3D Visualization for Long Pavement Surfaces with 1-mm Laser Imaging Data

With the recent development of advanced laser imaging technology, 3D pavement surface data can be collected at highway speed up to 60mph at 1-mm resolution. There is a need for pavement engineers and researchers to visualize the 3D data sets and understand the extent and level of pavement deterioration. To visualize such large amount of pavement data sets requires rendering of many gigabytes of raw data in a short period of time, such as less than 1 second. It is a challenging task and related research is lacking. In this paper, the geometry clipmap based algorithm, a 3D level-of-detail (LOD) technique, is implemented to visualize long 3D pavement surfaces with 1-mm data. This algorithm caches pavement height data in a set of nested grids to render long pavement surfaces, and provides visual continuity and uniform frames rate for steady rendering. As a result, the data size uploaded to video memory is reduced dramatically, while the detailed high-resolution 1mm pavement surface data remains. With a high performance graphic card, GPU based technique is subsequently implemented in the geometry clipmap framework to dramatically reduce the offloading work of CPU. In addition, the vertex normal calculation based on OpenGL Shading Language (GLSL) is applied in this paper to further decrease the data size required for 3D rendering. Finally, the visualization results and statistics of frames per second (FPS) are provided to demonstrate the visual performance and efficiency of the geometry clipmap based algorithm for long pavement visualization. This research can aid decision makers to better visualize pavement and airport runway surface to identify surface deterioration, check data continuity, assess data quality, and prepare data for pavement condition evaluation.