A Stellar Application of the IUE: Solar Feature Extraction

Here is a postscript version of this paper. [707 Kb]

ABSTRACT

The Image Understanding Environment (IUE) project is a five year program, sponsored by DARPA, to develop a common object-oriented software environment for the development of algorithms and application systems. This paper builds upon previous reports (Kohl and Mundy, 1994), (Mundy and IUE Committee, 1993a), (Mundy and IUE Committee, 1993b), (Mundy et. al., 1993) to review the design of this system and provides an overview of the implementation effort currently underway at Amerinex AI, Advanced Decision Systems, Carnegie Mellon University, and Colorado State University. This effort has the goal of a full C++ implementation of the IUE-Core by late 1995.

The ultimate goal of the IUE project is to provide a software infrastructure (class hierarchies, user interface tools, and IU algorithms) that is sufficient to support state-of-the-art research in image understanding. In order to verify that the design and implementation of the IUE meets this goal, we have begun the development of an IUE research application in the extraction and tracking of features in solar image data. This paper summarizes how the IUE class hierarchy supports this IU research effort and the way in which the current partial implementation of the IUE class hierarchy facilitates the development of robust and useful solutions in this domain.