Learning Combination of Evidence Functions in Object Recognition

D. Cook, L. Hall, L. Stark and K. Bowyer

This paper describes a learning system, OMLET, which helps to automate the construction of function-based object recognition systems. OMLET is designed to learn fuzzy membership functions which approximate the allowable ranges of physical dimensions such as width, area, relative orientation, etc. The learning is done by iterative error reduction on a set of labeled training examples. Results from a chair domain show that the system is capable of learning the fuzzy membership functions.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.