Fast Image Stitching for Continuous Casting Steel Billet Images

Chao Yung Hsu, Jliili Wei Huang, Li Wei Kang, Ming Fang Weng

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

To achieve the goal of industry 4.0 for steel industry, intelligent automated manufacturing becomes increasingly important. One of the key steps on the automatic production line of steel products is automatic inspection of surface detects. In this paper, we consider the problem of stitching continuous casting steel billet images captured from billet production line. It has been shown that integrated individually captured continuous billet images would be beneficial to vision-based surface inspection. This paper presents a fast image stitching fr amework for continuous billet images relying on homography estimation. Experimental results have demonstrated that our method achieves fast image stitching with accurately stitched results.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509027439
DOIs
Publication statusPublished - 2017 Jan 3
Externally publishedYes
Event2016 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2016 - Seoul, Korea, Republic of
Duration: 2016 Oct 262016 Oct 28

Publication series

Name2016 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2016

Conference

Conference2016 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2016
Country/TerritoryKorea, Republic of
CitySeoul
Period2016/10/262016/10/28

Keywords

  • Billet images
  • Defect inspection
  • Homography estimation
  • Linage stitching
  • Steel products

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Instrumentation

Fingerprint

Dive into the research topics of 'Fast Image Stitching for Continuous Casting Steel Billet Images'. Together they form a unique fingerprint.

Cite this