Use of Hierarchical Stepwise Optimization for the Segmentation of Cloud Features
Abstract
The Hierarchical Stepwise Optimization algorithm of Beaulieu and Goldberg (1989) is applied to the segmentation of satellite images into meaningful, large scale cloud features. GOES-W visible imagery is used. Several different forms of the cost function are explored in an attempt to improve the segmentation of Nov. 15, 1983 case. A new cost function is shown to result in a superior segmentation. Both HSWO versions are tested on six additional cases. A modification to the HSWO approach is suggested for further research.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1991
- Accession Number
- ADA238987
Entities
People
- J. E. Peak
Organizations
- Computer Sciences Corporation