A Hybrid Architecture for Feature Labelling of Oceanographic Satellite Images
Abstract
The objective of this project is to develop an automatic oceanographic image interpretation system. This system will aid the operator in identifying various oceanic features and will predict the dynamic behavior of these features. Presently these tasks are handled by experts in oceanography and obviously, a time consuming process. The proposed project has been divided into three phases: In Phase I the features present in the oceanographic image are identified. In this phase an efficient feature labeling algorithm will be implemented. In Phase II an Expert System that helps in predicting the dynamic behavior of the oceanic features will be developed. In Phase III these two modules will be integrated and the system will be tested with real-time data. The design and development of Phase I started in May 1988. Various techniques for feature labeling were studied. It was decided to use non-linear probabilistic relaxation labeling scheme for the oceanographic feature labeling problem.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 1989
- Accession Number
- ADA212236
Entities
People
- Sundaraja Sitharama Iyengar
Organizations
- Louisiana State University