Orthophoto imaging and GIS for seabed visualization and underwater archaeology

We present here the first step of an interdisciplinary work dealing with underwater photogrammetry and archaeological data management. In the framework of a phd project we develop a set of tools from underwater data capture to 3D underwater GIS for archaeological excavation. The phd project, managed by Julien Seinturier, is monitored by Odile Papini for the data fusion aspect and Pierre Drap for the underwater photogrammetrical aspect. The project is financed together by the French Region PACA and the COMEX firm, specialized in underwater exploration. In this paper we present the first step experimentation on and underwater archaeological site in Corsica managed by the DRASSM and excavated under the scientific responsibility of Franca Cibecchini. After a brief archaeological presentation of the site we'll start by a description of the photogrammetric survey phase, made during the excavation, in the same time that a traditional manual survey make thanks to the shallowness site aspect. The first step is an orthophoto production. Buoys and scale bar are present on the site in order to permit orientation an model scale. If the orthophoto produced can't replace completely the manual survey made in situe, (this kind of survey is, in the same time, an archaeological interpretation made by the expert in the best situation of observation) the archaeologist can have a lot of benefit using this kind of representation. The orthophoto is a photogrammetric product, easy to obtain once the photogrammetric orientation is done and considering a more or less plane seabed, enough accurate for a first site representation. Moreover, due to the small scale photograph the orthophoto generation carry out a lot of very interesting huge quantities of qualitative detail. Then we will present the way to use this site representation as an interface to the database, managing all the site archaeological data, from the artefact representation and localization to the bibliography and iconography linked. The two dimension of this stratigraphy approach allowed us the use of traditional and free GIS software as GeoTools and SVG GIS package. For each step of the excavation, referring to a different deepness of the same site, we can produce a 2D representation used as a database interface. The next step presented here, yet under development, is the data management in term of coherence, abstract theoretical model and data fusion. The artefact are carried out of the seabed and then analysed in laboratory month after the excavation process. So we have for each object, several information coming from different medium, for example underwater photogrammetry and direct observation in laboratory, we have to develop a strategy to merge these different data often obtained from different accuracy measure system. The last but not the least step of this project is developed with the Comex firm. We plan to develop an integrated photogrammetric tool, using a ROV, digital camera and video control, in order to be able to make this kind of survey completely diver less