現今電視擷取文字的方法是透過光學文字辨識來擷取文字資訊，而且都是把焦點放在電視節目視頻上，利用光學文字辨識來擷取電視節目視頻上顯示的文字。但是現在的數位電視節目不再只有視頻的部份，還增加了應用程式的顯示，也有獨立字幕的出現，因此在本研究中提出了一個文字擷取系統，除了電視節目視頻外，還考慮了應用程式所顯示的文字以及獨立字幕。由於過去尚未有人提及應用程式所顯示文字的擷取方式，因此在這裡提出一個簡易的架構來取得應用程式所顯示的文字資訊。
在這篇論文中，有別於一般純粹利用光學文字辨識的文字擷取方法，是利用MHP(Multimedia Home Platform)規範所定義的架構及環境來設計一個文字擷取系統，此擷取系統可以取得不同的媒體所顯示的文字，使用不同的方式擷取各種文字，此文字擷取系統能夠擷取在視頻播放以及執行應用程式時顯示在螢幕上的資訊。本研究實驗的平台是工作頻率1.73GHz個人電腦和工作頻率933MHz嵌入平台，實驗結果顯示，此文字擷取系統能夠在MHP適時擷取資訊，系統擷取所花費的時間不會超過0.5秒，若為應用程用程式顯示的文字，擷取正確率會提高至100%，其它部份的文字擷取正確率隨著光學文字辨識系統而改變，這裡利用Asprise光學文字辨識系統，正確率為91.3%，並且在最後實作數個MHP應用程式結合擷取文字達到更多的應用服務，證明此系統能夠共享文字以及開放程式介面。

Nowadays DTV text retrieval is using OCR to retrieve texts and it focused on DTV video only; it is mainly using OCR to retrieve texts of video on screen. But a DTV program has not only video part, it also displays texts generated by application programs and independent subtitles. In the related researches there is no text retrieval system for retrieving texts generated by application programs, therefore this research proposes a solution to this issue.
This research not only uses OCR to retrieve texts on DTV, it also bases on MHP specification defined architecture and environment to design a text retrieval system. The system can retrieve text from different sources, like videos, subtitles and applications. It can use different approaches to retrieve all texts and it can retrieve texts from video and applications shown on DTV screen. The experimentation platform is an 1.73GHz PC and 933MHz embedded platform.The results of experimentation show that the MHP text retrieval system enables retrieving information timely and get the recognition rate to 100% of texts generated by application programs. The recognition rate using Asprise OCR is 91.3%. Retrieval time is less than 0.5 second.

[1]
曾世憲、黃麗屏、許騰尹、張眾欣、粘永峰、謝志成, “家庭多媒體中心(HMC)—數位生活時代個人電腦發展趨勢與市場動態之調查研究”,經濟部培訓科技背景跨領域高級人才計畫 九十三年海外培訓成果發表會, 民國93年。
[2]
N. Akiyo and T. Katsumi, "Complementing your TV-viewing by web content automatically-transformed into TV-program-type content," in Proceedings of the 13th annual ACM international conference on Multimedia, Hilton, Singapore, 2005.
[3]
G. H. Alexander, J. Rong, and N. Tobun Dorbin, "Multi-modal information retrieval from broadcast video using OCR and speech recognition," in Proceedings of the 2nd ACM/IEEE-CS joint conference on Digital libraries, Portland, Oregon, USA, 2002.
[4]
Asprise OCR Java SDK. Available at: http://asprise.com/product/ocr/selector.php.
[5]
M. Carlo, S. Fabrizio, and S. Umberto, "A model of multimedia information retrieval." vol. 48: ACM, 2001, pp. 909-970.
[6]
P. Chengyuan and V. Petri, "A digital television navigator," in Proceedings of the eighth ACM international conference on Multimedia, Marina del Rey, California, United States, 2000.
[7]
Digital Video Broadcasting. Available at: http://www.dvb.org/.
[8]
Digital Video Broadcasting (DVB);Subtitling systems. European Broadcast Union. Available at: http://broadcasting.ru/pdf-standard-specifications/subtitling/dvb-sub/en300743.v1.2.1.pdf.
[9]
L. Feng and S. Jun, "An Interactive Media Platform Scheme for DTV Receiver Compliant with MHP," Consumer Electronics, IEEE Transactions on, vol. 53, pp. 1370-1377, 2007.
[10]
H. P. Fotschl and R. Plosch, "Interactive applications for the multimedia home platform," in Multimedia Software Engineering, 2002. Proceedings. Fourth International Symposium on, 2002, pp. 264-271.
