Skip to Main Content
Skip Nav Destination
ASME Press Select Proceedings
Intelligent Engineering Systems through Artificial Neural Networks Volume 18
Editor
Cihan H. Dagli
Cihan H. Dagli
Search for other works by this author on:
ISBN-10:
0791802823
ISBN:
9780791802823
No. of Pages:
700
Publisher:
ASME Press
Publication date:
2008

The ocean surface wind vector field is a key element for accurate short term weather forecasting. Those forecasts are needed to issue timely weather warnings to avoid major catastrophes. In recent years, considerable effort has been expended to measure and forecast ocean surface wind speed using data provided by satellites. Analyses based on these massive data sets should be based on a subset of the full retrieved data to make the process efficient for use with on-line algorithms.

This work builds upon successful application of support vector regression (SVR) and Voronoi tessellation to extract a data subset composed of support vectors. A pipeline method is developed to manage an on-line stream of the above mentioned data. Experiments show that the subsets reconstruct the wind vector field with high accuracy. Furthermore, the time required to generate the subset, using the pipeline, is ≈9% of that for thinning the whole data set.

Abstract
1. Introduction
2. Methodology
3. Experiments and results
4. Conclusions
Acknowledgements
References
This content is only available via PDF.
You do not currently have access to this chapter.
Close Modal

or Create an Account

Close Modal
Close Modal