This post is concerned with the photogrammetric survey strand of survey that we undertook (see first post) as part of the Altogether Archaeology project at Long Meg and Her Daughters stone circle near Penrith, in the Eden Valley, Cumbria.
In this instance we undertook to photograph 33 of the 69 individual stones consisting the stone circle, and including the Long Meg monolith itself. This process was undertaken by the Altogether Archaeology volunteers who photographed each individual stone from the ground, and by taking shots from above using a photographic mast. The intention was to take photographs from all around the stones so as to provide complete overlapping coverage of each individual stone. The background detail needed to be removed from each photograph and this was achieved both by masking it out on the computer and by having a white background held behind the individual stones during photography.
Altogether Archaeology volunteers were instrumental in going through the various processes using Agisoft Photoscan Pro software to produce nine 3D models of individual stones from the circle. The software combined the photographs, computed the positions that the images were taken from, generated a 3D mesh, and then the photographs were reapplied to the 3D mesh as a texture. Each individual photograph needed to be prepared, and this entailed masking out of any background detail still left in the photographs, and then the application of survey control. The end product was a series of 3D models that can be viewed as 2D plans or within Adobe Acrobat as 3D pdf files.
The photographs of the Long Meg red sandstone monolith came out particularly well as the bright sunlight slanted across the face of the stone containing the rock art motifs and gave sharp definition to the rock strata and any depressions in the surface.
Again many thanks must go to all of the Altogether Archaeology volunteers who braved the weather to take the photography on site, and to those luckier souls who were based in the nearby village hall processing the data.