Text-to-speech reading assistant device with scene text locator for the blind

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

One of the most anticipated devices for the blind is probably a wearable camera capable of translating the captured scene text into an alternative representation such as speech. However, finding signboards and capturing the text images in a form suitable for character recognition is difficult for the blind. We propose a framework of reading assistant device with a scene text locator that shows the text location by sound signals. By using a pair of base/offset notes, the accuracy of text localization has become much higher compared with our previous system.

Original languageEnglish
Title of host publicationAssistive Technology
Subtitle of host publicationFrom Research to Practice
Pages702-707
Number of pages6
DOIs
Publication statusPublished - 2013 Dec 9

Publication series

NameAssistive Technology Research Series
Volume33
ISSN (Print)1383-813X
ISSN (Electronic)1879-8071

Keywords

  • OCR (Optical Character Recognition)
  • Reading assistant
  • Scene text detection
  • Text locator
  • Text-to-speech

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Engineering (miscellaneous)

Fingerprint Dive into the research topics of 'Text-to-speech reading assistant device with scene text locator for the blind'. Together they form a unique fingerprint.

Cite this