Three-dimensional modeling of site elements by analytically processing image data contained in site photos

Three-dimensional (3D) modeling of site elements is beneficial to many construction management applications such as quantity takeoff and quality control. However, current practice lacks a solution that is computationally simple, cost effective, and easy to use in the field. This research proposes a 3D modeling method based on the mechanism of photogrammetry. It analytically processes the image data contained in site photos at a minimum in terms of image quantities and reference points. A minimum of five pairs of image points of the object from two photos taken at two different camera stations provide sufficient input data to analytically derive the positions and orientations of the camera stations. With more than five paired image points available, the least squares adjustment is utilized to refine the modeling result. The mathematical formulas are established for the proposed method, with the computational procedure and validity illustrated with examples of a simple box, a precast façade, and a residential building. These cases validate the effectiveness and efficiency of the proposed method in modeling construction site elements, particularly for dimensional measurement related applications, and help researchers and practitioners to better understand the impact of the baseline, the overlap percentage, and the quantity of overlapping photos on the outcome of the 3D modeling.