Automatic Feature Recognition Technique on Stone Monuments Using Visible and IR Photography

Visible photography has long been used in the field of architecture for documentation and restoration tasks. However, the automatic extraction and analysis of facade features, i. e. materials and alterations, is unusual in most of multidisciplinary teams involving documentation, conservation and restoration projects. This paper shows the profitability of applying digital image analysis on stone monuments, in order to provide exhaustive, global and accurate information from visible and near-infrared photographs. Output data are rectified and automatically classified images from which it is possible to obtain, among others, detailed positions of materials and degradations on facades, real measures of distances, areas of architectural features, as well as statistical documents of the accuracy achieved in the whole process of automatic recognition.