Laboratory and in situ assays of digital image analysis based protocols for biodeteriorated rock and mural paintings recording

Abstract

Rock art paintings, and in general mural paintings, are one of the many elements of cultural heritage complex systems. As the different elements of a system have diverse spatial positions, spatial recording allows understanding their interactions. Thus, a useful approach to mural paintings recording is to understand it as a microcartography issue, managing each element of the system as a cartographic coverage. The approach implemented emphasizes the utilization of data obtained by remote sensing techniques for extracting different kinds of information susceptible of being analysed, classified and plotted in a differentiate way by means of the possibility of reducing redundant data by Principal Component Analysis (PCA) and the elaboration of false-colour images from uncorrelated bands. A laboratory model was prepared in order to simulate biodeterioration of rock art. The samples were photographically recorded thereafter under different lighting conditions, and PCA applied to the resulting images. False-colour images obtained by combining Principal Component bands allowed us to reach results similar to those of an unsupervised classification. The method has been applied to Roman mural paintings from one of the tombs of Carmona Necropolis, obtaining good results.