Share Email Print

Proceedings Paper

Extraction of words from the national ID cards for automated recognition
Author(s): Md. Rezwan Akhter; Md. Hasanuzzaman Bhuiyan; Mohammad Shorif Uddin
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The government of Bangladesh introduced national ID cards in 2008 for all peoples of age 18 years and above. This card is now a de-facto identity document and finds diverse applications in vote casting, bank account opening, telephone subscribing as well as in many real life transactions and security checking. To get real fruits of this versatile ID card, automated retrieving and recognition of an independent person from this extra large national database is an ultimate necessity. This work is the first step to fill this gap in making the recognition in automated fashion. Here we have investigated an image analysis technique to extract the words that will be used in subsequent recognition steps. At first scanned ID card image is used as an input into the computer system and then the target text region is separated from the picture region. The text region is used for separation of lines and words on the basis of the vertical and horizontal projections of image intensity, respectively. Experimentation using real national ID cards confirms the effectiveness of our technique.

Paper Details

Date Published: 1 October 2011
PDF: 6 pages
Proc. SPIE 8285, International Conference on Graphic and Image Processing (ICGIP 2011), 828521 (1 October 2011); doi: 10.1117/12.913478
Show Author Affiliations
Md. Rezwan Akhter, Univ. of Liberal Arts Bangladesh (Bangladesh)
Md. Hasanuzzaman Bhuiyan, Ahsanullah Univ. of Science and Technology (Bangladesh)
Mohammad Shorif Uddin, Jahangirnagar Univ. (Bangladesh)

Published in SPIE Proceedings Vol. 8285:
International Conference on Graphic and Image Processing (ICGIP 2011)
Yi Xie; Yanjun Zheng, Editor(s)

© SPIE. Terms of Use
Back to Top