Ecosystem classification using artificial intelligence neural networks and very high spatial resolution satellite imagery

Keramitsoglou, I; Sarimveis, H; Kiranoudis, CT; Sifakis, N

dc.contributor.author	Keramitsoglou, I	en
dc.contributor.author	Sarimveis, H	en
dc.contributor.author	Kiranoudis, CT	en
dc.contributor.author	Sifakis, N	en
dc.date.accessioned	2014-03-01T02:49:45Z
dc.date.available	2014-03-01T02:49:45Z
dc.date.issued	2004	en
dc.identifier.issn	0277786X	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/34723
dc.subject	Classification	en
dc.subject	IKONOS	en
dc.subject	Neural networks	en
dc.subject	Radial basis function	en
dc.subject.other	Algorithms	en
dc.subject.other	Artificial intelligence	en
dc.subject.other	Data reduction	en
dc.subject.other	Forestry	en
dc.subject.other	Maximum likelihood estimation	en
dc.subject.other	Optical resolving power	en
dc.subject.other	Radial basis function networks	en
dc.subject.other	Satellites	en
dc.subject.other	Wetlands	en
dc.subject.other	Classification	en
dc.subject.other	IKONOS	en
dc.subject.other	Ecosystems	en
dc.title	Ecosystem classification using artificial intelligence neural networks and very high spatial resolution satellite imagery	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1117/12.511041	en
heal.identifier.secondary	http://dx.doi.org/10.1117/12.511041	en
heal.publicationDate	2004	en
heal.abstract	This study investigates the potential of classifying complex ecosystems by applying the radial basis function (RBF) neural network architecture, with an innovative training method, on multispectral very high spatial resolution satellite images. The performance of the classifier has been tested with different input parameters, window sizes and neural network complexities. The maximum accuracy achieved by the proposed classifier was 78%, outperforming maximum likelihood classification by 17%. Analysis showed that the selection of input parameters is vital for the success of the classifiers. On the other hand, the incorporation of textural analysis and/or modification of the window size do not affect the performance substantially. The new technique was applied to the area of Lake Kerkini (Greece), a wetland of great ecological value, included in the NATURA 2000 list of ecosystems.	en
heal.journalName	Proceedings of SPIE - The International Society for Optical Engineering	en
dc.identifier.doi	10.1117/12.511041	en
dc.identifier.volume	5232	en
dc.identifier.spage	228	en
dc.identifier.epage	236	en