Neurofuzzy Analysis of Cloud Motion from Time Series of Satellite Imagery
Neurofuzzy Analysis of Cloud Motion from Time Series of Satellite Imagery
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.
Newland, F.T.
e1b8f8f6-f064-4cd9-bdab-abdb38fe3054
October 1996
Newland, F.T.
e1b8f8f6-f064-4cd9-bdab-abdb38fe3054
Newland, F.T.
(1996)
Neurofuzzy Analysis of Cloud Motion from Time Series of Satellite Imagery
Record type:
Monograph
(Project Report)
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.
This record has no associated files available for download.
More information
Published date: October 1996
Additional Information:
Address: Department of Electronics and Computer Science
Organisations:
Electronics & Computer Science
Identifiers
Local EPrints ID: 250020
URI: http://eprints.soton.ac.uk/id/eprint/250020
PURE UUID: 4b9dbe68-fce2-48da-b5af-28bb052caddc
Catalogue record
Date deposited: 04 May 1999
Last modified: 20 Feb 2024 18:07
Export record
Contributors
Author:
F.T. Newland
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics