D. Westermann
(Kuppenheim, Germany)

In the last few years digital photography has been increasingly used on excavations. This has opened opportunities for the use of simple photo-rectification software to quickly create seemingly accurate plans and sections on site. However, on excavation, this supposed accuracy has often been uncritically accepted without testing the programs available against each other, looking, for example at accuracy of the results and ease of use.
This paper examines the principles behind the rectification of digital photos and reports the results of a series of comparative, analytical tests on the programmes “PhoToPlan” and various Auto Desk products.
The tests were carried out both in a laboratory environment and in the field and involved measuring the attainable accuracy for both pre-drawn grids and 3-dimentional real objects on excavations. The results demonstrate the possibilities and limitations of rectified digital photographs on excavations and give rise to interesting questions to the excavator as to the amount of accuracy he requires and the amount of time he is prepared to spend on getting it. Although it is true that no specialist a priori knowledge is required, as for example for laser scanning. It became clear during this work, that a certain amount of understanding of photographic principles and the algorithms involved in photorectification is necessary if adequate results are to be achieved.
The basis of this paper is the diploma thesis “Fotogrammetrie und digitale Fotografie in der grabungstechnischen Anwendung” presented at the Fachschule für Technik und Wirtschaft, Berlin (University of Applied Sciences).