An investigation into the factors affecting the accuracy of close range digital photogrammetry for the measurement of historic adobe wall ruins.

摘要

Structures constructed from unbaked earth, of which adobe is one of several forms, are found in abundance throughout the world. Like all structures they are subject to deterioration through time. The main causes of deterioration for adobe structures are the eroding effects of wind, rain, insects and animals. The preservation of historic adobe structures requires careful attention and evaluation of several factors, but the main concern is the loss of material due to erosion. In order to evaluate the condition and calculate the amount of erosion, accurate measurement of the structures is required. Digital photogrammetry was identified as a suitable method of measurement. The development of a theoretical model identified several factors that were hypothesized to have an effect on the accuracy and precision of digital photogrammetry.;An experiment was conducted to investigate how these factors affected the accuracy and precision. The experiment involved measuring a model wall that represented an eroded adobe wall using digital photogrammetry and comparing the measurements with measurements obtained using a measurement method whose accuracy and precision was assumed to be perfect. Multiple regression models were developed to identify which factors affected the accuracy and precision and the magnitude of the effect. Based on the results of the multiple regression models, recommendations were made for field measurement and for further research.