Abstract

This paper proposes an efficient approach to rare feature recognition from a boundary representation solid model with Fuzzy ART neural networks. A definition of rare feature is given, the complement coding is used at the preprocessing stage to solve the category proliferation problem, and the normalized input vector which is suitable for the Fuzzy ART neural network is adopted to represent the features. To learn the rare feature rapidly, fast learning is adopted in the Fuzzy ART neural network. Finally, a case study is given to verify the proposed approach.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call