Computer Detection of Low Contrast Targets.
Abstract
The problems of finding extremal edges, finding curvatures along the boundary of objects, and detecting color edges, at first sight appear unrelated. The formulation of these problems in the framework of vector fields and the application of this theoretical model to data both from the range and from the laboratory are the subjects of this paper. The paper is divided into three parts. In the first section we discuss external edges and their detection, section 2 discusses curvature and its measure and section 3 formulates the detection of color edges in vector field language and uses the theoretical results of sections 1 and 2 to detect color edges. Each section follows the same paradigm, namely: The problems is formulated as a problem in the theory of vector fields; The detection of a particular property is formulated as the computation of an integral over a vector field; and The theoretical results are applied to data obtained from the field or in the lab. An outline of the mathematical methods are contained and are based on classical vector analysis. Details of the mathematics used appear for supporting material on differential geometry.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 18, 1982
- Accession Number
- ADA117345
Entities
People
- Raul G. Machuca