OSA's Digital Library

Applied Optics

Applied Optics

APPLICATIONS-CENTERED RESEARCH IN OPTICS

  • Editor: Joseph N. Mait
  • Vol. 48, Iss. 35 — Dec. 10, 2009
  • pp: 6606–6620

Robust wafer identification recognition based on asterisk-shape filter and high–low score comparison method

Wei-Chih Hsu, Tsan-Ying Yu, and Kuan-Liang Chen  »View Author Affiliations


Applied Optics, Vol. 48, Issue 35, pp. 6606-6620 (2009)
http://dx.doi.org/10.1364/AO.48.006606


View Full Text Article

Enhanced HTML    Acrobat PDF (2072 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

Wafer identifications (wafer ID) can be used to identify wafers from each other so that wafer processing can be traced easily. Wafer ID recognition is one of the problems of optical character recognition. The process to recognize wafer IDs is similar to that used in recognizing car license-plate characters. However, due to some unique characteristics, such as the irregular space between two characters and the unsuccessive strokes of wafer ID, it will not get a good result to recognize wafer ID by directly utilizing the approaches used in car license-plate character recognition. Wafer ID scratches are engraved by a laser scribe almost along the following four fixed directions: horizontal, vertical, plus 45 ° , and minus 45 ° orientations. The closer to the center line of a wafer ID scratch, the higher the gray level will be. These and other characteristics increase the difficulty to recognize the wafer ID. In this paper a wafer ID recognition scheme based on an asterisk-shape filter and a high–low score comparison method is proposed to cope with the serious influence of uneven luminance and make recognition more efficiently. Our proposed approach consists of some processing stages. Especially in the final recognition stage, a template-matching method combined with stroke analysis is used as a recognizing scheme. This is because wafer IDs are composed of Semiconductor Equipment and Materials International (SEMI) standard Arabic numbers and English alphabets, and thus the template ID images are easy to obtain. Furthermore, compared with the approach that requires prior training, such as a support vector machine, which often needs a large amount of training image samples, no prior training is required for our approach. The testing results show that our proposed scheme can efficiently and correctly segment out and recognize the wafer ID with high performance.

© 2009 Optical Society of America

OCIS Codes
(100.0100) Image processing : Image processing
(100.5010) Image processing : Pattern recognition
(100.3008) Image processing : Image recognition, algorithms and filters

ToC Category:
Image Processing

History
Original Manuscript: February 24, 2009
Revised Manuscript: October 18, 2009
Manuscript Accepted: October 19, 2009
Published: December 1, 2009

Citation
Wei-Chih Hsu, Tsan-Ying Yu, and Kuan-Liang Chen, "Robust wafer identification recognition based on asterisk-shape filter and high-low score comparison method," Appl. Opt. 48, 6606-6620 (2009)
http://www.opticsinfobase.org/ao/abstract.cfm?URI=ao-48-35-6606

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Article level metrics are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited