Page 217 - Green - Maritime Archaeology: A Technical Handbook. 2nd ed
P. 217

196 Maritime Archaeology: A Technical Handbook, Second Edition
raphy or stereo SPOT imagery as input and produces ortho-rectified imagery, digital terrain models, and contour maps as output. Dynamic 3-D stereo visualization is also possible. Users can view the results of every stage of the photogrammetric process, including raw images, epipolar images in stereo, ortho images, and contours. Mono and stereo perspective views are possible. The software allows interactive editing of matching (x-parallax) results, however, the program is extremely expensive (U.S. $50,000). There are many other systems, that do the same at varying prices. The reader is referred to the listing of commercial terrain visualization software at http://www.tec.army.mil/TD/tvd/survey/survey_toc.html.
A. RESULTS
As mentioned earlier, early phases of survey work used multitape trilat- eration and the measurements were initially processed using the Rule Web program, later the 3H Consulting Ltd Site Surveyor 2 was used because it was more powerful and more flexible.
Control points—the points from which tape measurements were taken— were a complex issue. The points were mainly 10-mm steel rods driven into the rock with a nail attached to the rod to which the end of the tape measure was attached. Additionally, several towers were constructed to provide access to low lying areas on the site. However, these towers were likely to be accidentally caught by airlift hoses or lines and subjected to unusual stresses resulting in occasional position change. The location of the control points therefore needed to be continuously monitored; obviously, if the control point moved slightly, then all the subsequent measurements from that point would reflect this indicating that the point needed to be recalibrated.
Three control points defined the axis of the coordinate system on the site: zero (coordinate 0, 0, 0) and X (3.575, 0, 0.3) and A (0.899, -3.952, -0.061) downward was negative. There were a total of 22 control points on the site and these were surveyed using 225 individual measurements (including a depth measurement at each point). These data were then processed using Site Surveyor. Nine measurements were then rejected because their residuals were too high, and these data were then reprocessed. The resulting data gave an average residual of 6 mm with a root-mean-square (RMS) residual of 9mm.
Once the control points had been surveyed and fixed, excavators then used these points to measure to the objects. Usually, measurements were taken from at least four control pointsand were recorded by the diver and brought to the surface. Matthews then processed these data and the




























































































   215   216   217   218   219