q) w D ( q ) (0) The expermental results on the maxmum fuson, multplcaton fuson, and sum fuson are 86.5\87.74\88.20, whch are shown n Fgure 3(a). The expermental results on the adaptve weght and global average weght are shown n Fgure 3(b). From the Fgure 3(a) we can see that the sum fuson s much better than the other two. Furthermore, t can be seen from the Fgure 3(b) that the adaptve weght proposed n ths paper s of great performance. Table shows the performance comparson between the sngle-feature retreval and multple-feature retreval. As what can be seen from the table, the latter s performance s optmal. Expermental results show that the proposed adaptve mult-feature fuson method s effectve and has good performance. Fgure 3. (a) A comparson of the retreval performance based on the three fuson methods of maxmum fuson, product fuson and sum fuson. (b) The performance comparson between the adaptve weght proposed n ths paper and the global average weght.

7 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 7 of Table. On the Holdays dataset, the comparson of search results based on sngle features and fuson features s CNN color lbp GIST fuson map Retreval performance evaluaton On the Ubeach, Holdays, Corel-5 datasets,we carres out experments based on dfferent fuson features. The retreval results are shown n Fgure 4. From the fgure, we can see that on the Holdays dataset, fusng CNN, color, lbp, GIST features to search mage acheves the best performance. On the Ubeach dataset, optmal mage retreval performance can be obtaned by fusng CNN, color features. On the Corel-5 dataset, we get the best result by fusng CNN, color, lbp, GIST features. Fgure 4. On the Ubeach, Holdays, Corel-5 datasets, the experments based on dfferent fuson features are compared. Fgure (a) shows precson comparson retreved on the Corel5 dataset, here the number of smlar mages returned s 30. Fgure (b) shows the map comparson retreved on the Holdays dataset,. Fgure (c) shows the comparson of N-S values retreved on the Ubeach dataset. 3.3 The effectveness of herarchcal retreval On the Ubeach dataset, we conduct experments based on the drect retreval and herarchcal retreval respectvely. Fgure 5 shows that 00 mages are randomly selected as query mages. We compare the retreval tme based on bnary CNN feature and CNN feature, from what we can see that mage retreval based on the bnary CNN can acheve the effect of savng tme. Then we set the number of pre-retreved mages to 8 and randomly select 00 mages as query mage. Fgure 6 shows the comparson of the retreval tme and N-S score. Fgure 7 shows the query mages wth dfferent N_S value. Fgure 8 shows the retreval results obtaned by the herarchcal retreval and the drect retreval.

8 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 8 of Fgure 5. On the Ubeach dataset, 00 mages were randomly selected as query mages, the retreval tme of bnary CNN feature s compared to the retreval tme of CNN feature. Fgure 6. On Ubeach dataset, We randomly selected 00 mages as the retreval mage, and set the number of pre-retreved mages to 8. Comparson of drect and herarchcal retreval performance. (a)the comparson of retreval tme n each search. (b) The average tme of 00 searches comparson. (c) The N-S value of each search comparson. (d) The average N-S values of 00 searches comparson. Fgure 7. Query mages wth dfferent N-S value. (a) Query mages whose N-S value of the herarchcal retreval s greater than the drect retreval. (b) Query mages whose N-S value of the herarchcal retreval s less than the drect retreval.

