Accurate Overlay Text Extraction for Digital Video Analysis

Size: px
Start display at page:

Download "Accurate Overlay Text Extraction for Digital Video Analysis"

Transcription

1 Accurate Overlay Text Extracton for Dgtal Vdeo Analyss Dongqng Zhang, and Shh-Fu Chang Electrcal Engneerng Department, Columba Unversty, New York, NY (Emal: dqzhang, Abstract Ths report descrbes a system to detect and extract the overlay texts n dgtal vdeo. Dfferent from the prevous approaches, the system used a multple hypothess testng approach: The regon-of-nterests (ROI) probably contanng the overlay texts are decomposed nto several hypothetcal bnary mages usng color space parttonng; A groupng algorthm then s conducted to group the dentfed character blocks nto text lnes n each bnary mage; If the layout of the grouped text lnes conforms to the verfcaton rules, the boundng boxes of these grouped blocks are output as the detected text regons. Fnally, moton verfcaton s used to reduce false alarms. In order to acheve real tme speed, ROI localzaton s realzed usng compressed doman features ncludng DCT coeffcents and moton vectors n MPEG vdeos. The proposed method showed mpressve results wth average recall 96.9% and precson 71.6% n testng on dgtal News vdeos. 1. Introducton Vdeotext detecton and recognton has been dentfed as one of the key components for the vdeo retreval and analyss system. Vdeotext detecton and recognton can be used n many applcatons, such as semantc vdeo ndexng, summarzaton, vdeo survellance and securty, multlngual vdeo nformaton access, etc. Vdeotext can be classfed nto two broad categores: Graphc text and scene text. Graphc text or text overlay s the vdeotext added mechancally by vdeo edtors, examples nclude the news/sports vdeo capton, move credts etc. Scene texts, are the vdeotexts embedded n the real-world objects or scenes, examples nclude street name, car lcense number, the number/name on the back of a soccer player. Ths report s to address the problem of accurately detectng and extractng the graph vdeotexts for vdeotext recognton. Although the overlay text s manually added nto the vdeo, the experments showed they are even as hard to extract as many vdeo objects, such as face, people etc. Ths s due to the followng reasons: 1. Many overlay texts present n the cluttered scene background; 2. There s no consstent color dstrbuton for texts n dfferent vdeos. Consequently, the color-tone based approach wdely used n face or people detecton applcaton actually cannot be appled n text detecton.; 3. The sze of the text regons may be very small such that when the color segmentaton based approach s appled, the small text regon may merge nto the large non-text regons n ts vcnty. There has been much pror work for vdeotext detecton and extracton. T.Sato et al. [1] nvestgated supermposed capton recognton n News vdeo. They use spatal dfferental flter to localze the text regon, and sze/poston constrants to refne the

2 detected area. Ther algorthm s used n the specfc doman, such as CNN news. R.Lenhart et al. [2] gave an approach usng texture, color segmentaton, contrast segmentaton, and moton analyss. In ther system, color segmentaton by regon mergng s done n global vdeo frame wthout a localzaton process. Thus the accuracy of the extracton may heavly rely on the segmentaton accuracy. L et al. [3] uses Haar wavelet to decompose the vdeo frame nto subband mages. Neural network s used to classfy the mage blocks nto text or non-text based on the subband mages. The Haar wavelet flterng s essentally a process of texture energy extracton. Y. Zhong [4] et al, employs DCT coeffcents to localze text regons n MPEG vdeo I-frame. They dd not use color nformaton and layout verfcaton, therefore false alarms are nevtable n those texture-lke objects, such as buldng, crowd. M. Bertn [5] presented a text locaton method usng salent corner detecton. Salent corner detecton may produce false postves n the cluttered background wthout good verfcaton process. J. Shm and C. Dora et al. [6] uses a regon-based approach for text detecton. They use gray level mages generated from the color vdeo frames. And texture features are not used n detecton. L. Agnhotr and N. Dmtrova.[7] uses a texture-based approach to locate the text regon, wthout usng the color segmentaton or decomposton. A. Jan [8] presented a method for text locaton n mage and vdeo frames. They use color space decomposton and ntegraton, texture and moton model s not used for detecton. And a smple layout verfcaton method s used based on vertcal projecton profle. Some of these work dd not explctly address the problem of accurate text boundary extracton. But accurate boundary extracton of vdeotexts s mportant for recognton, because the recognton error due to ll extracton s often unrecoverable usng enhancement or language/knowledge model. Our method attempts to address the lmtatons of the prevous systems by avodng background dsturbance and reducng the false alarms. We use a multple hypothess testng approaches: The regon-of-nterests (ROI) probably contanng the overlay texts are decomposed nto several hypothetcal bnary mages usng color space parttonng; A groupng algorthm then s conducted to cluster the dentfed character blocks nto text lnes n each bnary mage; If the layout of the grouped text lnes conforms to the verfcaton rules, the boundng boxes of these grouped blocks are output as the detected text regons. Moton verfcaton s also used to reduce false alarms. In order to acheve real tme speed, ROI localzaton s realzed usng compressed doman features ncludng DCT coeffcents and moton vectors n MPEG vdeos. The overall system can be llustrated n the followng dagram: Vdeo Localzaton by Texture&Moton Color Space Parttonng Block Groupng &Layout Analyss Temporal Verfcaton Text Block Fgure 1. System Flowchart In the dagram, texture and moton analyss s used to localze the regon-of-nterests (ROIs) of the vdeotexts by extractng texture and moton energy from compressed doman features. The color space parttonng s to dvde the HSV color space nto a few of parttons, and the hypothetcal bnary mages are generated for each partton.

