Algoritmi di Image Analysis applicati alle immagini diagnostiche: nuove metodologie per l'analisi conoscitiva ed estrazione semi-automatica della mappatura del degrado

This work proposes a methodology of statistical analysis of diagnostic images aimed at improving their reading and facilitating the graphic transcription of the state of conservation of artistic assets, making it punctual and repeatable. As a case study we present a small oil painting on canvas of an anonymous author in a bad state of conservation. Using the proposed methodology, based on a semi-automatic approach of extraction of the areas of interest, we obtain several survey sheets related to the conservation status through which it is possible to perform zonal statistics to compute the percentage of damage. The operations shown can be applied to any type of diagnostic image, studying more objectively the state of conservation of any artifact.