Streaming Extraction of Elevation Contours from LIDAR Points

Air-borne laser range scanning technology (LIDAR) is able to quickly generate massive amounts of densely spaced points that sample the elevation of a terrain. We describe a streaming technique that is able to extract high-resolution iso-contour elevation lines from such large data sets using significantly less memory than the processed input points and the produced output lines. Optional on-the-fly simplification can makes the produced iso-contours more tractable for immediate visualization and smooth out scanning error. By keeping point, mesh, and line data in streaming formats containing “finalization tags,” we connect individual batch operations into a streaming processing pipeline that instantly begins producing final iso-contour output.