3 Character block groupng and layout analyss cluster the character-lke blocks nto text regons and layout analyss s performed to verfy f the text regons are able to form text lnes. The temporal consstency analyss then s conducted to elmnate the false alarms stayng n the vdeo frames wth too short duraton. The report s organzed as follows: secton 2 descrbes the localzaton algorthm to detect the Regon of Interests. Secton 3 descrbes the hypothetcal bnary mage generaton usng color space parttonng. Secton 4 descrbes a rule-based approach for character block groupng and usng layout analyss to do verfcaton. Secton 5 presents the temporal verfcaton procedure to elmnate false alarms. 2. Localzaton usng Compressed Doman Features A typcal sze of the vdeo frame s 320x240. Wthout the localzaton of the nterest regon, the detecton program s dffcult to realze realtme speed. Furthermore, the localzaton usng texture or moton features can flter out rrelevant regons that would result n false alarms. However, capablty of the moton texture based approach to extract the accurate boundary of the text regon s usually poor, therefore an algorthm relyng alone on these features may not be suted for accurate text detecton. The problem becomes more severe f the background s cluttered. 2.1 Texture Energy and Moton Energy Texture features may be the most wdely used features for vdeotext detecton. The ntuton behnd texture based approach s vdeotexts are often wth hgh contrast aganst ther background and wth sharp stroke edges. These features make the text lne hold hgh energy n the hgh frequency band of the Fourer spectrum. For many vdeos, such as news and sports vdeo, usng texture feature alone s able to detect most of the text regon from the vdeo frames, snce texts n these vdeos are delberately rendered wth hgh contrast for the audence. The method used here s smlar to that used by Y. Zhong and Jans [4]. The texture energy n the horzontal and vertcal drecton s extracted from the 8x8 DCT coeffcent block n a MPEG-1 vdeo: Eh ( x, y) = C 0 k ( x, y) E ( x, y) = v 1 k 6 C 1 k 6 k0 ( x, y) (1) Where,j s the coordnates of a DCT block. E h s called the horzontal Texture Energy Map and E v s called the vertcal Texture Energy Map. For some vdeo genres, lke news and sports, most of the vdeotexts are statc. Thus the moton features can also be used n text extracton for those statc text blocks, the method has been successfully used n [9] for extracton of sports vdeo score box. Here a measurement called Moton Energy (ME) s used to characterze the moton ntensty of the regons. ME s the length of the moton vector of each macro block n the B or P frame. All Moton Energes of macro blocks n B or P frame form Moton Energy Map (MEM), whch exhbts the moton ntensty at dfferent locatons n a vdeo frame.

4 2.2. Combnng Texture and Moton Energy Moton energy map s often unstable due to the naccuracy of the MPEG moton estmaton algorthm. Therefore one cannot rely alone on the features extracted from the MPEG moton vector. We use a jont measurement combnng both texture and moton energy. The combnaton occur n each I frame, where the DCT features come from the current I-frame, and the moton features come from the latest B or P frame (temporal morphologcal flter can be used to stablze the MEM usng multple B and P frames). Snce the moton energy map s extracted from the macro block. Ther resoluton s half of the texture map. In order to be combned wth texture energy map, they are upscaled to the dentcal sze wth the texture map. The combnaton makes the hgh jont measurement correspondng to hgh texture energy and low moton energy, thus we have the followng equaton: TM E > λ ) ( E > λ ) ( U ( E,2) < ) (2) = ( h 1 v 2 m λ3 Where E m the moton energy map, U ( E m,2) s the operator of upscalng by 2, whch s actually upsamplng followed by an nterpolaton operator. The constants λ 1, λ 2, λ 3 are the thresholds for the horzontal, vertcal texture energy and moton energy. The bnarzed map may be wth rregular boundary, but they can be further de-nosed usng morphologcal flters. Fgure 2 llustrate the texture energy map, moton energy map and combned measurements. (a) (b) Fg 2. Localzaton by Texture-Moton Analyss. (a) Frame wth Intensve moton (zoom n) (b) Frame wthout any moton. From left to rght: orgnal mage; texture energy map (after tresholdng); moton energy map (after negatve, upscalng, thresholdng); combned measurement (before morphologcal flter). 3. Color Space Parttonng for Hypothetcal Bnary Image Generaton

