Abstract
A probabilistic variant of the decision-based neural network (DBNN) that is meant to better estimate probability density functions corresponding to different pattern classes is developed. New learning rules for the probabilistic DBNN are derived. Experimental results demonstrate noticeable improvement in various performance measures such as recognition accuracies and, in particular, false acceptance/rejection rates. A multiple sensor fusion system for object classification is developed using probabilistic output values of the DBNN.
Original language | English (US) |
---|---|
Pages | 333-342 |
Number of pages | 10 |
State | Published - 1995 |
Event | Proceedings of the 5th IEEE Workshop on Neural Networks for Signal Processing (NNSP'95) - Cambridge, MA, USA Duration: Aug 31 1995 → Sep 2 1995 |
Other
Other | Proceedings of the 5th IEEE Workshop on Neural Networks for Signal Processing (NNSP'95) |
---|---|
City | Cambridge, MA, USA |
Period | 8/31/95 → 9/2/95 |
All Science Journal Classification (ASJC) codes
- Signal Processing
- Software
- Electrical and Electronic Engineering