9 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 9 of Fgure 8. The search results. The frst mage n the upper left corner s query mage, and the remanng mages are smlar mages. In accordance wth the smlarty from large to small, we arrange them from left to rght. The frst row s search results based on the drect retreval. The second row s search results based on the herarchcal retreval Comparson Table 2 shows the results of ths paper compared wth other papers. On the Corel-5 dataset, when the number of returned smlar mages s, the precson of our method s mproved by 4% compared wth the method n [3]. On the Ubeach dataset, the N-S value of our method s 3.73, whch s 0.04 lower than [3], but ncreased by about 0.8 compared wth [7] and [22], 0.3 compared wth [20], and 0.08 compared wth [9]. On the Holdays dataset, the map of mage retreval based on our method s 88.20%, whch s about 3.5% hgher than [3], [20] and [22]. Compared wth [8], [9] and [2], t ncreased by about 8%. Compared wth [7], ncreased by 8.9%. Table2. Comparson wth other methods wth post processng n mage retreval level. Model Corel5(top) Ubeach Holdays Ours Graph-densty [3] Global-CNN [7] MOP-CNN+PCA+Whtenng [8] Bag-8 + PCA [20] Dscusson In ths paper, we propose a herarchcal search method based on adaptve mult-feature fuson. Gven a query mage, frstly, we get the pre-retreval results accordng to bnary CNN feature. Then, we extract the color, lbp, GIST features of query mage and mages n the pre-retreval results and search mage based one every sngle feature respectvely. After gettng the accuracy of each retreval, we adjust ther weghts and gan the comprehensve measure. Fnally, the retreval results are obtaned accordng to the comprehensve measure. The proposed method has better effcency than tradtonal drect retreval, and mult-feature fuson can effectvely mprove the accuracy. In the comparson of the algorthms whch have acheved good performance n mage retreval, we fnd that our method has a good performance and generalzaton ablty for dfferent mage databases. In the future work, we wll consder usng our method n unsupervsed mage retreval work. Wth the ncrease of the number of network mages, there are many dffcultes n manual markng.

10 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v 0 of The tradtonal unsupervsed mage retreval system has low precson and weak generalzaton ablty. It s sgnfcant to consder the applcaton of our method n unsupervsed mage retreval. Acknowledgement:Ths research s partally supported by Natonal Natural Scence Foundaton of Chna (Grant No ). Author Contrbutons: X.Lu, Q.Wang, and X.Zhang conceved and desgned the experments; J.Wang and Y.Hou performed the experments; J.Wang, Y.Hou, and M.Yang analyzed the data; J.Wang and and Y.Hou wrote the paper. Conflcts of Interest: The authors declare no conflct of nterest. References. Hremath P S; Pujar J. Content Based Image Retreval Usng Color, Texture and Shape s. IEEE Computer Socety. 2007, [CrossRef] 2. Tamura B H; Mor S, Yamawak T. Texture features correspondng to vsual percepton. IEEE Trans. Syst. Man Cybern. 200, [CrossRef] 3. Lowe D G; Lowe D G. Dstnctve Image s from Scale-Invarant eyponts. Internatonal Journal of Computer Vson. 2004, 60(2):9-0.[ CrossRef] 4. Olva B A; Torralba. Modelng the shape of the scene: A holstc representaton of the spatal envelope. Int J Comput Vson. 200.,45 75.[CrossRef] 5. Zheng Y; Huang X; Feng S. An mage matchng algorthm based on combnaton of SIFT and the rotaton nvarant LBP. Journal of Computer-Aded Desgn & Computer Graphcs. 200, 22(2): [PubMed] 6. Yu J; Qn Z; Wan T; et al. ntegraton analyss of bag-of-features model for mage retreval. Neurocomputng. 203, 20(0): [CrossRef] 7. Wang X; Han T X; Yan S. An HOG-LBP human detector wth partal occluson handlng.computer Vson, 2009 IEEE 2th Internatonal Conference on. IEEE. 2009, [CrossRef] 8. Fagn; Ronald; umar; et al. Effcent smlarty search and classfcaton va rank aggregaton. Proceedngs of the 2003 ACM SIGMOD nternatonal conference on Management of data. 2003, [CrossRef] 9. Terrades O R; Valveny E; Tabbone S. Optmal Classfer Fuson n a Non-Bayesan Probablstc Framework. IEEE Transactons on Pattern Analyss & Machne Intellgence. 2008, 3(9): [CrossRef] 0. Zhang S; Yang M; Cour T; et al. Query specfc fuson for mage retreval. European Conference on Computer Vson. Sprnger-Verlag, 202, [CrossRef]. Ja D; Berg A C; L F F. Herarchcal semantc ndexng for large scale mage retreval. Computer Vson and Pattern Recognton (CVPR), 20 IEEE Conference on. 20, 32(4): [CrossRef] 2. Ln ; Yang H F; Hsao J H; et al. Deep learnng of bnary hash codes for fast mage retreval. IEEE Conference on Computer Vson and Pattern Recognton Workshops. 205, [CrossRef] 3. Zhang S; Yang M; Cour T; et al. Query specfc fuson for mage retreval. Computer Vson ECCV 202. Sprnger Berln Hedelberg. 202:, [CrossRef] 4. Nster D; Stewenus H. Scalable recognton wth a vocabulary tree IEEE Computer Socety Conference on Computer Vson and Pattern Recognton (CVPR'06). IEEE. 2006, 2: [CrossRef] 5. Jégou H; Douze M; Schmd C. Hammng embeddng and weak geometry consstency for large scale mage search-extended verson. 2008, HAL Id : nra , verson. [PubMed] 6. Zheng L; Wang S; Tan L; et al. Query-adaptve late fuson for mage search and person re-dentfcaton. Proceedngs of the IEEE Conference on Computer Vson and Pattern Recognton. 205, [CrossRef] 7. Babenko A; Slesarev A; Chgorn A; et al. Neural codes for mage retreval. European Conference on Computer Vson. 204, [CrossRef] 8. Gong Y; Wang L; Guo R; et al. Mult-scale orderless poolng of deep convolutonal actvaton features. European Conference on Computer Vson. 204, [CrossRef] 9. Babenko A; Lemptsky V. Aggregatng local deep features for mage retreval. Proceedngs of the IEEE Internatonal Conference on Computer Vson. 205, [CrossRef] 20. Perronnn F; Larlus D. Fsher vectors meet neural networks: A hybrd classfcaton archtecture. Proceedngs of the IEEE conference on computer vson and pattern recognton. 205, [CrossRef] 2. Zhang S; Yang M; Wang X; et al. Semantc-aware co-ndexng for mage retreval. Proceedngs of the IEEE Internatonal Conference on Computer Vson. 203, [CrossRef]