5 The dea of color space parttonng s based on the fact that most of the text overlays are of unform color or near unform color. Ths also means the color dstrbuton of a vdeotext lne s very localzed n the color space. On the hand, the background s often rendered wth hgh color contrast to the text overlay. Thus parttonng the color space nto a few of sub regons can separate the color layer of the text overlay from ts background. The color space based approach has been used by prevous algorthms, for example A.K.Jan [8] uses color space decomposton to separate the layer of the texts n mages. Also some approaches use color segmentaton method n color space or gray space, such as [2][6]. A problem of color segmentaton s that the number of the color clusters s hard to determne a pror. Another problem s the texts wth too small szes may be merged to the background object f the number of the color cluster s not selected properly. Here we use a straghtforward approach: we frst convert the RGB color space nto HSV color space. Afterwards we dvde the whole HSV space nto n m l cubes. Whch means the H drecton s dvded nto n segments, S drecton m segments, V drecton l segments (segments can be slghtly overlapped). Suppose a color vector n color space s denoted as C = ( H C, SC, VC ). Then the partton can be represented as color range from C to C. A bnary mage s generated for each color space partton as: B = I C ) ( I < C ) (3) ( Where I s the gven color mage. These bnary mages wll be verfed usng the character block groupng and layout analyss. 4. Character block groupng and Layout Analyss The dea of usng layout analyss s to verfy the bnary text regons n each generated bnary mage, and a text regon s dentfed f one of these bnary mages contans texts. The layout analyss procedure ncludes three phases: connected component analyss, groupng of the character blocks, layout verfcaton. In each bnary mage, the connected component analyss (CCA) algorthm wll be frst performed to generate the character blocks. The character block s the outmost boundng box of a gven bnary component after usng CCA. Sze and shape flters wll be performed to elmnate the blocks wth too small or too large szes, and regons wth abnormal aspect rato. The groupng algorthm uses the followng rules to cluster two blocks: 1. The blocks should be algned such that they are wth the same bottom lne. 2. The heghts of the blocks should be close to each other. 3. The blocks should be close enough, whch means the nearest dstance of two blocks should not exceed certan rato of the average heght. The groupng procedure would group the ndvdual blocks nto hypothetcal text lnes. Then the verfcaton procedure s performed n these text lnes. We use a smple verfcaton process: The character number n a text lne should exceed certan constant.

6 The typcal value of the constant s 3, because the lengths of most words n Englsh are larger than 3. One example of groupng and layout analyss s shown n Fgure 3. (1) Orgnal Vdeo Frame (2) ROIs after Localzaton (3) Text Detecton Results usng groupng and layout analyss (4) Character Groupng and layout analyss n ROIs Fgure 3. Character Block Groupng and Layout Analyss. 5. Temporal Consstency Verfcaton Multple frame verfcaton s to flter out the detecton false alarms usng temporal consstency verfcaton of the boundng boxes correspondng to the text lne regons. A text lne on vdeo usually stays on the vdeo wth a sgnfcantly long tme. But the false alarm regons are usually transent n one or two I-frames. Temporal consstency verfcaton s performed for the detected boundng boxes. It calculates the overlap rato of each boundng box n the current I-frame wth all boundng boxes n the prevous I-frame. And take the boundng box wth the largest overlap rato as the most lkely match (MLM). The overlap rato s calculated as followng: O( r, r ) 1 2 2a( r1 r2 ) a( r ) + a( r ) = (4) Where a(r) s the area of regon r. If the overlap rato wth the MLM s sgnfcant (larger than certan constant), then the two boundng boxes are regarded as temporarly consstent. N Boundng boxes are sad to be n-consstent f each of them are temporarly consstent wth the boundng box n the nearest I-frames n n consecutve I-frames. A vdeotext lne s detected f the text lne s n-consstent, where n s larger than a constant, otherwse t s classfed as a false alarm. 6. Experments The algorthm s tested usng NIST TREC-2002 benchmark, and capton detecton n News vdeo. TREC Vdeo Track s an open metrc-based evaluaton system amng at promotng communcaton and progress n dgtal vdeo retreval feld. The task of TREC-2002 s to 1 2

7 ndex and retreve TREC vdeo based on concept detecton (ncludng face, ndoor, outdoor, text overlay etc.) and query-retreval framework. The benchmark used hours vdeos for the development of concept detectors and 5.02 hours vdeos as the Feature Test Set (FTS) (.e. for feature extracton evaluaton). The testng results usng our developed method on 5 hours Feature Valdate (FV) set n TREC-2002 vdeo data showed average precson, whch compares wth the average precson of Dora, Shm and Bolle s system (DSB system) [6][11], and average precson of B.Tseng, C.Ln and J.Smth s fuson system results [10], whch combnes the descrbed system and the DSB system. The average precson of our system on the TREC Test Set s , whle the DSB system acheves precson of and the fused system [10] acheves precson. 1 Fgure 5 shows some text detecton results on TREC-2002 vdeo data. Fgure 5. Overlay Text Detecton Example of TREC 2002 Vdeos Yet another testng excrement s conducted on News vdeos, News vdeos nclude a large amount of overlay texts. The detecton and recognton of the text overlay s very useful for News vdeo retreval. The text overlay n News vdeos may present n varous postons, thus the fxed poston assumpton used n our sports vdeo applcaton [] does not work very well for general overlay text detecton. The expermental data nclude four News vdeos, three of whch are US news vdeos from three dfferent channels, one of whch s Tawan News vdeo. The overall length of the vdeos s 2.13 hours. The capton styles on these vdeos are dfferent from each other as well as fonts. These vdeos are of MPEG-1 format wth 320x240 resolutons. The overall length of vdeo s about 2 hours. The followng table shows the detecton results. Some of the detecton results are shown n Fgure 6. Table 1. Text Overlay Detecton n News Vdeo US 1 (Ch 7) US 2 (Ch 11) US 3 (Ch 2) TW 1 (TTV) Average Recall 97.2% (105/108) 95.7% (90/94) 95.7%(90/94) 98.1%(154/157) 96.9% Precson 77.2% (105/136) 58.4% (90/154) 62.5%(90/144) 86.3%(154/179) 71.6% Legends: Ch: Channel, US: Unted States, TW: Tawan The experments showed that the performance on News vdeo s better than the performance n TREC vdeo. Ths s because most of overlay texts n News vdeos are 1 Acknowledgement to Dr. Belle T. Tseng and Dr. Chng-Yung Ln to provde TREC-2002 text overlay testng results

