Neurofuzzy Analysis of Cloud Motion from Time Series of Satellite Imagery


Newland, F.T. (1996) Neurofuzzy Analysis of Cloud Motion from Time Series of Satellite Imagery.

Download

Full text not available from this repository.

Description/Abstract

This report outlines the first year s research into the application of neural networks and fuzzy logic to the extraction of wind motion vectors from time series of satellite imagery. To improve on current schemes for determining wind information from cloud motion, three new components of satellite image sequences have been used: context, cloud 'object' motion analysis and longer time series. A fuzzy system has been used to identify the motion context within an image sequence of 12 hours of satellite data. Techniques for the automatic extraction of cloud objects and regions have been looked at, and an array clustering algorithm defined on interest operators in an image sequence of 4 hours of data has been developed. A new neurofuzzy approach to cloud region identification and motion analysis is being developed for application to trial images supplied with operationally generated wind data, and is outlined in this report. A timetable for the remainder of the studentship is given.

Item Type: Monograph (Technical Report)
Additional Information: Address: Department of Electronics and Computer Science
Divisions: Faculty of Physical Sciences and Engineering > Electronics and Computer Science
ePrint ID: 250020
Date Deposited: 04 May 1999
Last Modified: 27 Mar 2014 19:50
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/250020

Actions (login required)

View Item View Item