11 Preprnts ( NOT PEER-REVIEWED Posted: 26 Aprl 207 do: /preprnts v of 22. Jégou H; Douze M; Schmd C. Improvng bag-of-features for large scale mage search. Internatonal Journal of Computer Vson. 200, 87(3): [CrossRef]

Fusion of Deep Features and Weighted VLAD Vectors based on Multiple Features for Image Retrieval

Fusion of Deep Features and Weighted VLAD Vectors based on Multiple Features for Image Retrieval MATEC Web of Conferences, 0500 (07) DTS-07 DO: 005/matecconf/070500 Fuson of Deep Features and Weghted VLAD Vectors based on Multple Features for mage Retreval Yanhong Wang,, Ygang Cen,, Lequan Lang,*,

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

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

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

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

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School

More information

Image Alignment CSC 767

Image Alignment CSC 767 Image Algnment CSC 767 Image algnment Image from http://graphcs.cs.cmu.edu/courses/15-463/2010_fall/ Image algnment: Applcatons Panorama sttchng Image algnment: Applcatons Recognton of object nstances

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

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

A CALCULATION METHOD OF DEEP WEB ENTITIES RECOGNITION

A CALCULATION METHOD OF DEEP WEB ENTITIES RECOGNITION A CALCULATION METHOD OF DEEP WEB ENTITIES RECOGNITION 1 FENG YONG, DANG XIAO-WAN, 3 XU HONG-YAN School of Informaton, Laonng Unversty, Shenyang Laonng E-mal: 1 fyxuhy@163.com, dangxaowan@163.com, 3 xuhongyan_lndx@163.com

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

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

KIDS Lab at ImageCLEF 2012 Personal Photo Retrieval

KIDS Lab at ImageCLEF 2012 Personal Photo Retrieval KD Lab at mageclef 2012 Personal Photo Retreval Cha-We Ku, Been-Chan Chen, Guan-Bn Chen, L-J Gaou, Rong-ng Huang, and ao-en Wang Knowledge, nformaton, and Database ystem Laboratory Department of Computer

