Optical Phenomena in Computer Vision,

Abstract

Computer vision programs are based on some kind of model of the optical world, in addition to whatever significance they may have in terms of human vision, algorithms, architectures, etc. There is a school of research that addresses this aspect of computer vision directly, by developing mathematical models of the optics and geometry of image formation and applying these models in image understanding algorithms. In this paper, we examine the optical phenomena that have been analyzed in computer vision and suggest several topics for future research. The three topics that have received the most attention are shading (and glossiness), color, and shadows. Shape-from-shading research, while producing many interesting algorithms and research results, has primarily been based on very simplified models of glossiness. Since realistic gloss models exist within the optics community, we can expect improved computer vision algorithms in the future. Color work in the past has similarly concentrated on developing sophisticated algorithms for exploiting very simple color models, but a more realistic analysis technique has recently been proposed. Shadows have been used by a number of people for simple analysis such as locating buildings in aerial photographs, and a more complex theory already exists that relates surface orientations to shapes of shadows in the image.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1984
Accession Number
ADA152970

Entities

People

  • S. A. Shafer

Organizations

  • University of Rochester

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Aerial Photographs
  • Artificial Intelligence
  • Cameras
  • Cognitive Science
  • Computer Graphics
  • Computer Science
  • Computer Vision
  • Detectors
  • Geometry
  • Image Processing
  • Light Sources
  • Materials
  • Optical Phenomena
  • Optical Properties
  • Optics
  • Photographs
  • Three Dimensional

Fields of Study

  • Computer science

Readers

  • Computer Vision.
  • Technical Research and Report Writing.

Technology Areas

  • AI & ML