Integration of lidar and airborne imagery for realistic visualization of 3d urban environments

In its report "State of the World Cities 2006/07", the United Nations states that the number of urban area inhabitants is expected to rise to almost 5 billion by 2030, constituting roughly 62 % of the global population. To avoid social and environmental problems arising from this rapid urbanization, federal and local governments must have quick access to accurate and current geo-spatial information to assist in understanding, planning, and managing a continuously changing environment. Therefore, it has been recognized that the research and academic communities should develop practical and accurate tools for assisting urban planners and environmental managers in exploiting geo-spatial data acquired by modern remote sensing systems. This paper presents a framework for integrating photogrammetric and LiDAR data for realistic visualization of 3D urban environments. We will start by illustrating the benefits of integrating photogrammetric and LiDAR data. Then, the paper will proceed by introducing a methodology for the registration of LiDAR and photogrammetric data to a common reference frame. More specifically, conjugate features will be used to geo-reference the imagery relative to the LiDAR reference frame. True orthophoto generation techniques will be incorporated to link the spectral and positional information in the imagery and LiDAR data while circumventing problems encountered when dealing with large scale geo-spatial data over urban environments. Finally, generated orthophotos will be draped over the LiDAR surface model for realistic visualization of an urban environment. Several techniques for refining building boundaries in the LiDAR data and in the orthoimage will be presented for the purpose of improving the quality of the derived visualization