2007 · 91 citations · 51 references
ComputingGeoinformaticsDigital Elevation ModelingGeometric ModelingRemote Sensing3D ReconstructionGeometry ProcessingPhotogrammetryPoint Cloud ProcessingSurface ReconstructionTerrain SegmentationActive MeasurementAirborne Laser ScanningDigital Terrain Model SegmentationGeometric ReconstructionLaser ScanningObject Surface ReconstructionShapeTerrestrial Laser Scanning
This paper reviews the current state of laser scanning from airborne and terrestrial platforms for geometric reconstruction of object shape and size. The current performance figures of sensor systems are presented in an overview. Next, their calibration and the orientation of the acquired point clouds is discussed. For airborne deployment this is usually one step, whereas in the terrestrial case laboratory calibration and registration of point clouds are (still) two distinct, independent steps. As laser scanning is an active measurement technology, the interaction of the emitted energy with the object surface has influences on the range measurement. This has to be considered in order to explain geometric phenomena in the data. While the problems, e.g. multiple scattering, are understood well, there is currently a lack of remedies. Then, in analogy to the processing chain, segmentation approaches for laser scanning data are reviewed. Segmentation is a task relevant for almost all applications. Likewise, DTM (digital terrain model) reconstruction is relevant for many applications of airborne laser scanning, and is therefore discussed, too. This paper reviews the main processing steps necessary for many applications of laser scanning.
51
Martin A. Fischler, Robert C. Bolles · Communications of the ACM · 1981 · 25.5K citations · Full text
Efficient variants of the ICP algorithm
Szymon Rusinkiewicz, Marc Levoy · 2002 · 3.7K citations