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

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

A Clustering Algorithm for Key Frame Extraction Based on Density Peak

A Clustering Algorithm for Key Frame Extraction Based on Density Peak Journal of Computer and Communcatons, 2018, 6, 118-128 http://www.scrp.org/ournal/cc ISSN Onlne: 2327-5227 ISSN Prnt: 2327-5219 A Clusterng Algorthm for Key Frame Extracton Based on Densty Peak Hong Zhao

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

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity Journal of Sgnal and Informaton Processng, 013, 4, 114-119 do:10.436/jsp.013.43b00 Publshed Onlne August 013 (http://www.scrp.org/journal/jsp) Corner-Based Image Algnment usng Pyramd Structure wth Gradent

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

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

UB at GeoCLEF Department of Geography Abstract

UB at GeoCLEF Department of Geography   Abstract UB at GeoCLEF 2006 Mguel E. Ruz (1), Stuart Shapro (2), June Abbas (1), Slva B. Southwck (1) and Davd Mark (3) State Unversty of New York at Buffalo (1) Department of Lbrary and Informaton Studes (2) Department

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

The Study of Remote Sensing Image Classification Based on Support Vector Machine

The Study of Remote Sensing Image Classification Based on Support Vector Machine Sensors & Transducers 03 by IFSA http://www.sensorsportal.com The Study of Remote Sensng Image Classfcaton Based on Support Vector Machne, ZHANG Jan-Hua Key Research Insttute of Yellow Rver Cvlzaton and

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

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

Robust Shot Boundary Detection from Video Using Dynamic Texture

Robust Shot Boundary Detection from Video Using Dynamic Texture Sensors & Transducers 204 by IFSA Publshng, S. L. http://www.sensorsportal.com Robust Shot Boundary Detecton from Vdeo Usng Dynamc Teture, 3 Peng Tale, 2 Zhang Wenjun School of Communcaton & Informaton

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

Modular PCA Face Recognition Based on Weighted Average

Modular PCA Face Recognition Based on Weighted Average odern Appled Scence odular PCA Face Recognton Based on Weghted Average Chengmao Han (Correspondng author) Department of athematcs, Lny Normal Unversty Lny 76005, Chna E-mal: hanchengmao@163.com Abstract

More information

Content-Based Bird Retrieval using Shape context, Color moments and Bag of Features

Content-Based Bird Retrieval using Shape context, Color moments and Bag of Features www.ijcsi.org 101 Content-Based Brd Retreval usng Shape context, Color moments and Features Bahr abdelkhalak 1 and hamd zouak 2 1 Faculty of Scences, Unversty Chouab Doukkal, Equpe: Modélsaton mathématque

More information

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented

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

CO-SPARSITY REGULARIZED DEEP HASHING FOR IMAGE INSTANCE RETRIEVAL I2R 1, UPMC 2

CO-SPARSITY REGULARIZED DEEP HASHING FOR IMAGE INSTANCE RETRIEVAL I2R 1, UPMC 2 CO-SPARSITY REGULARIZED DEEP HASHING FOR IMAGE INSTANCE RETRIEVAL Je Ln,1, Olver Morère,1,2, Vjay Chandrasekhar,1, Antone Vellard,2, Hanln Goh 1 I2R 1, UPMC 2 ABSTRACT In ths work, we tackle the problem

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

Remote Sensing Image Retrieval Algorithm based on MapReduce and Characteristic Information

Remote Sensing Image Retrieval Algorithm based on MapReduce and Characteristic Information Remote Sensng Image Retreval Algorthm based on MapReduce and Characterstc Informaton Zhang Meng 1, 1 Computer School, Wuhan Unversty Hube, Wuhan430097 Informaton Center, Wuhan Unversty Hube, Wuhan430097

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

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING An Improved K-means Algorthm based on Cloud Platform for Data Mnng Bn Xa *, Yan Lu 2. School of nformaton and management scence, Henan Agrcultural Unversty, Zhengzhou, Henan 450002, P.R. Chna 2. College

More information

A Simple Methodology for Database Clustering. Hao Tang 12 Guangdong University of Technology, Guangdong, , China

A Simple Methodology for Database Clustering. Hao Tang 12 Guangdong University of Technology, Guangdong, , China for Database Clusterng Guangdong Unversty of Technology, Guangdong, 0503, Chna E-mal: 6085@qq.com Me Zhang Guangdong Unversty of Technology, Guangdong, 0503, Chna E-mal:64605455@qq.com Database clusterng

