Decay regions segmentation from color images of ancient monuments using fast marching method

Abstract This work faces the problem of detecting decay zones from color images of stone materials. In Cultural Heritage, the extraction of degradation regions from images of ancient monuments represents an important step forward in studying and analyzing the state of preservation of historical buildings. Generally the decay diagnosis is provided by “naked eye” analysis done by expert scientists “walking around” the artifact and recording the conservation state of each individual element they observe. In addition to this kind of investigation, the application of an image segmentation strategy to color images of stony materials can be used in order to extract regions characterized by a visible chromatic alteration, changes in color, for example, as oxidation or concretion. This paper features a color image segmentation approach founded on the fast marching numerical method. We have applied this technique for its possibility to work locally, that is, only the contour of the region under study is processed. In addition to this method, we present a global approach, that is, the possibility to extract decay regions from the entire image; these regions are spatially disconnected but with similar colorimeter value. The main aim of the present work is to provide a tool that helps the expert to contour the degraded regions. In this sense even if the results of the proposed procedure depend on the expert evaluation, the approach can be a contribution to improve the efficiency of the boundary detection process. The study case concerns the impressive remains of the Roman Theatre in the city of Aosta (Italy). In the image segmentation process the color space L * a * b * is utilized.