Automated Cartographic Feature Attribution Using Panchromatic and Hyperspectral Imagery

Abstract

This report summarizes the primary accomplishments made during the second year of the DARPA Automated Population of Geographic Databases (APGD) program. The first part summarizes work in acquiring and registering the Hyperspectral Digital Imagery Collection Experiment (HYDICE) data. It also describes experiments in classifying the HYDICE data using standard maximum-likelihood techniques and explores the development and results of the new spectral angle mapper. The second part covers work begun on aggregating the surface material maps to reduce the number of polygons required without degrading classification accuracy.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2000
Accession Number
ADA380211

Entities

People

  • David M. Mckeown Jr.
  • J. C. Mcglone
  • Jefferey A. Shulelt
  • Steven D. Cochran
  • Wilson A. Harvey

Organizations

  • Carnegie Mellon University

Tags

Communities of Interest

  • Engineered Resilient Systems

DTIC Thesaurus Topics

  • Accuracy
  • Acquisition
  • Classification
  • Computer Science
  • Coordinate Systems
  • Data Acquisition
  • Data Fusion
  • Databases
  • Detectors
  • Discriminant Analysis
  • Feature Extraction
  • Geometry
  • Hyperspectral Imagery
  • Materials
  • Mathematical Models
  • Spectra
  • Three Dimensional

Readers

  • Computer Vision.
  • Image Processing and Computer Vision.
  • Technical Research and Report Writing.