Skip to main navigation Skip to search Skip to main content

Sub-pixel coded structured light stripe boundary location using weighted centroid method

  • School of Computer Science and Technology, Harbin Institute of Technology
  • Harbin Normal University

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

Abstract

Accurate stripe boundary location is crucial for stripe coded structured light based 3D reconstruction. Intersecting point method might be the most accurate for stripe boundary location. But this method requires project a group of normal and a group of inverse patterns, and thus can not be used in some time- or resource-constrained applications. In this paper, we propose an improved sub-pixel location scheme based on weighted centroid algorithm. In the proposed scheme, a green slit of a pixel width is first added between every stripe to avoid false detection and omitted detection. Then, universal gravity-based edge detection and weighted centroid method are utilized for sub-pixel location. Compared with intersecting point method, the proposed scheme achieves comparable detection accuracy with only half number of the projected patterns.

Original languageEnglish
Title of host publication2010 International Conference on Biomedical Engineering and Computer Science, ICBECS 2010
DOIs
StatePublished - 2010
Externally publishedYes
Event2010 International Conference on Biomedical Engineering and Computer Science, ICBECS 2010 - Wuhan, China
Duration: 23 Apr 201025 Apr 2010

Publication series

Name2010 International Conference on Biomedical Engineering and Computer Science, ICBECS 2010

Conference

Conference2010 International Conference on Biomedical Engineering and Computer Science, ICBECS 2010
Country/TerritoryChina
CityWuhan
Period23/04/1025/04/10

Keywords

  • Edge detection
  • Gray code
  • Structured light

Fingerprint

Dive into the research topics of 'Sub-pixel coded structured light stripe boundary location using weighted centroid method'. Together they form a unique fingerprint.

Cite this