Texture mapping and true orthophoto generation of 3D objects

Abstract Photogrammetry provides, in contrast to other surveying me thods, a complete coverage of the object or region by image data. For a long time, the main purpo se f surveys was to extract geometrical information such as elevation models, distanc es, or point coordinates from images. Since the first computers became readily available, more pro spects have opened to use this data and the photorealistic visualization of the acquired side c ame into clearer focus. This application was unique and proved a big advantage for photogrammetry in c omparison with other surveying techniques. Over many years, the representation of 2.5D dat a, such as digital elevation models derived from aerial images, was the development goal. With t he improvement of the computer systems and the available algorithms, photogrammetry pave d th way to acquire 3D objects in aerial photogrammetry and close range applications.Photogrammetry provides, in contrast to other surveying me thods, a complete coverage of the object or region by image data. For a long time, the main purpo se f surveys was to extract geometrical information such as elevation models, distanc es, or point coordinates from images. Since the first computers became readily available, more pro spects have opened to use this data and the photorealistic visualization of the acquired side c ame into clearer focus. This application was unique and proved a big advantage for photogrammetry in c omparison with other surveying techniques. Over many years, the representation of 2.5D dat a, such as digital elevation models derived from aerial images, was the development goal. With t he improvement of the computer systems and the available algorithms, photogrammetry pave d th way to acquire 3D objects in aerial photogrammetry and close range applications. The change from 2.5D to 3D lead to an enormous change in the req uirements concerning the algorithms used. Due to the much higher complexity in handli ng 3D data as compared to 2.5D data, a large number of existing approaches and algorithms w ere no longer usable. This work was motivated by the need to develop new algorithms to visualize 3D data in combination with the acquired image data. The procedure to do this is called textu r mapping. The basic idea behind texture mapping is to attach the image information onto the g eometrical data. The result is a photorealistic visualization of the acquired object. Even at the time of the start of this work, a wide range of algorithms were available to perform texture mappi ng. The problem was that the existing algorithms were developed with the focus on a different data configuration and requirement. Most of the existing texture mapping algorithms, that could hand le 3D data were developed for computer vision applications, e.g. real time visualizations. In tho se applications, the available data is quite different than the data we are focused to handle in this work. For example, one difference is the number of available images. In our applications, an obje ct should be covered with a usefull number of images in a high quality. In contrast, most of the ex isting approaches use data with low resolution images, acquired using video devices. This w ork is aimed at analyzing existing approaches concerning their usability to process photogra mmetric datasets. Even in the best cases, most of the algorithms had to be adapted to the special needs o f the photogrammetric specifications and a number of new algorithms had to be developed for necessa ry processing steps. This thesis presents algorithms and software modules cover ing the workflow for texture mapping from the point of provided images and surface models to the ph oto-realistic textured 3D model. The completed work results in a wide variety of algorithms of whi ch there are two types of algorithms distinguished: The first group of algorithms handles geometric issues: • A vector algebra-based visibility analysis to eliminate th disadvantages of exiting approaches like z-buffer or ray-tracing, and • a multi-image texture mapping to determine the best possibl e texture source for every surface element.