Linear features extraction with an orientation constrained probabilistic hough transform

Kuan Tsung Chang, Tian Yuan Shih

Research output: Contribution to journalArticlepeer-review

Abstract

This study proposed a Hough Transform algorithm based on the probabilistic scheme termed as the Orientation Constrained Probabilistic Hough Transform. The orientation constraints for segmentation are applied to form compact and reliable sampling subsets. This process is subsequently followed by constrained searching. Numerical experiments are performed with both synthetic and real datasets indicate that the proposed method performs better than other algorithms in terms of correctness and omission rate. Moreover, computational time deemed necessary is much less than that for other algorithms.

Original languageEnglish
Pages (from-to)157-168
Number of pages12
JournalInternational Journal of Image and Graphics
Volume8
Issue number1
DOIs
StatePublished - 1 Jan 2008

Keywords

  • Computational performance
  • Constrained searching
  • Geometric parameterization
  • Line pattern recognition
  • Random sampling

Fingerprint

Dive into the research topics of 'Linear features extraction with an orientation constrained probabilistic hough transform'. Together they form a unique fingerprint.

Cite this