Polygonization of Remote Sensing Classification Maps by Mesh Approximation - Laboratoire Interdisciplinaire des Sciences du Numérique Access content directly
Conference Papers Year : 2017

Polygonization of Remote Sensing Classification Maps by Mesh Approximation

Abstract

The ultimate goal of land mapping from remote sensing image classification is to produce polygonal representations of Earth's objects, to be included in geographic information systems. This is most commonly performed by running a pix-elwise image classifier and then polygonizing the connected components in the classification map. We here propose a novel polygonization algorithm, which uses a labeled triangular mesh to approximate the input classification maps. The mesh is optimized in terms of an 1 norm with respect to the classifiers's output. We use a rich set of optimization operators , which includes a vertex relocator, and add a topology preservation strategy. The method outperforms current approaches , yielding better accuracy with fewer vertices.
Fichier principal
Vignette du fichier
Maggiori_ICIP2017.pdf (2.7 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01530460 , version 1 (31-05-2017)

Identifiers

  • HAL Id : hal-01530460 , version 1

Cite

Emmanuel Maggiori, Yuliya Tarabalka, Guillaume Charpiat, Pierre Alliez. Polygonization of Remote Sensing Classification Maps by Mesh Approximation. ICIP 2017 - IEEE International Conference on Image Processing, Sep 2017, Beijing, China. pp.5. ⟨hal-01530460⟩
486 View
472 Download

Share

Gmail Facebook X LinkedIn More