Reconstruction of tree structure from multi-scale measurement data

Two methods are proposed for three-dimensional (3D) reconstruction of tree structure from multi-scale measurement data. The proposed reconstruction methods include reconstruction from digitized characteristic points of the tree and reconstruction from scanned point cloud of a whole tree. Both methods used real measurement data as such multi-scale data needed to be acquired from the targeted tree. At organ scale, laser scanner was used to obtain point cloud data from the surface of an organ. At individual tree scale, 3D digitizer and large-scale 3D scanner was used to measure the data of a whole tree. While at colony scale, several morphological characteristics were measured from the plant colony. Methods for reconstructing the 3D structure of a tree from these multi-scale measuring data were described in detail. A quantitative evaluation combing reconstructed accuracy and time consuming for data measurement was given to compare different reconstruction strategies.