Scene Text Detection & Language Translation Using SWT and EBMT

DOI

10.2991/iccasp-16.2017.79How to use a DOI?

Keywords

Word segmentation, text extraction, text detection, character recognition, localization.

Abstract

In recent years, the popularity of portable devices for capturing images, video, text extraction, etc. become key problems to detect and recognize the text in images. Extraction of text information from images or scene involves binary, text detection, text localization, word segmentation, and enhancement and character recognition. But there are variations involved in text such as font style, font size, orientation, alignment, reflections and illumination effect, hybrid background and low contrast image make text extraction process too difficult and more challenging. A large number of approaches and methods have been proposed to address this problem but still none of them are perfect. This paper presents an effective approach where we come up with hybrid approach that combines SWT with OCR for text detection and recognition in an image and EBMT to translate detected text into Hindi language. General challenges for performing scene text detection are also discussed.

Copyright

© 2017, 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  - S. Khandait
AU  - P. Khandait
AU  - P. Jambhulkar
PY  - 2016/12
DA  - 2016/12
TI  - Scene Text Detection & Language Translation Using SWT and EBMT
BT  - Proceedings of the International Conference on Communication and Signal Processing 2016 (ICCASP 2016)
PB  - Atlantis Press
SP  - 540
EP  - 548
SN  - 1951-6851
UR  - https://doi.org/10.2991/iccasp-16.2017.79
DO  - 10.2991/iccasp-16.2017.79
ID  - Khandait2016/12
ER  -