Please use this identifier to cite or link to this item: https://doi.org/10.1145/1710035.1710095
Title: Robust pre-processing techniques for OCR applications on mobile devices
Authors: Chang, L.Z. 
ZhiYing, S.Z. 
Keywords: Mobile device
OCR
Optical character recognition
Otsu threshold
Pre-processing
Issue Date: 2009
Source: Chang, L.Z.,ZhiYing, S.Z. (2009). Robust pre-processing techniques for OCR applications on mobile devices. Proceedings of the 6th International Conference on Mobile Technology, Application and Systems, Mobility '09. ScholarBank@NUS Repository. https://doi.org/10.1145/1710035.1710095
Abstract: Text detection and recognition in natural images using a single mobile device is becoming relevant due to the increasing interest in Optical Character Recognition (OCR) applications. OCR application on mobile devices is no longer a dream due to the advancement in mobile technology. There are many ongoing researches in this field. Most of the researches focus on the OCR engine of the application and there is not much focus on the processing stage of the OCR application. Pre-processing plays a major role in optimizing the image for character recognition. Thus, in this paper, two pre-processing techniques are proposed for the mobile OCR application. The first technique helps to locate a sign in a natural image efficiently, while the second technique implements Otsu's threshold algorithm to convert images into binary image. These techniques are implemented in an OCR mobile application which is developed using desktop open sources library. After implementation, the OCR application is tested with 50 sign images to verify the accuracy. Experimental results have demonstrated that these techniques can significantly improve the OCR accuracy and decrease the overall computation time. Copyright © 2009 ACM.
Source Title: Proceedings of the 6th International Conference on Mobile Technology, Application and Systems, Mobility '09
URI: http://scholarbank.nus.edu.sg/handle/10635/71687
ISBN: 9781605585369
DOI: 10.1145/1710035.1710095
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

2
checked on Feb 20, 2018

Page view(s)

87
checked on Feb 16, 2018

Google ScholarTM

Check

Altmetric


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