Title :
A content based Internet search engine for analysis and archival of MPEG-1 compressed newsfeeds
Author :
Kao, Odej ; Joubert, Gerhard R.
Author_Institution :
Dept. of Comput. Sci., Tech. Univ. Clausthal, Germany
Abstract :
In this paper a specialised search engine for TV news editors as part of a system for analysis and presentation of MPEG compressed newsfeeds is discussed. The editor is supplied with additional material relating to particular news items, which he can, for example, use to prepare the accompanying text of newscasts. The key words for the search engine are extracted automatically from the compressed MPEG newsfeeds. This is a complex process including a video analysis phase, image preprocessing of the frames with the textual information and finally character recognition. The extracted keywords can on the one hand be used by news editors to search for particular news items in a newsfeed and on the other to search for additional information on the Internet. All information supplied to news editors is presented in a well-structured and clear fashion by the ClaViPS system based on HTML pages
Keywords :
Internet; code standards; content-based retrieval; data compression; hypermedia markup languages; optical character recognition; search engines; telecommunication standards; video coding; ClaViPS system; HTML; Internet search engine; MPEG-1 compressed newsfeeds; TV news editors; character recognition; content based retrieval; image preprocessing; keywords; newscasts; video analysis; Character recognition; Data mining; Image analysis; Image coding; Information analysis; Internet; Search engines; TV; Transform coding; Video compression;
Conference_Titel :
Multimedia and Expo, 2000. ICME 2000. 2000 IEEE International Conference on
Conference_Location :
New York, NY
Print_ISBN :
0-7803-6536-4
DOI :
10.1109/ICME.2000.871565
