Application of multifractal analysis to segmentation of water bodies in optical satellite images

Victor Manuel San Martin; Alejandra Figliola

doi:10.4279/pip.090007

Authors

Victor Manuel San Martin National Scientific and Technical Research Council (CONICET), Argentina. Institute of Human Development, National University of General Sarmiento, Juan María Gutiérrez 1150, (1613) Los Polvorines, Buenos Aires, Argentina.
Alejandra Figliola National Scientific and Technical Research Council (CONICET), Argentina. Institute of Human Development, National University of General Sarmiento, Juan María Gutiérrez 1150, (1613) Los Polvorines, Buenos Aires, Argentina.

DOI:

https://doi.org/10.4279/pip.090007

Keywords:

Multifractal Analysis, Remote Sensing Images, Segmentation

Abstract

In this paper we study the characteristics of multifractal spectra obtained for remote sensing images through the coarse theory and proposed a method for segmentation of water bodies based on it.
In first place, spectra of self-similar images created with Iterated Function Systems are calculated and compared with their statistical spectra. Then, optical remote sensing images are studied, emphasizing the differences between their spectra and those obtained for synthetic images. Punctually, attention is focused on the concavity of real image spectra and on its interpretation in terms of the analyzed images. This lead us to the proposition of a segmentation method for water bodies.
The method is tested and the results are compared with water masks of the regions under study. Comments are made about the limitations of the proposed method.

Received: 29 May 2017, Accepted: 20 September 2017; Edited by: C. Muravchik; DOI: http://dx.doi.org/10.4279/PIP.090007

Cite as: V M San Martin, A Figliola, Papers in Physics 9, 090007 (2017)

This paper, by V M San Martin, A Figliola, is licensed under the Creative Commons Attribution License 3.0.