Kernel Multi-Metric Learning for Multi-Channel Transient Acoustic Signal Classification

Abstract

In this paper, we propose a kernel multi-metric learning algorithm for multi-channel transient acoustic signal classification. The proposed method learns a set of metrics jointly for multi-channel transient acoustic signals in a kernel-induced feature space to exploit the non-linearity of the data for improving the classification performance. An effective algorithm is developed for the task of learning multiple metrics in the kernel space. By learning the multiple metrics jointly within a single unified optimization framework, we can learn better metrics to integrate the multiple channels of the signal for a joint classification. Experimental results compared with classical as well as recent algorithms on real-world acoustic datasets verified the effectiveness of the proposed method.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2012
Accession Number
ADA571140

Entities

People

  • Haichao Zhang
  • Nasser M. Nasrabadi
  • Thomas Huang
  • Yanning Zhang

Organizations

  • Northwestern Polytechnical University

Tags

Communities of Interest

  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Acoustic Detectors
  • Acoustic Signals
  • Algorithms
  • Classification
  • Coefficients
  • Computer Science
  • Detection
  • Detectors
  • Hilbert Space
  • Kernel Functions
  • Learning
  • Measurement
  • Military Research
  • Sensor Networks
  • Supervised Machine Learning
  • Training

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.

Technology Areas

  • Space