Non-conventional algorithm for representing and recognizing complicated two-dimensional objects

D. C. Douglas Hung

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


A non-conventional method for representing and recognizing shape descriptions is described. For any shape S, a modified Cobweb form representation, which is referred to as a shell descriptor (SD), is established for the most significant vertex V. The shell descriptor is then used as a registered description for the shape S at vertex V. This method is computationally efficient as compared to most of the conventional techniques in shape recognition. The new algorithm provides an effective and reliable tool for shape representation and classification, especially for complicated objects such as logos.

Original languageEnglish (US)
Pages (from-to)495-504
Number of pages10
JournalPattern Recognition
Issue number4
StatePublished - Apr 1993

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence


  • Complicated shape recognition
  • Logo inspection
  • Shell descriptor
  • Significant vertex

Cite this