[11]
M. Hisashi, M. Mitsuru, S. Zoran, M. Qiang, A. Tadashi, and T. Katsumi, "Proposal of integrated search engine of web and TV contents," in Proceedings of the 15th international conference on World Wide Web, Edinburgh, Scotland, 2006.
[12]
Home Audio / Video Interoperability. Available at: http://www.havi.org/.
[13]
Z. Hongguang, Z. Shibao, and Y. Jianghai, "A personalized TV guide system compliant with MHP," Consumer Electronics, IEEE Transactions on, vol. 51, pp. 731-737, 2005.
[14]
Java Media Framework. Sun Microsystems. Available at:
http://java.sun.com/javase/technologies/desktop/media/jmf/reference/docs/index.html
[15]
Java Virtual Machine SPEC. Available at: http://java.sun.com/.
[16]
J. John, "DVB-MHP/Java TV data transport mechanisms," in Proceedings of the Fortieth International Conference on Tools Pacific: Objects for internet, mobile and embedded applications, Sydney, Australia, 2002.
[17]
R. Lienhart and A. Wernicke, "Localizing and segmenting text in images and videos," Circuits and Systems for Video Technology, IEEE Transactions on, vol. 12, pp. 256-268, 2002.
[18]
H. Liu, "Content-Based TV Sports Video Retrieval Based on Audio-Visual Features and Text Information," in Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence, 2004.
[19]
J. C. Lopez-Ardao, C. Lopez, A. Gil, R. Diaz, A. Fernandez, M. Fernandez, A. Suarez, and J. Munoz, "Experiences from implementing a MHP receiver," in Video/Image Processing and Multimedia Communications 4th EURASIP-IEEE Region 8 International Symposium on VIPromCom, 2002, pp. 441-446.
[20]
A. Messer, A. Kunjithapatham, N. Phuong, P. Rathod, M. Sheshagiri, D. Cheng, and S. Gibbs, "Internet Search on TV," in Consumer Communications and Networking Conference, 2008. CCNC 2008. 5th IEEE, 2008, pp. 1240-1241.
[21]
MHP Specification. European Telecommunications Standards Institute. Available at: http://www.mhp.org/fullspeclist.htm.
[22]
C. Ming-yu and H. Alexander, "Multi-modal classification in digital news libraries," in Proceedings of the 4th ACM/IEEE-CS joint conference on Digital libraries, Tuscon, AZ, USA, 2004.
[23]
Multimedia Home Platform. European Telecommunications Standards Institute. Available at: http://www.mhp.org/.
[24]
OpenMHP. Available at: http://www.openmhp.org/.
[25]
Optical character recognition. Available at: http://en.wikipedia.org/wiki/Optical_character_recognition.
[26]
J. Piesing, "The DVB Multimedia Home Platform (MHP) and Related Specifications," Proceedings of the IEEE, vol. 94, pp. 237-247, 2006.
[27]
T. Sabine, S. Ansgar, A. Melanie, and B. Susanne, "Media gallery TV: view and shop your photos on interactive digital television," in Proceedings of the 13th annual ACM international conference on Multimedia, Hilton, Singapore, 2005.
[28]
Specification for conveying ITU-R System B Teletext in DVB bitstreams. European Broadcasting Union. Available at: http://broadcasting.ru/pdf-standard-specifications/multiplexing/dvb-txt/en300472.v1.3.1.pdf.
[29]
F. Stefano, R. Marco, and A. Johannes, "Living the TV revolution: unite MHP to the web or face IDTV irrelevance!," in Proceedings of the 15th international conference on World Wide Web, Edinburgh, Scotland, 2006.
[30]
XleTView. Available at: http://xletview.sourceforge.net/.
[31]
C. Yang, P. Wang, C. Lin, and T. Hou, "Personalized iDTV Program in Multimedia Home Platform," in Advanced Information Networking and Applications - Workshops, 2008. AINAW 2008. 22nd International Conference on, 2008, pp. 473-476.
[32]
Z. Yantao, D. Lingyu, T. Qi, and J. S. Jin, "TV Commercial Classification by using Multi-Modal Textual Information," in Multimedia and Expo, 2006 IEEE International Conference on, 2006, pp. 497-500.
[33]
K. S. Yildirim, A. Ugur, and A. C. Kinaci, "Design and Implementation of a Software Presenting Information in DVB Subtitles in Various Forms," Consumer Electronics, IEEE Transactions on, vol. 53, pp. 1656-1660, 2007.
[34]
Z. Zhen-bo, L. Jun, P. Liang, and N. Hong, "Implementation of an advanced TV Navigation System," in Signal Processing, 2006 8th International Conference on, 2006.