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

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

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

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

TARGET RECOGNITION ALGORITHM BASED ON SALIENT CONTOUR FEATURE SEGMENTS

TARGET RECOGNITION ALGORITHM BASED ON SALIENT CONTOUR FEATURE SEGMENTS U.P.B. Sc. Bull., Seres C, Vol. 81, Iss. 1, 2019 ISSN 2286-3540 TARGET RECOGNITION ALGORITHM BASED ON SALIENT CONTOUR FEATURE SEGMENTS Janhu SONG 1, Yungong LI 2, Yanju LIU 3 *, Yang YU 4, Zhe YIN 5 A

More information

Query Clustering Using a Hybrid Query Similarity Measure

Query Clustering Using a Hybrid Query Similarity Measure Query clusterng usng a hybrd query smlarty measure Fu. L., Goh, D.H., & Foo, S. (2004). WSEAS Transacton on Computers, 3(3), 700-705. Query Clusterng Usng a Hybrd Query Smlarty Measure Ln Fu, Don Hoe-Lan

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Combination of Color and Local Patterns as a Feature Vector for CBIR

Combination of Color and Local Patterns as a Feature Vector for CBIR Internatonal Journal of Computer Applcatons (975 8887) Volume 99 No.1, August 214 Combnaton of Color and Local Patterns as a Feature Vector for CBIR L.Koteswara Rao Asst.Professor, Dept of ECE Faculty

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

MULTI-VIEW ANCHOR GRAPH HASHING

MULTI-VIEW ANCHOR GRAPH HASHING MULTI-VIEW ANCHOR GRAPH HASHING Saehoon Km 1 and Seungjn Cho 1,2 1 Department of Computer Scence and Engneerng, POSTECH, Korea 2 Dvson of IT Convergence Engneerng, POSTECH, Korea {kshkawa, seungjn}@postech.ac.kr

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

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 Recognition by Fusing Binary Edge Feature and Second-order Mutual Information

Face Recognition by Fusing Binary Edge Feature and Second-order Mutual Information Face Recognton by Fusng Bnary Edge Feature and Second-order Mutual Informaton Jatao Song, Bejng Chen, We Wang, Xaobo Ren School of Electronc and Informaton Engneerng, Nngbo Unversty of Technology Nngbo,

More information

Semantic Image Retrieval Using Region Based Inverted File

Semantic Image Retrieval Using Region Based Inverted File Semantc Image Retreval Usng Regon Based Inverted Fle Dengsheng Zhang, Md Monrul Islam, Guoun Lu and Jn Hou 2 Gppsland School of Informaton Technology, Monash Unversty Churchll, VIC 3842, Australa E-mal:

More information

Kernel Collaborative Representation Classification Based on Adaptive Dictionary Learning

Kernel Collaborative Representation Classification Based on Adaptive Dictionary Learning Internatonal Journal of Intellgent Informaton Systems 2018; 7(2): 15-22 http://www.scencepublshnggroup.com/j/js do: 10.11648/j.js.20180702.11 ISSN: 2328-7675 (Prnt); ISSN: 2328-7683 (Onlne) Kernel Collaboratve

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Computer Aided Drafting, Design and Manufacturing Volume 25, Number 2, June 2015, Page 14

Computer Aided Drafting, Design and Manufacturing Volume 25, Number 2, June 2015, Page 14 Computer Aded Draftng, Desgn and Manufacturng Volume 5, Number, June 015, Page 14 CADDM Face Recognton Algorthm Fusng Monogenc Bnary Codng and Collaboratve Representaton FU Yu-xan, PENG Lang-yu College

More information

Object-Based Techniques for Image Retrieval

Object-Based Techniques for Image Retrieval 54 Zhang, Gao, & Luo Chapter VII Object-Based Technques for Image Retreval Y. J. Zhang, Tsnghua Unversty, Chna Y. Y. Gao, Tsnghua Unversty, Chna Y. Luo, Tsnghua Unversty, Chna ABSTRACT To overcome the

