Joint Use of Dynamical Classifiers and Ambiguity Plane Features

Abstract

This paper argues for using ambiguity plane features within dynamic statistical models for classification problems. The relative contribution of the two model components are investigated in the context of acoustically monitoring cutter wear during milling of titanium, an application where it is known that standard static classification techniques work poorly. Experiments show that explicit modeling of long-term context via a hidden Markov model state improves performance, but mainly by using this to augment sparsely labeled training data. An additional performance gain is achieved by using the shorter-term context of ambiguity plane features.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2001
Accession Number
ADA436824

Entities

People

  • G. D. Bernard
  • L. Atlas
  • M. Ostendorf
  • O. Cetin
  • R. Fish
  • S. Sukittanon

Organizations

  • University of Washington

Tags

Communities of Interest

  • Air Platforms

DTIC Thesaurus Topics

  • Abstracts
  • Ambiguity
  • Case Studies
  • Classification
  • Elements
  • Frequency
  • Hidden Markov Models
  • Information Operations
  • Machine Learning
  • Markov Models
  • Military Research
  • Models
  • Monitoring
  • Stationary
  • Titanium
  • Training

Readers

  • Computational Modeling and Simulation
  • Computer Vision.