Design and Test of Algorithms for the Evaluation of Modern Sensors in Close-Range Photogrammetry
Karsten Scheibe · 2008
The thesis describes a general approach for scanning and visualizing panoramic (360°) indoor scenes. It combines range data acquired by a laser rangefinder with color pictures captured by a rotating CCD line camera. The thesis defines coordinate systems of both sensors, specifies the fusion of range and color data acquired by both sensors, and reports about different alternatives for visualizing the generated three-dimensional data sets. Compared to earlier publications, this approach also utilizes an improved method for calculating the spatial (geometric) correspondence between laser diode of the laser rangefinder and the principal point of the rotating CCD line camera. Calibration is also a subject in this thesis. A least-square minimization-based approach is proposed for the rotating CCD line camera. The thesis also deals with particular problems occurring in the process of merging range data and color texture, such as achieving straight edges or planarity in the scene where supported by range or color data. Finally the thesis verifies the possibility to combine terrestrial panoramic images with airborne data. Altogether, the thesis verifies that the described multi sensor approach allows a digitization of indoor or outdoor scenes, illustrates important progress, but also indicates areas of future research.