More information

Video Copy Detection Based on Fusion of Spatio-temporal Features

Video Copy Detection Based on Fusion of Spatio-temporal Features Vdeo Copy Detecton Based on Fuson of Spato-temporal Features BAO We, JI Lxn, GAO Shln, LI Xng, Lu Lxong Natonal Dgtal swtchng System Engneerng & Technologcal R&D Center Zhengzhou, Chna 3202004075@163.com

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

SRBIR: Semantic Region Based Image Retrieval by Extracting the Dominant Region and Semantic Learning

SRBIR: Semantic Region Based Image Retrieval by Extracting the Dominant Region and Semantic Learning Journal of Computer Scence 7 (3): 400-408, 2011 ISSN 1549-3636 2011 Scence Publcatons SRBIR: Semantc Regon Based Image Retreval by Extractng the Domnant Regon and Semantc Learnng 1 I. Felc Raam and 2 S.

More information

Research and Application of Fingerprint Recognition Based on MATLAB

Research and Application of Fingerprint Recognition Based on MATLAB Send Orders for Reprnts to reprnts@benthamscence.ae The Open Automaton and Control Systems Journal, 205, 7, 07-07 Open Access Research and Applcaton of Fngerprnt Recognton Based on MATLAB Nng Lu* Department

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

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

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively Regularized Nearest Points for Set Based Recognition Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,

More information

Feature-based image registration using the shape context

Feature-based image registration using the shape context Feature-based mage regstraton usng the shape context LEI HUANG *, ZHEN LI Center for Earth Observaton and Dgtal Earth, Chnese Academy of Scences, Bejng, 100012, Chna Graduate Unversty of Chnese Academy

More information

Face Recognition using 3D Directional Corner Points

Face Recognition using 3D Directional Corner Points 2014 22nd Internatonal Conference on Pattern Recognton Face Recognton usng 3D Drectonal Corner Ponts Xun Yu, Yongsheng Gao School of Engneerng Grffth Unversty Nathan, QLD, Australa xun.yu@grffthun.edu.au,

More information

PRÉSENTATIONS DE PROJETS

PRÉSENTATIONS DE PROJETS PRÉSENTATIONS DE PROJETS Rex Onlne (V. Atanasu) What s Rex? Rex s an onlne browser for collectons of wrtten documents [1]. Asde ths core functon t has however many other applcatons that make t nterestng

More information

HCMX: AN EFFICIENT HYBRID CLUSTERING APPROACH FOR MULTI-VERSION XML DOCUMENTS

HCMX: AN EFFICIENT HYBRID CLUSTERING APPROACH FOR MULTI-VERSION XML DOCUMENTS HCMX: AN EFFICIENT HYBRID CLUSTERING APPROACH FOR MULTI-VERSION XML DOCUMENTS VIJAY SONAWANE 1, D.RAJESWARA.RAO 2 1 Research Scholar, Department of CSE, K.L.Unversty, Green Felds, Guntur, Andhra Pradesh

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

Discriminative Dictionary Learning with Pairwise Constraints

Discriminative Dictionary Learning with Pairwise Constraints Dscrmnatve Dctonary Learnng wth Parwse Constrants Humn Guo Zhuoln Jang LARRY S. DAVIS UNIVERSITY OF MARYLAND Nov. 6 th, Outlne Introducton/motvaton Dctonary Learnng Dscrmnatve Dctonary Learnng wth Parwse

More information

AUTOMATIC RECOGNITION OF TRAFFIC SIGNS IN NATURAL SCENE IMAGE BASED ON CENTRAL PROJECTION TRANSFORMATION

AUTOMATIC RECOGNITION OF TRAFFIC SIGNS IN NATURAL SCENE IMAGE BASED ON CENTRAL PROJECTION TRANSFORMATION AUTOMATIC RECOGNITION OF TRAFFIC SIGNS IN NATURAL SCENE IMAGE BASED ON CENTRAL PROJECTION TRANSFORMATION Ka Zhang a, Yehua Sheng a, Pefang Wang b, Ln Luo c, Chun Ye a, Zhjun Gong d a Key Laboratory of

