Morphological Segmentation of Multispectral Images for Land Cover Mapping

This paper presents an unsupervised segmentation method applied to multispectral satellite images especially SPOT images. The main objective of this work is to combine spectral and contextual information in order to extract the most important cartographic regions. We choose a mathematical morphology context. Previous morphological works are usually interested in one type of land covering area. The proposed technique globalizes the problem by considering all the important regions to perform complete and automatic multispectral satellite images cartography.

[1]  A.J. Plaza,et al.  Automated selection of results in hierarchical segmentations of remotely sensed hyperspectral images , 2005, Proceedings. 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. IGARSS '05..

[2]  Pierre Soille,et al.  Morphological Image Analysis , 1999 .

[3]  Selim Aksoy,et al.  Morphological Segmentation of Urban Structures , 2007 .

[4]  Pierre Soille,et al.  Advances in mathematical morphology applied to geoscience and remote sensing , 2002, IEEE Trans. Geosci. Remote. Sens..

[5]  Peijun Li,et al.  Evaluation of multiscale morphological segmentation of multispectral imagery for land cover classification , 2004, IGARSS 2004. 2004 IEEE International Geoscience and Remote Sensing Symposium.