Numerous fields of application effort the detection of pose changesfor 3 dimensional objects in six degrees of freedom (6 DoF). Automaticprocedures that exploit 2D images for the detection of pose changescan be used for example for tracking object movements, for qualitycontrol or for the verification of the alignment of patients in radiationtreatment devices. In this contribution we present two differentsolutions for the detection of pose changes that base on the comparisonof two 2D images resulting from the projection of an object in thenew pose and a 3D volume of the same object in a known referencealignment. Whereas for the first solution we use an object wherewe can clearly extract landmarks useable as reference positions forthe determination of the object’s alignment, we provide a secondsolution for objects where these landmarks cannot be extracted, whichis involved automatically if necessary. In this case grey value basedpose estimation is conducted by registering the computationally projectedreference 3D volume to the 2D images. As reference data for the objectwith known alignment, CT slices will The study highlights a new method for the delineation of tree crownsand the detection of stem positions of single trees from dense airborne LIDAR data. At first, we combine a method for surface reconstruction,which robustly interpolates the canopy height model (CHM) from the LIDAR data, with a watershed algorithm. Stem positionsof the tallest trees in the tree segments are subsequently derived from the local maxima of the CHM. Additional stem positionsin the segments are detected in a 3-step algorithm. First, all the points between the ground and the crown base height are separated.Second, possible stem points are found by hierarchicallyclustering these points. Third, the stem is reconstructed with a robustRANSAC-based adjustment of the stem points. The method was applied to small-footprint full waveform data, which have a pointdensity of 25 points per m². The detection rate for coniferous trees is 61 % and for deciduous trees 44 %, respectively. 7 % of thedetected trees are false positives. The mean positioning error is0.92 cm, whereas the additional stem detection improves the tree positionon average by 22 cm. The analysis of waveform data inthe tree structure shows that the intensity and pulse width discriminatestem points, crown points and ground points significantly. Moreover, the mean intensity of stem points turned out to be the mostsalient feature for the discrimination of coniferous and deciduous trees.e used, as they are providedfor the alignment of patients in radiation treatment devices. TwoX-ray images of the same object in an unknown pose can then be comparedto the reference data to determine the respective pose change, whichmay consist of 3 rotations and 3 translations. Using both approachesto determine patient misalignments in treatment devices shows, thatboth methods result in highly accurate pose detections and that thesecond method, despite being less accurate and more time consuming,is an appropriate solution in cases where landmark detection fails.     «

Numerous fields of application effort the detection of pose changesfor 3 dimensional objects in six degrees of freedom (6 DoF). Automaticprocedures that exploit 2D images for the detection of pose changescan be used for example for tracking object movements, for qualitycontrol or for the verification of the alignment of patients in radiationtreatment devices. In this contribution we present two differentsolutions for the detection of pose changes that base on the comparisonof two 2D images resul...     »