Skip to Main content Skip to Navigation
Conference papers

Segmentation and mathematical morphology applied to X-ray microtomographic images of building limestones

Abstract : Historical monuments contributing to our cultural heritage undergo their environment which alters and destroys them. The observed macro-scale effects are the results of physical mechanisms (dissolution, transport, precipitation) occurring in the complex porous structure, at the pore scale. Understanding of weathering processes and more generally of water transfer properties of building stone requires a detailed knowledge of the pore and solid phase distribution. Indeed, quantitative physical models rely on a precise and relevant description of the microstructure. Thus, it is important to characterize stones originating from carriers (unweathered) and from weathered buildings as an initial step of research. X-ray microtomography is a good choice technology to achieve such a goal and to obtain relevant topological information. However it requires the identification of the various phases constituting the porous medium. This segmentation step (sometimes underestimate in the literature) is of great importance to assure the quality of the approach. So, in the case of low contrast images of multiphasic medium this analysis requires a particular attention. We propose a new approach in the field of building stone conservation based on mathematical morphology to segment such images and obtain quantitative descriptions of stones (porosity, autocorrelation functions, chord distributions ...).
Document type :
Conference papers
Complete list of metadata
Contributor : Nathalie Pothier <>
Submitted on : Wednesday, September 27, 2006 - 5:11:03 PM
Last modification on : Friday, April 5, 2019 - 8:08:15 PM


  • HAL Id : hal-00101648, version 1



Olivier Rozenbaum, Emmanuel Le Trong, Jean-Louis Rouet, Ary Bruand. Segmentation and mathematical morphology applied to X-ray microtomographic images of building limestones. GEOX - 2nd International Workshop on X-RAY CT for GEOMATERIALS, 2006, Auusois - Grenoble, France. ⟨hal-00101648⟩



Record views