Viewer-centred feature recognition

M. G L Sommerville, D. E R Clark, J. R. Corney

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

8 Citations (Scopus)


Feature Recognition systems generally use `Object-Centred' models (eg B-Rep and CSG) to represent components and search for features. The authors present a novel `Viewer-Centred' approach to feature recognition, influenced by their work on Aspect Graphs. The motivation behind this is an attempt to create a more robust system, capable of recognizing a wide range of features. Two advantages of the system over conventional techniques are first that it is able to recognize generic DP-features on objects with a range of non-planar faces and second, it can find assembled features. This paper reports the current state of development of the system. The methodology of the approach is detailed and contrasted with previously published work in this area. Results for a variety of test cases are also given.

Original languageEnglish
Title of host publicationSymposium on Solid Modeling and Applications - Proceedings
Number of pages5
Publication statusPublished - 1995
EventProceedings of the 3rd Symposium on Solid Modeling and Applications - Salt Lake City, UT, USA
Duration: 17 May 199519 May 1995


ConferenceProceedings of the 3rd Symposium on Solid Modeling and Applications
CitySalt Lake City, UT, USA


Dive into the research topics of 'Viewer-centred feature recognition'. Together they form a unique fingerprint.

Cite this