Text region extraction algorithm on steel making process

  • Authors:

  • SungHoo Choi;Jong Pil Yun;KeunHwi Koo;JongHyun Choi;Sang Woo Kim

  • Affiliations:

  • Division of Electrical and Computer Engineering, Pohang University of Science and Technology, Pohang, Korea;Division of Electrical and Computer Engineering, Pohang University of Science and Technology, Pohang, Korea;Division of Electrical and Computer Engineering, Pohang University of Science and Technology, Pohang, Korea;Division of Electrical and Computer Engineering, Pohang University of Science and Technology, Pohang, Korea;Division of Electrical and Computer Engineering, Pohang University of Science and Technology, Pohang, Korea

  • Venue:
  • ROCOM'08 Proceedings of the 8th WSEAS International Conference on Robotics, Control and Manufacturing Technology
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes about the extraction of slab management numbers (SMNs) from captured still images for automation of steel making process. In order to prevent the product mixing in each steel making stage, automatic recognition system of the SMNs is essential. Moreover, the SMNs extraction algorithm which has robust performance is necessary because it affects seriously to the performance of the entire recognition system. For several decades, many researchers have researched about the scene text extraction from images and thus many algorithms exist. But, these algorithms are not well suited for our application because captured still images have much noise and especially there is not enough time to apply the time-consuming state-of-the-art algorithms. In this paper, we propose a localization method of text region candidates by using text features and selection method of true text among the text candidates. Finally, experimental results show that our fast algorithm is operates reliably.