A discourse on geometric feature recognition from cad models

Jami J. Shah, David Anderson, Yong Se Kim, Sanjay Joshi

Research output: Contribution to journalArticlepeer-review

173 Scopus citations


This paper discusses the past 25 years of research in feature recognition. Although a great variety of feature recognition techniques have been developed, the discussion here focuses on the more successful ones. These include graph based and “hint” based methods, convex hull decomposition, and volume decomposition-recomposition techniques. Recent advances in recognizing features with free form features are also presented. In order to benchmark these methods, a frame of reference is created based on topological generality, feature interactions handled, surface geometry supported, pattern matching criteria used, and computational complexity. This framework is used to compare each of the recognition techniques. Problems related to domain dependence and multiple interpretations are also addressed. Finally, some current research challenges are discussed.

Original languageEnglish (US)
Pages (from-to)41-51
Number of pages11
JournalJournal of Computing and Information Science in Engineering
Issue number1
StatePublished - Mar 2001

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Industrial and Manufacturing Engineering


Dive into the research topics of 'A discourse on geometric feature recognition from cad models'. Together they form a unique fingerprint.

Cite this