Integrated Class Structures for Image Pattern Recognition and Computer Graphics,
Abstract
Research efforts in image pattern recognition and computer graphics face two kinds of software problems. First is the intrinsic complexity of the algorithms developed in the course of the research. Second is the design and construction of the researcher's software toolbox including fundamental operations, standards for data storage and communication, and interfaces to rapidly changing sets of display and interaction devices. The complexity problem is intrinsic to the subject matter and objectives, and it is the proper domain of the researcher. This paper describes several techniques we have developed while designing an integrated object-oriented software toolbox for image pattern recognition and interactive computer graphics research. Design criteria for the system include (1) pervasive integration of constructs, (2) maximum flexibility for researchers using the system, (3) minimum user effort to invoke the facilities, and (4) purity of the object-oriented design.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1987
- Accession Number
- ADA192504
Entities
People
- James M. Coggins
Organizations
- University of North Carolina at Chapel Hill