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

A SCENE RECOGNITION ALGORITHM BASED ON MULTI-INSTANCE LEARNING

A SCENE RECOGNITION ALGORITHM BASED ON MULTI-INSTANCE LEARNING A SCENE RECOGNITION ALGORITHM BASED ON MULTI-INSTANCE 2 1 Tao Wang, 2 Wenqng Chen and 3 Balng Wang 1 Department of Computer Scence and Technology,Shaoxng Unversty College of of Computer Informaton Scence

More information

Face Recognition Method Based on Within-class Clustering SVM

Face Recognition Method Based on Within-class Clustering SVM Face Recognton Method Based on Wthn-class Clusterng SVM Yan Wu, Xao Yao and Yng Xa Department of Computer Scence and Engneerng Tong Unversty Shangha, Chna Abstract - A face recognton method based on Wthn-class

More information

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

More information

SIFT Match Verification by Geometric Coding for Large-Scale Partial-Duplicate Web Image Search

SIFT Match Verification by Geometric Coding for Large-Scale Partial-Duplicate Web Image Search SIFT Match Verfcaton by Geometrc Codng for Large-Scale Partal-Duplcate Web Image Search WENGANG ZHOU and HOUQIANG LI, Unversty of Scence and Technology of Chna YIJUAN LU, Texas State Unversty QI TIAN,

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

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM PERFORMACE EVALUAIO FOR SCEE MACHIG ALGORIHMS BY SVM Zhaohu Yang a, b, *, Yngyng Chen a, Shaomng Zhang a a he Research Center of Remote Sensng and Geomatc, ongj Unversty, Shangha 200092, Chna - yzhac@63.com

More information

3D Face Recognition Fusing Spherical Depth Map and Spherical Texture Map

3D Face Recognition Fusing Spherical Depth Map and Spherical Texture Map Journal of Computer and Communcatons, 14, *, ** Publshed Onlne **** 14 n ScRes. http://www.scrp.org/journal/jcc http://dx.do.org/1.436/jcc.14.***** 3D Face Recognton Fusng Sphercal Depth Map and Sphercal

More information

RESEARCH ON EQUIVALNCE OF SPATIAL RELATIONS IN AUTOMATIC PROGRESSIVE CARTOGRAPHIC GENERALIZATION

RESEARCH ON EQUIVALNCE OF SPATIAL RELATIONS IN AUTOMATIC PROGRESSIVE CARTOGRAPHIC GENERALIZATION RESEARCH ON EQUIVALNCE OF SPATIAL RELATIONS IN AUTOMATIC PROGRESSIVE CARTOGRAPHIC GENERALIZATION Guo Qngsheng Du Xaochu Wuhan Unversty Wuhan Unversty ABSTRCT: In automatc cartographc generalzaton, the

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and

More information

International Conference on Applied Science and Engineering Innovation (ASEI 2015)

International Conference on Applied Science and Engineering Innovation (ASEI 2015) Internatonal Conference on Appled Scence and Engneerng Innovaton (ASEI 205) Desgn and Implementaton of Novel Agrcultural Remote Sensng Image Classfcaton Framework through Deep Neural Network and Mult-

More information

Professional competences training path for an e-commerce major, based on the ISM method

Professional competences training path for an e-commerce major, based on the ISM method World Transactons on Engneerng and Technology Educaton Vol.14, No.4, 2016 2016 WIETE Professonal competences tranng path for an e-commerce maor, based on the ISM method Ru Wang, Pn Peng, L-gang Lu & Lng

More information

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram

Shape Representation Robust to the Sketching Order Using Distance Map and Direction Histogram Shape Representaton Robust to the Sketchng Order Usng Dstance Map and Drecton Hstogram Department of Computer Scence Yonse Unversty Kwon Yun CONTENTS Revew Topc Proposed Method System Overvew Sketch Normalzaton

More information

ABSTRACT 1. INTRODUCTION

ABSTRACT 1. INTRODUCTION Robust SIFT-Based Descrptor for Vdeo Classfcaton Razyeh Salarfard, Mahshd lsadat Hossen, Mahmood Karman and Shohreh Kasae Department of Computer Engneerng, Sharf Unversty of Technology Tehran, Iran BSTRCT

