Temporal Extrapolation of Satellite Imagery for Short-Range Cloud Forecasting
Small Business Information
5951 Encina Road, Suite 208, Goleta, CA, 93117
John G. Devore
AbstractVisidyne proposes to develop an expert, neural net based, system for extrapolating satellite imagery forward in time to provide short-range forecasts of cloud cover. In Phase I Visidyne proposes to investigate the limits of the temporal extrapolation of satellite cloud imagery. For locations in a variety of climate zones, Visidyne will determine how long overlying clouds can be tracked backward in time in satellite images. If these times are sufficiently long, Visidyne will propose a Phase II effort to develop a short-range forecasting system employing an expert, neural net based, system for extrapolating satellite cloud imagery forward in time. The primary benefit of Phase I will be information regarding the utility of developing a short-range cloud forecasting scheme based on the extrapolation of satellite imagery. If the derived time scales are sufficiently long, then experience gained in Phase I will be used to develop a forecasting scheme, which will be proposed in Phase II. Such a forecasting scheme could be adapted for a variety of civilian uses such as in scheduling LANDSAT use or planning cloud seeding activities.
* information listed above is at the time of submission.