Digital image analysis based study, recording, and protection of painted rock art. Some Iberian experiences

This work reviews different approaches, with the common point of using digital image analysis techniques imported from the research field of environmental remote sensing, used the last years by our research group for obtaining information from prehistoric painted rock panels. The obtained results are relevant for the definition of the particular environments of different panels (the complex series of elements composing the natural systems in which the rock art constitute one of the parts, and the relations and synergies connecting all them), for the definition of the rock art itself (in terms of composition, taphonomical history or typology), or simply for improving the vision of faded images, helping in the task of making a secure tracing of the panel. This paper presents several examples in which image uncorrelation by principal components analysis, mixed digital classification, and auto-tracing helps understanding the panels.