8 wth clean background and hgh contrast. Fgure 6.shows some examples of text detecton results n News vdeo. Fgure 6. Examples of Overlay Text Detecton Results on Dgtal News Vdeo 7. Conclusons and Future Work The report gves a rule-based approach to detect and extract the text lnes on the vdeo frames. The system ncludes the compressed doman feature processng to extract Regon of Interests, color space parttonng, regon groupng and layout analyss, At last temporal verfcaton s used to enhance the stablty and reduce the false alarms. Future work s to extend the rule-based approach to the probablstc framework to general the accuracy of the whole system. 8. Acknowledgement Acknowledgement to Dr. Belle L. Tseng and Dr. Chng-Yung Ln, IBM Watson Research Center for helpful dscusson and testng results on TREC-2002 vdeo data. 9. Reference [1] T. Sato, T. Kanade, E. Hughes, and M. Smth, "Vdeo OCR: Indexng Dgtal News Lbrares by Recognton of Supermposed Captons", Multmeda Systems, 7: , [2] R.Lenhart, W. Effelsberg, "Automatc text segmentaton and text recognton for vdeo ndexng", Multmeda System, [3] H. L, D. Doermann, O. Ka, Automatc text detecton and trackng n dgtal vdeo, IEEE Transacton. on Image processng, Vol 9, No. 1, January [4] Y. Zhong, H, Zhang, and A. K.Jan, "Automatc Capton Localzaton n Compressed Vdeo", IEEE Trans on PAMI, Vol 22. No.4 Aprl [5] M. Bertn, C. Colombo and A. D. Bmbo, "Automatc capton localzaton n vdeos usng salent ponts". In Proceedngs IEEE Internatonal Conference on Multmeda and Expo ICME 2001, Toko, Japan, August 2001.

9 [6] J.C. Shm, C. Dora and R. Bolle, Automatc Text Extracton from Vdeo for Content-Based Annotaton and Retreval," n Proc. 14th Internatonal Conference on Pattern Recognton, vol 1, pp , Brsbane, Australa, August [7] L. Agnhotr and N. Dmtrova. Text Detecton n Vdeo Segments. Proc. Of workshop on Content Based access to Image and Vdeo Lbrares, pp , June [8] A.K. Jan and B.Yu, automatc Text Locaton n Images and Vdeo Frames, Pattern Recognton, vol.31, no.12, pp , [9] D. Zhang, and S.F. Chang, General and Doman-specfc Technques for Detectng and Recognzng Supermposed Text n Vdeo, Proceedng of Internatonal Conference on Image Processng, Rochester, New York, USA. [10] B.L.Tseng, C.Y. Ln, D. Zhang and J.R. Smth, Improved Text Overlay Detecton n Vdeos Usng a Fuson-Based Classfer, IBM T.J. Watson Research Center, Yorktown Heghts, New York, [11] C. Dora, Enhancements to Vdeotext Detecton Algorthms, Confdence Measures, and TREC 2002 performance, IBM T.J. Watson Research Center, Yorktown Heghts, New York, September 2002.

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

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

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

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

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

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

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

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

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

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

Hybrid Non-Blind Color Image Watermarking

Hybrid Non-Blind Color Image Watermarking Hybrd Non-Blnd Color Image Watermarkng Ms C.N.Sujatha 1, Dr. P. Satyanarayana 2 1 Assocate Professor, Dept. of ECE, SNIST, Yamnampet, Ghatkesar Hyderabad-501301, Telangana 2 Professor, Dept. of ECE, AITS,

More information

Real-Time View Recognition and Event Detection for Sports Video

Real-Time View Recognition and Event Detection for Sports Video Real-Tme Vew Recognton and Event Detecton for Sports Vdeo Authors: D Zhong and Shh-Fu Chang {dzhong, sfchang@ee.columba.edu} Department of Electrcal Engneerng, Columba Unversty For specal ssue on Multmeda