More information

Recommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm

Recommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm Recommended Items Ratng Predcton based on RBF Neural Network Optmzed by PSO Algorthm Chengfang Tan, Cayn Wang, Yuln L and Xx Q Abstract In order to mtgate the data sparsty and cold-start problems of recommendaton

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

An Evaluation of Divide-and-Combine Strategies for Image Categorization by Multi-Class Support Vector Machines

An Evaluation of Divide-and-Combine Strategies for Image Categorization by Multi-Class Support Vector Machines An Evaluaton of Dvde-and-Combne Strateges for Image Categorzaton by Mult-Class Support Vector Machnes C. Demrkesen¹ and H. Cherf¹, ² 1: Insttue of Scence and Engneerng 2: Faculté des Scences Mrande Galatasaray

More information

1. Introduction. Abstract

1. Introduction. Abstract Image Retreval Usng a Herarchy of Clusters Danela Stan & Ishwar K. Seth Intellgent Informaton Engneerng Laboratory, Department of Computer Scence & Engneerng, Oaland Unversty, Rochester, Mchgan 48309-4478

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

Fast Computation of Shortest Path for Visiting Segments in the Plane

Fast Computation of Shortest Path for Visiting Segments in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 4 The Open Cybernetcs & Systemcs Journal, 04, 8, 4-9 Open Access Fast Computaton of Shortest Path for Vstng Segments n the Plane Ljuan Wang,, Bo Jang

More information

Orthogonal Complement Component Analysis for Positive Samples in SVM Based Relevance Feedback Image Retrieval

Orthogonal Complement Component Analysis for Positive Samples in SVM Based Relevance Feedback Image Retrieval Orthogonal Complement Component Analyss for ostve Samples n SVM Based Relevance Feedback Image Retreval Dacheng Tao and Xaoou Tang Department of Informaton Engneerng The Chnese Unversty of Hong Kong {dctao2,

More information

Clustering Algorithm of Similarity Segmentation based on Point Sorting

Clustering Algorithm of Similarity Segmentation based on Point Sorting Internatonal onference on Logstcs Engneerng, Management and omputer Scence (LEMS 2015) lusterng Algorthm of Smlarty Segmentaton based on Pont Sortng Hanbng L, Yan Wang*, Lan Huang, Mngda L, Yng Sun, Hanyuan

More information

A Similarity Measure Method for Symbolization Time Series

A Similarity Measure Method for Symbolization Time Series Research Journal of Appled Scences, Engneerng and Technology 5(5): 1726-1730, 2013 ISSN: 2040-7459; e-issn: 2040-7467 Maxwell Scentfc Organzaton, 2013 Submtted: July 27, 2012 Accepted: September 03, 2012

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

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

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

What Is the Most Efficient Way to Select Nearest Neighbor Candidates for Fast Approximate Nearest Neighbor Search?

What Is the Most Efficient Way to Select Nearest Neighbor Candidates for Fast Approximate Nearest Neighbor Search? IEEE Internatonal Conference on Computer Vson What Is the Most Effcent Way to Select Nearest Neghbor Canddates for Fast Approxmate Nearest Neghbor Search? Masakazu Iwamura, Tomokazu Sato and Koch Kse Graduate

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

Data-dependent Hashing Based on p-stable Distribution

Data-dependent Hashing Based on p-stable Distribution Data-depent Hashng Based on p-stable Dstrbuton Author Ba, Xao, Yang, Hachuan, Zhou, Jun, Ren, Peng, Cheng, Jan Publshed 24 Journal Ttle IEEE Transactons on Image Processng DOI https://do.org/.9/tip.24.2352458

More information

Querying by sketch geographical databases. Yu Han 1, a *

Querying by sketch geographical databases. Yu Han 1, a * 4th Internatonal Conference on Sensors, Measurement and Intellgent Materals (ICSMIM 2015) Queryng by sketch geographcal databases Yu Han 1, a * 1 Department of Basc Courses, Shenyang Insttute of Artllery,

More information