Dotted and curved line character segmentation

Khader Mohammad, Sos Agaian, Hani Saleh

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper presents new methodology for addressing curvature and segmentation in text as applied to dotted line images. Several improvements are provided, including methods and procedures for handling (a) curvature in text detection and segmentation, (b) rotation angle determinations, (c) isolation of only the character pixels in the image, (d) use of a fill algorithm to solidify the dotted character, and (e) discernment of segmented characters which may be printed too closely or that touch each other. Applying the new algorithms under various lighting conditions against 90 water bottle images (from Ozarka® and Dasani®) with differing text yielded a 93% accuracy rate. Additionally, the fill algorithm presented here improved recognition by more than 20% as compared to non-filled characters.

Original languageBritish English
Title of host publicationProceedings of SPIE-IS and T Electronic Imaging - Intelligent Robots and Computer Vision XXVIII
Subtitle of host publicationAlgorithms and Techniques
DOIs
StatePublished - 2011
EventIntelligent Robots and Computer Vision XXVIII: Algorithms and Techniques - San Francisco, CA, United States
Duration: 24 Jan 201125 Jan 2011

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume7878
ISSN (Print)0277-786X

Conference

ConferenceIntelligent Robots and Computer Vision XXVIII: Algorithms and Techniques
Country/TerritoryUnited States
CitySan Francisco, CA
Period24/01/1125/01/11

Keywords

  • Angled text
  • Character recognition
  • Connected characters
  • Dotted character
  • Fill pixels image
  • Rotation
  • Segmentation

Fingerprint

Dive into the research topics of 'Dotted and curved line character segmentation'. Together they form a unique fingerprint.

Cite this