Analysis of Accretion and Deletion at Boundaries in Dynamic Scenes.

Abstract

Locating object bundaries in images is an important but difficult problem. Intensity-based edge detection provides ambiguous or misleading boundary information in many situations, such as textured regions. Motion-based techniques can provide more reliable results in these cases. At object boundaries where occlusion occurs, surface regions will typically appear or disappear over time when motion is present. These regions of changing visibility may be used to indicate both object boundaries and the side of the boundary corresponding to the occluded surface. Thus, in dynamic scenes, regions of surface accretion or deletion can be found using matching technique similar to those used to determine optical flow in an image sequence. Regions in one frame that are not adequately matched by any region in previous frames correspond to accretion. Regions that have no matches in subsequent frames correspond to deletion. In either case, an occlusion boundary is present. Furthermore, by associating accretion or deletion regions with a surface on one side of a boundary, it is possible to determine which side of the boundary is being occluded. This association can be based purely on visual motion - the accretion or deletion region moves with the same image velocity as the remaining visible surface to which it is attached.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 1984
Accession Number
ADA145944

Entities

People

  • K. M. Mutch
  • W. B. Thompson

Organizations

  • University of Minnesota

Tags

Communities of Interest

  • Air Platforms

DTIC Thesaurus Topics

  • Boundaries
  • Change Detection
  • Computer Science
  • Computers
  • Contracts
  • Detection
  • Detectors
  • Disparities
  • Displacement
  • Identification
  • Image Processing
  • Intensity
  • Minnesota
  • Pattern Recognition
  • Recognition
  • Security
  • Sequences

Readers

  • Computer Vision.
  • Ocean-Atmosphere Mesoscale Modeling, Data Assimilation, and Flux Boundary Layers
  • Regression Analysis.