Geometric, Algebraic and Topological Structure for Signal and Image Processing

Abstract

Military information collected from streaming video of real-time unmanned aerial reconnaissance, battlefield operations or from multi-spectral satellite imagery, for example, requires intelligent processing by analysts to reduce response time and increase accuracy of interpretation. This research concerns the understanding of, i.e., modeling and classifying of, information in data. Of primary focus has been development of new algorithms for processing large data sets. In particular the proposed techniques emphasize exploiting all available data rather than reducing the data before processing. We have made fundamental progress in the classification problem via encoding sets of images as points on parameter spaces as well as in the model fitting problem using radial basis functions that represent geometry in data.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 19, 2007
Accession Number
ADA477141

Entities

People

  • Michael Kirby

Organizations

  • Colorado State University

Tags

Communities of Interest

  • Autonomy
  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Aerial Reconnaissance
  • Algorithms
  • Classification
  • Computer Vision
  • Data Sets
  • Dimensionality Reduction
  • Geometry
  • Image Processing
  • Image Recognition
  • Information Processing
  • Information Science
  • Low Resolution
  • Machine Learning
  • Mathematics
  • Pattern Recognition
  • Signal Processing

Readers

  • Computer Vision.
  • Systems Analysis and Design

Technology Areas

  • Autonomy
  • Space