Text Detection in Born-Digital Images Using IT-LBP
Fine text detection plays a crucial role in a text detection algorithm as it is capable of removing the false alarms while keeping the detected text lines in coarse text detection. Good performance of a machine learning-based fine text detection heavily depends on the powerful feature to depict the...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SAGE Publishing
2014-03-01
|
Series: | Journal of Algorithms & Computational Technology |
Online Access: | https://doi.org/10.1260/1748-3018.8.1.127 |
id |
doaj-c9cc45bbb9b14fb1a6693b52abc6164b |
---|---|
record_format |
Article |
spelling |
doaj-c9cc45bbb9b14fb1a6693b52abc6164b2020-11-25T03:16:58ZengSAGE PublishingJournal of Algorithms & Computational Technology1748-30181748-30262014-03-01810.1260/1748-3018.8.1.127Text Detection in Born-Digital Images Using IT-LBPChao Zeng0Wenjing Jia1Xiangjian He2Liming Zhang3 Research Centre for Innovation in IT Services and Applications (iNEXT) University of Technology, Sydney Research Centre for Innovation in IT Services and Applications (iNEXT) University of Technology, Sydney Research Centre for Innovation in IT Services and Applications (iNEXT) University of Technology, Sydney Faculty of Science and Technology, University of MacauFine text detection plays a crucial role in a text detection algorithm as it is capable of removing the false alarms while keeping the detected text lines in coarse text detection. Good performance of a machine learning-based fine text detection heavily depends on the powerful feature to depict the characteristics of text. In this paper, a novel texture-based descriptor, named IT-LBP, is proposed by considering horizontal, vertical, diagonal and anti-diagonal directions of character strokes to better describe the texture of text. The new feature demonstrates its superiority by comparing with other texture-based features. The new feature is used to train an SVM classifier to further filter out non-text candidates. The ICDAR 2011 born-digital image dataset is used to evaluate and demonstrate the performance of the proposed method. Following the same performance evaluation criteria, the proposed method outperforms the winner algorithm of the ICDAR 2011 Robust Reading Competition Challenge 1.https://doi.org/10.1260/1748-3018.8.1.127 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Chao Zeng Wenjing Jia Xiangjian He Liming Zhang |
spellingShingle |
Chao Zeng Wenjing Jia Xiangjian He Liming Zhang Text Detection in Born-Digital Images Using IT-LBP Journal of Algorithms & Computational Technology |
author_facet |
Chao Zeng Wenjing Jia Xiangjian He Liming Zhang |
author_sort |
Chao Zeng |
title |
Text Detection in Born-Digital Images Using IT-LBP |
title_short |
Text Detection in Born-Digital Images Using IT-LBP |
title_full |
Text Detection in Born-Digital Images Using IT-LBP |
title_fullStr |
Text Detection in Born-Digital Images Using IT-LBP |
title_full_unstemmed |
Text Detection in Born-Digital Images Using IT-LBP |
title_sort |
text detection in born-digital images using it-lbp |
publisher |
SAGE Publishing |
series |
Journal of Algorithms & Computational Technology |
issn |
1748-3018 1748-3026 |
publishDate |
2014-03-01 |
description |
Fine text detection plays a crucial role in a text detection algorithm as it is capable of removing the false alarms while keeping the detected text lines in coarse text detection. Good performance of a machine learning-based fine text detection heavily depends on the powerful feature to depict the characteristics of text. In this paper, a novel texture-based descriptor, named IT-LBP, is proposed by considering horizontal, vertical, diagonal and anti-diagonal directions of character strokes to better describe the texture of text. The new feature demonstrates its superiority by comparing with other texture-based features. The new feature is used to train an SVM classifier to further filter out non-text candidates. The ICDAR 2011 born-digital image dataset is used to evaluate and demonstrate the performance of the proposed method. Following the same performance evaluation criteria, the proposed method outperforms the winner algorithm of the ICDAR 2011 Robust Reading Competition Challenge 1. |
url |
https://doi.org/10.1260/1748-3018.8.1.127 |
work_keys_str_mv |
AT chaozeng textdetectioninborndigitalimagesusingitlbp AT wenjingjia textdetectioninborndigitalimagesusingitlbp AT xiangjianhe textdetectioninborndigitalimagesusingitlbp AT limingzhang textdetectioninborndigitalimagesusingitlbp |
_version_ |
1724633874892324864 |