More information

MULTIRESOLUTION TEXT DETECTION IN VIDEO FRAMES

MULTIRESOLUTION TEXT DETECTION IN VIDEO FRAMES MULTIRESOLUTION TEXT DETECTION IN VIDEO FRAMES Maros Anthmopoulos, Basls Gatos and Ioanns Pratkaks Computatonal Intellgence Laboratory, Insttute of Informatcs and Telecommuncatons Natonal Center for Scentfc

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

More information

DETECTING TEXT IN VIDEO FRAMES

DETECTING TEXT IN VIDEO FRAMES DETECTING TEXT IN VIDEO FRAMES M. Anthmopoulos, B. Gatos, I. Pratkaks, S. J. Perantons Computatonal Intellgence Laboratory, Insttute of Informatcs and Telecommuncatons, Natonal Center for Scentfc Research

More information

Editorial Manager(tm) for International Journal of Pattern Recognition and

Editorial Manager(tm) for International Journal of Pattern Recognition and Artfcal Intellgence Edtoral Manager(tm) for Internatonal Journal of Pattern Recognton and Manuscrpt Draft Manuscrpt Number: Ttle: TEXT LOCALIZATION IN COMPLEX COLOR DOCUMENTS Artcle Type: Research Paper

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

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

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

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

Long-Term Moving Object Segmentation and Tracking Using Spatio-Temporal Consistency

Long-Term Moving Object Segmentation and Tracking Using Spatio-Temporal Consistency Long-Term Movng Obect Segmentaton Trackng Usng Spato-Temporal Consstency D Zhong Shh-Fu Chang {dzhong, sfchang}@ee.columba.edu Department of Electrcal Engneerng, Columba Unversty, NY, USA Abstract The

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

A Bayesian Framework for Fusing Multiple Word Knowledge Models in Videotext Recognition

A Bayesian Framework for Fusing Multiple Word Knowledge Models in Videotext Recognition A Bayesan Framework for Fusng Multple Word Knowledge Models n Vdeotext Recognton DongQng Zhang and Shh-Fu Chang Department of Electrcal Engneerng, Columba Unversty New York, NY 0027, USA. {dqzhang, sfchang}@ee.columba.edu

More information

Novel Pattern-based Fingerprint Recognition Technique Using 2D Wavelet Decomposition

Novel Pattern-based Fingerprint Recognition Technique Using 2D Wavelet Decomposition Mathematcal Methods for Informaton Scence and Economcs Novel Pattern-based Fngerprnt Recognton Technque Usng D Wavelet Decomposton TUDOR BARBU Insttute of Computer Scence of the Romanan Academy T. Codrescu,,

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

Enhanced Watermarking Technique for Color Images using Visual Cryptography

Enhanced Watermarking Technique for Color Images using Visual Cryptography Informaton Assurance and Securty Letters 1 (2010) 024-028 Enhanced Watermarkng Technque for Color Images usng Vsual Cryptography Enas F. Al rawashdeh 1, Rawan I.Zaghloul 2 1 Balqa Appled Unversty, MIS

More information

Palmprint Feature Extraction Using 2-D Gabor Filters

Palmprint Feature Extraction Using 2-D Gabor Filters Palmprnt Feature Extracton Usng 2-D Gabor Flters Wa Kn Kong Davd Zhang and Wenxn L Bometrcs Research Centre Department of Computng The Hong Kong Polytechnc Unversty Kowloon Hong Kong Correspondng author:

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

Audio Content Classification Method Research Based on Two-step Strategy

Audio Content Classification Method Research Based on Two-step Strategy (IJACSA) Internatonal Journal of Advanced Computer Scence and Applcatons, Audo Content Classfcaton Method Research Based on Two-step Strategy Sume Lang Department of Computer Scence and Technology Chongqng

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

Analysis of Continuous Beams in General

Analysis of Continuous Beams in General Analyss of Contnuous Beams n General Contnuous beams consdered here are prsmatc, rgdly connected to each beam segment and supported at varous ponts along the beam. onts are selected at ponts of support,

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

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

Feature Selection for Target Detection in SAR Images

Feature Selection for Target Detection in SAR Images Feature Selecton for Detecton n SAR Images Br Bhanu, Yngqang Ln and Shqn Wang Center for Research n Intellgent Systems Unversty of Calforna, Rversde, CA 95, USA Abstract A genetc algorthm (GA) approach

More information

Face Tracking Using Motion-Guided Dynamic Template Matching

Face Tracking Using Motion-Guided Dynamic Template Matching ACCV2002: The 5th Asan Conference on Computer Vson, 23--25 January 2002, Melbourne, Australa. Face Trackng Usng Moton-Guded Dynamc Template Matchng Lang Wang, Tenu Tan, Wemng Hu atonal Laboratory of Pattern

More information

Efficient Content Representation in MPEG Video Databases

Efficient Content Representation in MPEG Video Databases Effcent Content Representaton n MPEG Vdeo Databases Yanns S. Avrths, Nkolaos D. Doulams, Anastasos D. Doulams and Stefanos D. Kollas Department of Electrcal and Computer Engneerng Natonal Techncal Unversty

