Reducing satellite imagery classification uncertainty through spatiotemporal reasoning.

Abstract

The natural habitat constantly endures both inherent natural and human-induced influences. Remote sensing has been providing monitoring oriented solutions regarding the natural Earth surface, by offering a series of tools and methodologies which contribute to the sustainable environmental management. Processing and analysis of multi-temporal satellite images for the observation of the land use/land changes includes often classification and change-detection techniques. These error prone procedures are influenced mainly by the distinctive characteristics of the study areas, the remote sensing systems limitations and the image analysis methodologies and algorithms. The present study takes advantage of the temporal continuity of multi-temporal classified images, in order to improve the results of the classification in a forest environment, based on reasoning rules. Computational tools are developed in order to disclose the alterations in land use changes and offer a spatial reference to the pressures that the classes endure and impose between them. Moreover, by disseminating areas that are susceptible to misclassification, we can propose a specific target site selection for training during the process of supervised classification. The underlying objective is to contribute to the understanding and analysis of anthropogenic and environmental factors that influence land use changes. The developed algorithms have been tested upon Landsat satellite image time series, depicting the National Park of Ainos in Kefallinia, Greece, where the unique in the world Abies Chephalonica grows. Along the minor changes and pressures indicated, our algorithms have successfully captured historical fire incidents. Overall, the results have shown that the use of the suggested procedures can contribute towards the advancement of the classification performance of satellite images and support the existing knowledge regarding the pressure among land-use changes. Θέμα: EGU2014 Letter of Acceptance Από: egu2014@copernicus.org (egu2014@copernicus.org) Προς: gmiliar@yahoo.com; Ημeρομηνία: 11:37 π.μ. Τeτάρτη, 5 Φeβρουαρίου 2014 Dear Panagiotis Partsinevelos, We are pleased to inform you about the acceptance of your following Abstract for the EGU General Assembly 2014: EGU2014-15356 Reducing uncertainty on satellite image classification through spatiotemporal reasoning by Panagiotis Partsinevelos et al. accepted in NP1.3/GI1.7 The organizers will schedule your contribution and decide about its presentation type by 12 March 2014 and you will be informed accordingly. Detailed information on the pre-registration as well as the hotel booking are provided on the conference webpage http://www.egu2014.eu In case any questions arise, please contact us! Kind regards, Katja Ganger Copernicus Meetings egu2014@copernicus.org on behalf of the Programme Committee Chair