Abstract
This paper investigates the application of a particular type of probabilistic neural networks, namely radial basis function (RBF) networks, to detecting cloud in NOAA/AVHRR images. Based on the images collected from the East China Sea, the paper compares the performance of RBF networks with that of traditional multi-layer perceptrons (MLPs). The main results show that RBF networks are able to handle complex atmospheric and oceano-graphic phenomena while MLPs could not. The internal representation of the RBF networks and MLPs are also detailed in this paper.
Original language | English |
---|---|
Title of host publication | International Geoscience and Remote Sensing Symposium (IGARSS) |
Pages | 2373-2375 |
Number of pages | 3 |
Publication status | Published - 1 Dec 2001 |
Externally published | Yes |
Event | 2001 International Geoscience and Remote Sensing Symposium (IGARSS 2001) - University of New South Wales, Sydney, NSW, Australia Duration: 9 Jul 2001 → 13 Jul 2001 |
Conference
Conference | 2001 International Geoscience and Remote Sensing Symposium (IGARSS 2001) |
---|---|
Country/Territory | Australia |
City | Sydney, NSW |
Period | 9/07/01 → 13/07/01 |
ASJC Scopus subject areas
- Computer Science Applications
- General Earth and Planetary Sciences