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

A MODEL-BASED BOOK BOUNDARY DETECTION TECHNIQUE FOR BOOKSHELF IMAGE ANALYSIS

A MODEL-BASED BOOK BOUNDARY DETECTION TECHNIQUE FOR BOOKSHELF IMAGE ANALYSIS A MODEL-BASED BOOK BOUNDARY DETECTION TECHNIQUE FOR BOOKSHELF IMAGE ANALYSIS Ej TAIRA, Sech UCHIDA, Hroak SAKOE Graduate School of Informaton Scence and Electrcal Engneerng, Kyushu Unversty Faculty of

More information

MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS XUNYU PAN

MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS XUNYU PAN MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS by XUNYU PAN (Under the Drecton of Suchendra M. Bhandarkar) ABSTRACT In modern tmes, more and more

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

A Multi-step Strategy for Shape Similarity Search In Kamon Image Database

A Multi-step Strategy for Shape Similarity Search In Kamon Image Database A Mult-step Strategy for Shape Smlarty Search In Kamon Image Database Paul W.H. Kwan, Kazuo Torach 2, Kesuke Kameyama 2, Junbn Gao 3, Nobuyuk Otsu 4 School of Mathematcs, Statstcs and Computer Scence,

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

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

Intelligent Video Display to Raise Quality of Experience on Mobile Devices

Intelligent Video Display to Raise Quality of Experience on Mobile Devices Intellgent Vdeo Dsplay to Rase Qualty of Experence on Moble Devces Changck Km* a, Jaeseung Ko a, Ilkoo Ahn a, Muhammad Usman a, Jae Hoon Kwon b, Jeongrok Park b, Young Hun Joo b, and Yun Je Oh b a Vsual

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

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

COMPLEX WAVELET TRANSFORM-BASED COLOR INDEXING FOR CONTENT-BASED IMAGE RETRIEVAL

COMPLEX WAVELET TRANSFORM-BASED COLOR INDEXING FOR CONTENT-BASED IMAGE RETRIEVAL COMPLEX WAVELET TRANSFORM-BASED COLOR INDEXING FOR CONTENT-BASED IMAGE RETRIEVAL Nader Safavan and Shohreh Kasae Department of Computer Engneerng Sharf Unversty of Technology Tehran, Iran skasae@sharf.edu

More information

Grading Image Retrieval Based on DCT and DWT Compressed Domains Using Low-Level Features

Grading Image Retrieval Based on DCT and DWT Compressed Domains Using Low-Level Features Journal of Communcatons Vol. 0 No. January 0 Gradng Image Retreval Based on DCT and DWT Compressed Domans Usng Low-Level Features Chengyou Wang Xnyue Zhang Rongyang Shan and Xao Zhou School of echancal

More information

Adaptive Silhouette Extraction and Human Tracking in Dynamic. Environments 1

Adaptive Silhouette Extraction and Human Tracking in Dynamic. Environments 1 Adaptve Slhouette Extracton and Human Trackng n Dynamc Envronments 1 X Chen, Zhha He, Derek Anderson, James Keller, and Marjore Skubc Department of Electrcal and Computer Engneerng Unversty of Mssour,

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

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole Appled Mathematcs, 04, 5, 37-3 Publshed Onlne May 04 n ScRes. http://www.scrp.org/journal/am http://dx.do.org/0.436/am.04.584 The Research of Ellpse Parameter Fttng Algorthm of Ultrasonc Imagng Loggng

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

Efficient Video Coding with R-D Constrained Quadtree Segmentation

Efficient Video Coding with R-D Constrained Quadtree Segmentation Publshed on Pcture Codng Symposum 1999, March 1999 Effcent Vdeo Codng wth R-D Constraned Quadtree Segmentaton Cha-Wen Ln Computer and Communcaton Research Labs Industral Technology Research Insttute Hsnchu,

More information

Key-Selective Patchwork Method for Audio Watermarking

Key-Selective Patchwork Method for Audio Watermarking Internatonal Journal of Dgtal Content Technology and ts Applcatons Volume 4, Number 4, July 2010 Key-Selectve Patchwork Method for Audo Watermarkng 1 Ch-Man Pun, 2 Jng-Jng Jang 1, Frst and Correspondng

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

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images Internatonal Journal of Informaton and Electroncs Engneerng Vol. 5 No. 6 November 015 Usng Fuzzy Logc to Enhance the Large Sze Remote Sensng Images Trung Nguyen Tu Huy Ngo Hoang and Thoa Vu Van Abstract

More information

Identify the Attack in Embedded Image with Steganalysis Detection Method by PSNR and RGB Intensity

Identify the Attack in Embedded Image with Steganalysis Detection Method by PSNR and RGB Intensity Internatonal Journal of Computer Systems (ISSN: 394-1065), Volume 03 Issue 07, July, 016 Avalable at http://www.jcsonlne.com/ Identfy the Attack n Embedded Image wth Steganalyss Detecton Method by PSNR

More information

An Image Compression Algorithm based on Wavelet Transform and LZW

An Image Compression Algorithm based on Wavelet Transform and LZW An Image Compresson Algorthm based on Wavelet Transform and LZW Png Luo a, Janyong Yu b School of Chongqng Unversty of Posts and Telecommuncatons, Chongqng, 400065, Chna Abstract a cylpng@63.com, b y27769864@sna.cn

