3D reconstruction of Chinese hickory tree for dynamics analysis

Vibratory harvest is the most appropriate and efficient way to harvest Chinese hickory (Carya cathayensis Sarg.) nuts because of the hilly growth environment and scattered shooting architecture. To develop a mechanical harvester that can be adapted to different Chinese hickory trees, a three-dimensional (3D) model of the tree should be obtained and its dynamics parameters identified. This paper introduces an approach to reconstruct a 3D Chinese hickory tree without foliage using two images with different view angles, and then illustrates a dynamics analysis for the reconstructed tree. Two-dimensional (2D) branches were roughly segmented from the image by the “stripe programming” method, and refined by the ribbon snake model. After extracting the skeleton of 2D branches, a binary-directed tree representing 2D tree structure was established. Under the topology and epipolar constraints, the branch correspondences between two images could be easily found. To improve the accuracy, curve reconstruction under the minimum curvature constraint instead of point-based reconstruction was employed. The diameters of branches from the base to the tip were parameterised as a linear equation. Using the 3D model of the Chinese hickory tree, its dynamics analysis was solved by finite element software (ANSYS Workbench). Compared with the results of the impact tests, a predicted optimal vibration frequency of trunk shaker can be found eventually. It indicates that dynamics analysis using 3D model of tree reconstructed by the proposed method is effective.