A PROPOSED MULTIMODAL CONVOLUTIONAL NEURAL NETWORKS (CNNS) SOLUBLE SALTS MAPPING SYSTEM OF THE SUBSURFACE: THE CASE OF THE WALL PAINTING AT THE ROYAL BOX AT HERODIUM, ISRAEL

Abstract. In 2008, excavations at Herodium revealed magnificent secco wall paintings and stucco decorations adorning the central chamber top of royal theatre. The paintings, dated to first century B.C.E., have been preserved up a height 6 meters. However, shortly after discovery, salts weathering structural faults caused severe damages decorations. conservation process restore lasted almost decade. These efforts helped stabilize state painting, but in very fragile manner, while deterioration factors are still present, any slight change condition enclosure, could damage paintings. This study is aimed assisting conservators developing tool that will offer glance hidden threats subsurface, by help protect historic monuments from salt weathering. paper describe an innovative methodology with particular emphasis on novel multimodal convolutional neural networks (CNNs) technologies data non-destructive testing (NDT) for detection mapping soluble subsurface ancient Prior preforming system protocol situ, laboratory simulation was carried out thermochemical behaviour salts, chlorides sulphates, within different subsets. preliminary results be presented this paper.