Title :
A decision method for object identification in robotics
Author :
Ghallab, M. ; Giralt, G.
Author_Institution :
University of California at Berkeley, Berkeley, California
Abstract :
In this paper we propose a method for generating a deterministic classification tree in the Sequential Pattern Recognition problem. At first, features are partitioned and structured in a set of decision rules. Then this set is translated in an optimal decision tree which satisfies a constraint on the maximum probability of misclassification.
Keywords :
Cameras; Decision trees; Laboratories; Pattern recognition; Robots; Sensor phenomena and characterization;
Conference_Titel :
Decision and Control including the 17th Symposium on Adaptive Processes, 1978 IEEE Conference on
Conference_Location :
San Diego, CA, USA
DOI :
10.1109/CDC.1978.268032
