Research on the Recognition Technology of Letter String on Label

Yu Chan Xie; Wei Jin

doi:10.4028/www.scientific.net/AMR.476-478.876

Paper Titles

Role of Interface of Epoxy/Clay Nanocomposites and its Effect on Structure-Property Relationship
p.859

Synthesis and Characterization of Bacterial Cellulose/Calcium Silicate Composites
p.863

A Multi-Degree-of-Freedom System Modal Parameters Wavelet Transform Identification Method
p.867

LBM Application in Two-Phase Fluid Flow with Large Density Ratio
p.871

Research on the Recognition Technology of Letter String on Label
p.876

Fabrication and Characterization of ZnO-Based Nano Hetero Structure by Using a 2-Step Hydrothermal Growth Method
p.881

The Research Progress of Soccer Learning Mechanisms for Robot
p.886

Magnet Skew Technology Investigation for the Reduction of PMLSM Thrust Ripple
p.890

Pass Design Method for Rolling Titanium Alloy Rod
p.894

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 476-478Research on the Recognition Technology of Letter...

Research on the Recognition Technology of Letter String on Label

Abstract:

A letters recognition method based on a small area of the feature matching is proposed, in which the unqualified letter in the string were recognized based on its geometric feature. In this method, firstly confirm label’s position by the extracted straight line characters. Then, rotate and zoom image according to the prior knowledge of label. Acquire the position of the string on the label and cut them out. And then single letters are segmented out using the projection method in the area of the string. Finally, compare geometrical characteristics of the extracted letters with its template to recognize and eliminate unqualified labels. The matching process proceeded only in the small region with letters and there is no need to search the irrelevant information. For this algorithm, so much time has been saved compared with traditional letter recognition method and projection method which is used for single character segmentation is very quick also. Theory analysis and experiment showed that the method can segment unqualified characters robustly and quickly.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 476-478)

Pages:

876-880

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.476-478.876

Citation:

Cite this paper

Online since:

February 2012

Authors:

Yu Chan Xie, Wei Jin

Keywords:

Letter Recognition, Projection Method, Straight Line

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Li Guoping, Lu Changhou, Li Jianmei: Proc of the 6th World Congress on Control and Automation.Dalian,China ,Vol.(2)( 2006) P.175.

DOI: 10.1109/wcica.2006.1713992

Google Scholar

[2] Zhang Ping,Zhu Zhenghong:Jounal of Hefei University of Technology.Vol.30(2007),P.1293.

Google Scholar

[3] Liao Qiang, Zhou Yi,Mi Lin,Xu Zhongjun:Jounal of Chongqing University,Vol.(25)(2002),P.1.

Google Scholar

[4] Song Huaibo,Lu Changhou,Wang Fuchuan . PR&AI. Vol. 21 (2008), P.824.

Google Scholar

[5] Li Zuo,Wang Shuhua,Cai Shijie:A character recognition approach based on feature line necessary sufficient condition detection.Li. Journal of Software. Vol.13 (2002), P.85.

Google Scholar

[6] Hough P．V．C． A method and means for recognizing complex patterns, U.S. Patent 3,069,654.1962．(2001)

Google Scholar