Distributed processing of Dutch AHN laser altimetry changes of the built-up area

The evolution and spreading of data capturing methods ranging from simple GPS devices like smart-phones to large scale imaging equipment – including very high resolution hyperspectral cameras LiDAR resulted in an exponential growth the amount spatial maintained by companies organizations. At same time for extracting information such are often behind efficiency. In this paper we analyse possibilities nation-wide change detection massive airborne laser altimetry point clouds, based on digital elevation models generated them. proposed workflow distinguishes modifications built-up area other changes noise. Our methodology combines different processing algorithms: object detection, noise filtering, morphological operations clustering. method is designed dynamically extensive datasets a spatially partitioned input dataset easily parallelized manner. Favourable visualizations aggregated representations results examined, followed discussion feasible validation methods. As demonstration showcase implemented distributed evaluation our full Dutch archive exceeding several terabytes storage space using high-performance computing environment. While average execution was 47 h desktop computer, solution only took less than 2.4 complete. output validated against building layer TOP10NL topographic dataset, proving 70% accuracy over 90% urban areas. result analysis shows that Netherlands experienced volume 912.33 km3 between acquisition AHN-2 AHN-3.