A Method of Caption Detection in News Video

DOI

10.2991/icmt-13.2013.61How to use a DOI?

Keywords

news video· caption detection· edge detection·Otsu

Abstract

News video is one of the most important media for people to get information. However, it is an urgent problem to find the useful information from a huge amount of news video efficiently and correctly. The caption in news video highly summarizes the related news story and can be used for effective retrieval. In this paper, based on the feature of news video, a method of caption detection in news video is proposed. Firstly, the key frames with captions are detected by using color and edge information. Then, the caption text is extracted by Otsu algorithm. The experiment results show that all the caption frames can be detected and combined with OCR software, the proposed method can give an average recognition rate of 87.3%.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Huang He
AU  - Shi Ping
PY  - 2013/11
DA  - 2013/11
TI  - A Method of Caption Detection in News Video
BT  - Proceedings of 3rd International Conference on Multimedia Technology(ICMT-13)
PB  - Atlantis Press
SP  - 495
EP  - 502
SN  - 1951-6851
UR  - https://doi.org/10.2991/icmt-13.2013.61
DO  - 10.2991/icmt-13.2013.61
ID  - He2013/11
ER  -