Machine Vision as a Materials Conservation Technology

Abstract

Conservation of archaeological monuments and artifacts is in need of precise and efficient techniques for materials condition evaluation in order to ensure accurate diagnosis followed by effective curative intervention. Machine Vision, already extensively used in the manufacturing industry, is introduced as a preservation technology to fulfill that need. Material degradation manifests itself as localized variations in the optical response (e.g., color & intensity) from the affected surface. Once captured through an appropriate recording device, the image of the surface is documented, enhanced, analyzed and displayed by use of Digital Image Processing (DIP) techniques. The resulting map of materials condition for the surface is demonstrably comprehensive, quantitative & accurate, and the process is fast and economical. Field work was done at the site of Mayan ruins in Tulum, Mexico, on the shore of the Caribbean sea. This paper focuses on a piece of structural rock having a large number of surface pores as the primary form of degradation. Employing DIP procedures, the material condition of the surface is quantitatively represented and characterized in terms of size as well as spatial distribution of the damage features. On a theoretical note, it is postulated that the sand from the beach, airborne by the sea breeze, had been the principal erosion agent. And a phenomenological model of the weathering mechanism has been developed centering on the fluid mechanical concept of terminal velocity. Close agreement between the field work data and the theoretical construct demonstrates the additional merit of the adapted technology as enabling a verifiable, and hence more meaningful, diagnosis of the weathering mechanism.