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

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

Information Hiding Watermarking Detection Technique by PSNR and RGB Intensity

Information Hiding Watermarking Detection Technique by PSNR and RGB Intensity www..org 3 Informaton Hdng Watermarkng Detecton Technque by PSNR and RGB Intensty 1 Neha Chauhan, Akhlesh A. Waoo, 3 P. S. Patheja 1 Research Scholar, BIST, Bhopal, Inda.,3 Assstant Professor, BIST, Bhopal,

More information

An Efficient Background Updating Scheme for Real-time Traffic Monitoring

An Efficient Background Updating Scheme for Real-time Traffic Monitoring 2004 IEEE Intellgent Transportaton Systems Conference Washngton, D.C., USA, October 3-6, 2004 WeA1.3 An Effcent Background Updatng Scheme for Real-tme Traffc Montorng Suchendra M. Bhandarkar and Xngzh

More information

Shape-adaptive DCT and Its Application in Region-based Image Coding

Shape-adaptive DCT and Its Application in Region-based Image Coding Internatonal Journal of Sgnal Processng, Image Processng and Pattern Recognton, pp.99-108 http://dx.do.org/10.14257/sp.2014.7.1.10 Shape-adaptve DCT and Its Applcaton n Regon-based Image Codng Yamn Zheng,

More information

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval

Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval Fuzzy -Means Intalzed by Fxed Threshold lusterng for Improvng Image Retreval NAWARA HANSIRI, SIRIPORN SUPRATID,HOM KIMPAN 3 Faculty of Informaton Technology Rangst Unversty Muang-Ake, Paholyotn Road, Patumtan,

More information

Multiple Image Thumbnailing

Multiple Image Thumbnailing Multple Image Thumbnalng Ganlug Coccaa, Ramondo Schettna a DISCo - Dpartmento d Informatca Sstemstca e Comuncazone, Unverstà degl stud d Mlano-Bcocca, Vale Sarca 336, Mlano, Italy ABSTRACT We have desgned

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

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

Lecture 13: High-dimensional Images

Lecture 13: High-dimensional Images Lec : Hgh-dmensonal Images Grayscale Images Lecture : Hgh-dmensonal Images Math 90 Prof. Todd Wttman The Ctadel A grayscale mage s an nteger-valued D matrx. An 8-bt mage takes on values between 0 and 55.

More information

We Two Seismic Interference Attenuation Methods Based on Automatic Detection of Seismic Interference Moveout

We Two Seismic Interference Attenuation Methods Based on Automatic Detection of Seismic Interference Moveout We 14 15 Two Sesmc Interference Attenuaton Methods Based on Automatc Detecton of Sesmc Interference Moveout S. Jansen* (Unversty of Oslo), T. Elboth (CGG) & C. Sanchs (CGG) SUMMARY The need for effcent

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

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

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 NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION

A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION Mhaela Gordan *, Constantne Kotropoulos **, Apostolos Georgaks **, Ioanns Ptas ** * Bass of Electroncs Department,

More information

A Novel Accurate Algorithm to Ellipse Fitting for Iris Boundary Using Most Iris Edges. Mohammad Reza Mohammadi 1, Abolghasem Raie 2

A Novel Accurate Algorithm to Ellipse Fitting for Iris Boundary Using Most Iris Edges. Mohammad Reza Mohammadi 1, Abolghasem Raie 2 A Novel Accurate Algorthm to Ellpse Fttng for Irs Boundar Usng Most Irs Edges Mohammad Reza Mohammad 1, Abolghasem Rae 2 1. Department of Electrcal Engneerng, Amrabr Unverst of Technolog, Iran. mrmohammad@aut.ac.r

More information

Height restriction barriers detection from traffic scenarios using stereo-vision

Height restriction barriers detection from traffic scenarios using stereo-vision Heght restrcton barrers detecton from traffc scenaros usng stereo-vson Mara-Irna Barbu, Ion Gosan, Tberu Marța Computer Scence Department Techncal Unversty of Cluj-Napoca, Romana Irna.M.Barbu@gmal.com,

More information

A New Knowledge-Based Face Image Indexing System through the Internet

A New Knowledge-Based Face Image Indexing System through the Internet Ne Knoledge-ased Face Image Indexng System through the Internet Shu-Sheng La a Geeng-Neng You b Fu-Song Syu c Hsu-Me Huang d a General Educaton Center, Chna Medcal Unversty, Taan bc Department of Multmeda

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

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

Video Classification and Retrieval with the Informedia Digital Video Library System

Video Classification and Retrieval with the Informedia Digital Video Library System Vdeo Classfcaton and Retreval wth the Informeda Dgtal Vdeo Lbrary System A. Hauptmann, R. Yan, Y. Q, R. Jn, M. Chrstel, M. Derthck, M.-Y. Chen, R. Baron, W.-H. Ln, and T. D. Ng. Carnege Mellon Unversty,

