Skip to main content
padlock icon - secure page this page is secure

An Audio Based Real Time Text Detection and Recognition Approach for Visually Impaired People

Buy Article:

$106.81 + tax (Refund Policy)

Text detection and character recognition in natural images are widespread, yet unsolved issues are available in computer vision. In most of the existing techniques, the texts were extracted from restored word image, which depends on the height difference between texts for text recognition. However, the text recognition accuracy using such techniques was low. Moreover, some existing techniques may not produce an audio-based output, so that it is a most challenging task to the visually impaired people. Therefore, this paper proposed an audio based real time text detection and recognition approach with the aim to provide guidance to the visually impaired people. This paper defines the system design and utilized five steps to recognize the characters. In the case of blind people, finding the text region is an essential problem that must be addressed. Because, it cannot be assumed that the learned image includes only characters. At first, the proposed system tries to recognize the characters and provide the audio solutions to the visually impaired people. Initially, image preprocessing is done based on the Gaussian filtering algorithm and histogram equalization. This paper introduces the Gradient Based Superpixel Segmentation (GBSS) algorithm to segment the text portions alone. The texture and shape features of the image are collected based on Convoluted Local Tetra Patterns (CLTrP), Histogram of Oriented Gradient (HOG) with the set of shape features. Then, the optimal features are extracted using the Fuzzy based Particle Swarm Optimization (FPSO) algorithm. Moreover, this paper introduces the Distance based Relevance Vector Machine (DRVM) algorithm to accurately recognize the characters. At last, the system provides an audio output to assist the blind persons for their day-to-day activities. The performance of the proposed method is compared with the existing SVM and RVM approaches, the results show that the proposed method yields higher precision, recall and accuracy than the existing algorithms.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
No Metrics

Keywords: Convoluted Local Tetra Patterns (CLTrP); Distance Based Relevance Vector Machine (DRVM); Fuzzy Based Particle Swarm Optimization (FPSO); Support Vector Machine (SVM); Text Detection and Character Recognition

Document Type: Research Article

Affiliations: Department of Electronics and Communication Engineering, Sudharsan Engineering College, Pudukkottai 622501, Tamilnadu, India

Publication date: August 1, 2016

More about this publication?
  • Journal of Computational and Theoretical Nanoscience is an international peer-reviewed journal with a wide-ranging coverage, consolidates research activities in all aspects of computational and theoretical nanoscience into a single reference source. This journal offers scientists and engineers peer-reviewed research papers in all aspects of computational and theoretical nanoscience and nanotechnology in chemistry, physics, materials science, engineering and biology to publish original full papers and timely state-of-the-art reviews and short communications encompassing the fundamental and applied research.
  • Editorial Board
  • Information for Authors
  • Submit a Paper
  • Subscribe to this Title
  • Terms & Conditions
  • Ingenta Connect is not responsible for the content or availability of external websites
  • Access Key
  • Free content
  • Partial Free content
  • New content
  • Open access content
  • Partial Open access content
  • Subscribed content
  • Partial Subscribed content
  • Free trial content
Cookie Policy
Cookie Policy
Ingenta Connect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more