Abstract
An overview is given of a hierarchical fuzzy ATR system that takes fuzzy data, segments out both a set of foundation silhouettes and sets of internal hot spots, uses a fuzzy feature to represent shape for all segments, uses structural features for the internal hot spots, and uses a two step fuzzy classification algorithm to give guesses with degrees of certainty. The process achieves a hardened classification accuracy of 98.5%.
| Original language | English (US) |
|---|---|
| Pages | 309-313 |
| Number of pages | 5 |
| State | Published - 1992 |
| Event | First International Conference on Fuzzy Theory and Technology Proceedings, Abstracts and Summaries - Duration: Oct 14 1992 → Oct 18 1992 |
Other
| Other | First International Conference on Fuzzy Theory and Technology Proceedings, Abstracts and Summaries |
|---|---|
| Period | 10/14/92 → 10/18/92 |
ASJC Scopus subject areas
- Information Systems
- Software
- Computational Theory and Mathematics
- Artificial Intelligence