Please use this identifier to cite or link to this item:
Title: Extraction of Text from Images and Videos
Keywords: text localization, text detection, text enhancement, text recognition, scene text, video text
Issue Date: 23-Jan-2014
Source: PHAN QUY TRUNG (2014-01-23). Extraction of Text from Images and Videos. ScholarBank@NUS Repository.
Abstract: With the rapid growth of the Internet, there is an increasing demand for text extraction in images and videos, which facilitates retrieval in image/video databases. However, text extraction is a challenging problem due to the unconstrained colors, backgrounds and alignments of the characters. This thesis describes novel methods for text extraction in natural scene images and in videos. We first propose two works for text localization. The first one uses gap features to localize scene text while the second one employs skeletonization to pick up video text. We then develop two works for text enhancement. The first one performs binarization at the character level for single-frame enhancement. The second one integrates the multiple instances of a video text for multiple-frame enhancement. Finally, we propose a method for recognizing scene text with perspective distortion. Experimental results show that the proposed methods compare favourably to the state-of-the-art on several public datasets.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
PhanQT.pdf5.44 MBAdobe PDF



Page view(s)

checked on Dec 11, 2017


checked on Dec 11, 2017

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.