Features for describing images

Features for describing images

internal features

• formal parameters:
  • color perception HVC (type hue, luminosity value, saturation chroma)
  • granularity, texture,
  • Eigenvalues (used for fingerprints)
  • transforms
    • Fourier
    • Wavelets -- relatable to semantics (James Wang)
    • template match
      • blocks for fields, lines without sharp bends for RR tracks
      • presence of airports
      • presence of airplanes (generalized cones)
      • image understanding

      Previous slide

      Next slide

      Back to the first slide

      View Graphic Version