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

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

2-Dimensional Image Representation. Using Beta-Spline

2-Dimensional Image Representation. Using Beta-Spline Appled Mathematcal cences, Vol. 7, 03, no. 9, 4559-4569 HIKARI Ltd, www.m-hkar.com http://dx.do.org/0.988/ams.03.3359 -Dmensonal Image Representaton Usng Beta-plne Norm Abdul Had Faculty of Computer and

More information

Multiple Frame Motion Inference Using Belief Propagation

Multiple Frame Motion Inference Using Belief Propagation Multple Frame Moton Inference Usng Belef Propagaton Jang Gao Janbo Sh The Robotcs Insttute Department of Computer and Informaton Scence Carnege Mellon Unversty Unversty of Pennsylvana Pttsburgh, PA 53

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

Takahiro ISHIKAWA Takahiro Ishikawa Takahiro Ishikawa Takeo KANADE

Takahiro ISHIKAWA Takahiro Ishikawa Takahiro Ishikawa Takeo KANADE Takahro ISHIKAWA Takahro Ishkawa Takahro Ishkawa Takeo KANADE Monocular gaze estmaton s usually performed by locatng the pupls, and the nner and outer eye corners n the mage of the drver s head. Of these

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

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

Description of NTU Approach to NTCIR3 Multilingual Information Retrieval

Description of NTU Approach to NTCIR3 Multilingual Information Retrieval Proceedngs of the Thrd NTCIR Workshop Descrpton of NTU Approach to NTCIR3 Multlngual Informaton Retreval Wen-Cheng Ln and Hsn-Hs Chen Department of Computer Scence and Informaton Engneerng Natonal Tawan

More information

Content Based Retrieval

Content Based Retrieval 2010 2nd Internatonal Conference on Sgnal rocessng Systems (ICSS) A Novel Commercal Break Detecton and Automatc Annotaton of TV rograms for Abstract In ths paper, we present a novel approach for automatc

More information

Resolving Ambiguity in Depth Extraction for Motion Capture using Genetic Algorithm

Resolving Ambiguity in Depth Extraction for Motion Capture using Genetic Algorithm Resolvng Ambguty n Depth Extracton for Moton Capture usng Genetc Algorthm Yn Yee Wa, Ch Kn Chow, Tong Lee Computer Vson and Image Processng Laboratory Dept. of Electronc Engneerng The Chnese Unversty of

More information

Machine Learning. Topic 6: Clustering

Machine Learning. Topic 6: Clustering Machne Learnng Topc 6: lusterng lusterng Groupng data nto (hopefully useful) sets. Thngs on the left Thngs on the rght Applcatons of lusterng Hypothess Generaton lusters mght suggest natural groups. Hypothess

More information

3D vector computer graphics

3D vector computer graphics 3D vector computer graphcs Paolo Varagnolo: freelance engneer Padova Aprl 2016 Prvate Practce ----------------------------------- 1. Introducton Vector 3D model representaton n computer graphcs requres

More information

Hybrid Body Representation for Integrated Pose Recognition, Localization and Segmentation

Hybrid Body Representation for Integrated Pose Recognition, Localization and Segmentation Hybrd Body Representaton for Integrated Pose Recognton, Localzaton and Segmentaton Cheng Chen and Guolang Fan School of Electrcal and Computer Engneerng Oklahoma State Unversty, Stllwater, OK 74078 {c.chen,

More information

Multi-view 3D Position Estimation of Sports Players

Multi-view 3D Position Estimation of Sports Players Mult-vew 3D Poston Estmaton of Sports Players Robbe Vos and Wlle Brnk Appled Mathematcs Department of Mathematcal Scences Unversty of Stellenbosch, South Afrca Emal: vosrobbe@gmal.com Abstract The problem

More information

MPEG-7 Pictorially Enriched Ontologies for Video Annotation

MPEG-7 Pictorially Enriched Ontologies for Video Annotation MPEG-7 Pctorally Enrched Ontologes for Vdeo Annotaton C. Grana, R.Vezzan, D. Bulgarell, R. Cucchara Dpartmento d Ingegnera dell Informazone Unverstà degl Stud d Modena e Reggo Emla Abstract. A system for

More information

A high precision collaborative vision measurement of gear chamfering profile

A high precision collaborative vision measurement of gear chamfering profile Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) A hgh precson collaboratve vson measurement of gear chamferng profle Conglng Zhou, a, Zengpu Xu, b, Chunmng

More information

EFFICIENT H.264 VIDEO CODING WITH A WORKING MEMORY OF OBJECTS

EFFICIENT H.264 VIDEO CODING WITH A WORKING MEMORY OF OBJECTS EFFICIENT H.264 VIDEO CODING WITH A WORKING MEMORY OF OBJECTS A Thess presented to the Faculty of the Graduate School at the Unversty of Mssour-Columba In Partal Fulfllment of the Requrements for the Degree

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

USING GRAPHING SKILLS

USING GRAPHING SKILLS Name: BOLOGY: Date: _ Class: USNG GRAPHNG SKLLS NTRODUCTON: Recorded data can be plotted on a graph. A graph s a pctoral representaton of nformaton recorded n a data table. t s used to show a relatonshp

More information