Contour-shape based reconstruction of fragmented, 1600 BC wall paintings

Abstract

In this paper a novel general methodology is introduced for the computer-aided reconstruction of the magnificent wall-paintings of the Greek island Thera (Santorini), painted in the middle of the second millennium BC. These wall-paintings are excavated in fragments and, as a result, their reconstruction is a painstaking and a time-consuming process. Therefore, in order to facilitate and speed up this process a proper system has been developed based on the introduced methodology. According to this methodology each fragment is photographed, its picture is introduced to the computer, its contour is obtained and subsequently all fragments contours are compared in a manner proposed herein. Both the system and the methodology presented here, extract the maximum possible information from the contour shape of fragments of an arbitrary initially unbroken plane object, to point out possible fragments matching. This methodology has been applied to two excavated fragmented wall-paintings consisting of 262 fragments, with full success but most important it has been used to reconstruct, for the first time, unpublished wall-paintings parts from a set of 936 fragments. 2 A. INTRODUCTION-PROBLEM DESCRIPTIONThe discovery of the wall-paintings at Akrotiri of the Greek island Thera (Santorini), is of outstanding importance for human knowledge of the early Aegean world and not only. According to prominent archaeologists these wall-paintings rank alongside the greatest archaeological discoveries.The late professor Marinatos originated the excavations, which are now successfully continued by Professor Christos Doumas. As with the treasures of Pompeii and Herculaneum, the wall-paintings of Thera were preserved due to the seal of the pumice from the great eruption of a volcano [1]. As a rule, the walls decorated with paintings no longer survive. They collapsed together with their painted coat before the volcanic eruption, due to particularly strong earthquakes. Thus, a single painting is usually scattered into many fragments mixed with the fragments of other wall-paintings, too. The restoration of the wall-paintings from the fragments is a very painstaking and time consuming process frequently demanding many months or even years of dedicated, experienced personnel work for a single wallpainting restoration. Therefore, the development of a system that will contribute to the automatic restoration of these wall-paintings is of fundamental importance for this archaeological research, but for many others too, which face the problem of an image reconstruction from excavated fragments.Each excavated wall-painting fragment after being cleaned, is being photographed with a very strict protocol, so that very similar illumination conditions, a fixed distance of the fragment plane from the camera focus and minimal photo distortion are ensured. Subsequently, the obtained image is processed and eventually each photographed fragment is embedded into a white background frame, which we call the absolute frame of reference of the specific fragmen...

Document reconstruction using dynamic programming

“…Leitão and Stolfi [5] propose an algorithm based on incremental dynamic programming to reconstruct ceramic tiles. Interesting results also have been reported by Papaodysseus et al [6] where the focus is the reconstruction of archaeological wall-paintings.…”

Virtual and Physical Recomposition of Fragmented Ecclesiastical Frescoes Using a Photogrammetric Approach

“…In the domain of virtual restoration of frescoes, most examples focus on the automatic matching of scattered bidimensional datasets, based on shape properties of pictorial content of decoration (Papaodysseus et al, 2002;Renna et al, 2005;Fornasier et al, 2005;Riccio et al, 2015) and tridimensional datasets (Huang 2006;Brown 2010). Although advanced image processing techniques are being incorporated in most recent systems, semi-automatic solutions where cultural heritage professional are involved still represent the most reliable approach (Scopigno at al., 2011).…”

A graph algorithmic framework for the assembly of shredded documents

“…Another closely related problem is the reconstruction of wall paintings which is addressed by Papaodysseus et al [10]. They rely on contour shape-based fragment matching.…”