UXO Discrimination Study Former Spencer Artillery Range

Abstract

The main technical objective of the Former Spencer Artillery Range demonstration is to validate and substantially automate the SIG learning process using next-generation electromagnetic induction (EMI) sensor data for discriminating targets-of-interest. This process includes three major components, feature extraction, site learning and excavation. At Camp Spencer SIG tested three methods for generating dig lists that vary by the amount of training data required. The SIG discrimination process has proven effective at providing efficient site discrimination with minimal training data. Former Spencer Artillery Range provides another example of this. The addition of adaptive modeling approaches that use an a priori metric of site complexity has increased effectiveness of this approach.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 2013
Accession Number
ADA591761

Entities

People

  • Larry Carin
  • Levi Kennedy
  • Todd Jobe

Organizations

  • Signal Innovations Group, Inc.

Tags

Communities of Interest

  • Weapons Technologies

DTIC Thesaurus Topics

  • Data Sets
  • Detection
  • Detectors
  • Dimensionality Reduction
  • Electromagnetic Induction
  • Excavation
  • Feature Extraction
  • Feature Selection
  • Gaussian Distributions
  • Information Science
  • Machine Learning
  • Munitions
  • Probability
  • Remote Sensing
  • Signal Processing
  • Supervised Machine Learning
  • Unexploded Ammunition

Readers

  • Distributed Systems and Data Platform Development
  • Instructional Design and Training Evaluation.
  • Military/Explosive Ordnance Disposal (EOD) Technology

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks