By Topic

Bar code reading from images captured by camera phones

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Kongqiao Wang ; Nokia Res. Center ; Yanming Zou ; Hao Wang

Bar codes are being widely used in many fields for applications of great commercial value. By encoding a series of characters or symbols, bar codes are able to both carry explicit information and a database key. Nowadays, The availability of imaging phones provides people a mobile platform for decoding bar code rather than the use of the conventional scanner which is lack of mobility. However, the short-distance capture of bar codes using an imaging phone inevitably makes bar code images blurred, meanwhile, these images are contaminated heavily with noises. Hence, it is a challenge for automatic bar code reading by imaging phones in such applications. In this paper, research effort on the algorithms of bar code reading by real NOKIA imaging phone products is proposed and EAN-13, a widely used 1-D bar code standard, is taken as an example to show the efficiency of the method. The method, of course, can be extended to other bar code standards without much effort. A wavelet-based bar code area location and knowledge-based bar code character segmentation scheme is applied to extract bar code characters under poor image quality of real conditions. Then the waveforms of the 12 marked divisions are input to the decoding engine, which is called statistical recognition block, and final decoding decision is made. Training of the statistical classifiers is based on the modified GLVQ (generalised learning vector quantization) method and the initial feature extraction is based on LDA (linear discriminant analysis). Training samples are from the database contains over 1,100 bar code images taken by an imaging phone and the sample set is extended by manually shifting (distortion) of the original samples to cover more possibilities of occurrence. Nearly 300 EAN-13 bar code images taken by imaging phone (NOKIA 3650) without micro-lens are tested to prove the effectiveness of the proposed method. The entire symbol recognition rate is 85.62%, which is desirable for the first kick-off - - of the attempt to implement bar code reading applications in the camera phone products. Bar code images taken with micro-lens or optical zoom functionality are also tested and the entire symbol recognition rate is nearly hundred percent

Published in:

Mobile Technology, Applications and Systems, 2005 2nd International Conference on

Date of Conference:

15-17 Nov. 2005