Spatially-Coherent Non-Linear Dimensionality Reduction and Segmentation of Hyper-Spectral Images (PREPRINT)

Abstract

Non-linear dimensionality reduction and vector segmentation of hyper-spectral images is investigated in this letter. The proposed framework takes into account the nonlinear nature of high dimensional hyper-spectral images, and projects onto a lower dimensional space via a spatially-coherent locally linear embedding technique. The spatial coherence is introduced by comparing individual pixels based on their local surrounding neighborhood structure. This neighborhood concept is also extended to the segmentation and classification stages using a modified vector angle distance. We present the underlying concepts of the proposed framework and experimental results showing the significant classification improvements.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2006
Accession Number
ADA478645

Entities

People

  • Anish Mohan
  • Edward Bosch
  • Guillermo Sapiro

Organizations

  • University of Minnesota

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Artificial Satellites
  • Classification
  • Clustering
  • Computer Vision
  • Data Analysis
  • Data Reduction
  • Data Sets
  • Detectors
  • Dimensionality Reduction
  • Embedding
  • Equations
  • Factor Analysis
  • Geospatial Intelligence
  • Jet Propulsion
  • Ores

Fields of Study

  • Physics

Readers

  • Image Processing and Computer Vision.
  • Operations Research
  • Systems Analysis and Design

Technology Areas

  • Space