A new video text detection method

Authors:
Jie Yuan;Baogang Wei;Weiming Lu;Lidong Wang
Affiliations:
Zhejiang University , Hangzhou, China;Zhejiang University , Hangzhou, China;Zhejiang University , Hangzhou, China;Hangzhou Normal University , Hangzhou, China
Venue:
Proceedings of the 11th annual international ACM/IEEE joint conference on Digital libraries
Year:
2011

Citing 8
Cited 0

Automatic text segmentation and text recognition for video indexing

Multimedia Systems
Text Detection in Images Based on Unsupervised Classification of High-Frequency Wavelet Coefficients

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 1 - Volume 01
Text Extraction from Web Images Based on A Split-and-Merge Segmentation Method Using Colour Perception

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 2 - Volume 02
Text Detection in Images Based on Unsupervised Classification of Edge-based Features

ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Color-based clustering for text detection and extraction in image

Proceedings of the 15th international conference on Multimedia
A new approach for overlay text detection and extraction from complex video scene

IEEE Transactions on Image Processing
Fast and robust text detection in images and video frames

Image and Vision Computing
A comprehensive method for multilingual video text detection, localization, and extraction

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Nowadays, digital libraries contain more and more videos in them, and how to organize and retrieve those videos effectively has become very urgent. Text in videos is a very meaningful clue for video semantic understanding, and it can be used for video organization and retrieval. However, existing text recognizing methods can not deal with multilingual texts or texts embedded in a complex background very well. In this paper, we propose a novel video text detection method. Edge detection and candidate region extraction are firstly used to obtain all rough candidate text regions, and then region refinement is used to obtain the accurate location of each region. Based on our observation that a real text region has a uniform distribution with its non-zero pixels in its binary image, an entropy filter is used to remove non-text regions. Experiments on various videos show that our method is effective and robust against different languages, background complexities and font styles.