GRAPHICAL DATA PROCESSING RESEARCH STUDY AND EXPERIMENTAL INVESTIGATION

Abstract

The trainability of learning machi es is discussed. A non-mathematical descriptive treatment is pr sented, covering the fundamentals of elementary machines. An attempt was made to develop an intuitive apprecia ion of w y these machines work. A mathematical description is then given of the geometrical interpretation of successive transformations by which a pattern at the machine input becomes a point in the classification space, and of a test for linear separability. Task E was completed with a survey of indexing and r trieval of non-alphanumeric graphical information for ili ary purposes. There appear to be no basic methodological problems to prevent t e development of an index that provides complete and sys ematic reference to all significant topological features or man-made objects shown on printed maps and aerial photographs. (Author)

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1961
Accession Number
AD0273440

Entities

People

  • A.e. Brain
  • N.j. Nilsson

Organizations

  • SRI International

Tags

DTIC Thesaurus Topics

  • Aerial Photographs
  • Classification
  • Coverings
  • Data Processing
  • Image Processing
  • Images
  • Information Processing
  • Learning
  • Photographic Images
  • Photographic Materials
  • Photographs
  • Photography

Readers

  • Neural Network Machine Learning.
  • Snow Cover Descriptors for Reptiles and Their Illustrations.
  • Theoretical Analysis.

Technology Areas

  • Space