Spatially constrained clustering over GIS generated suitability maps

Panagiotis Partsinevelos; Kostas Papadakis

doi:10.1117/12.2194432

Translator Disclaimer

19 June 2015 Spatially constrained clustering over GIS generated suitability maps

Panagiotis Partsinevelos, Kostas Papadakis

Author Affiliations +

Proceedings Volume 9535, Third International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2015); 95351O (2015) https://doi.org/10.1117/12.2194432
Event: Third International Conference on Remote Sensing and Geoinformation of the Environment, 2015, Paphos, Cyprus

Abstract

An abundance of GIS and Remote Sensing based spatial analysis studies result in various types of suitability maps, where selected regions are classified according to application driven qualitative or quantitative rules. Often, upon the resulting classified regions which define spatially constrained classes, users intent to position facilities in order to satisfy a series of demand sites spread throughout the study area. This fine tuning procedure, not tackled under classic clustering and location analysis algorithms, is addressed through the extension of k-means algorithm, by restricting cluster centers inside a priori outlined regions, while minimizing distance metrics towards demand locations. Experimentation in both synthetic and real based datasets shows the applicability of the approach and demonstrates the overall performance of the algorithm.

Citation Download Citation

Panagiotis Partsinevelos and Kostas Papadakis "Spatially constrained clustering over GIS generated suitability maps", Proc. SPIE 9535, Third International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2015), 95351O (19 June 2015); https://doi.org/10.1117/12.2194432

ACCESS THE FULL ARTICLE