SKELETONIZATION OF 3D PLANT POINT CLOUD USING A VOXEL BASED THINNING ALGORITHM

Understanding the point clouds of plants is crucial for the plant phenotyping. However, it is challenging due to a number of factors such as complicated structures and imaging noise. The primary objective of this project is to simplify the complicated 3D structure of the plant point cloud data into 1D curved skeleton. The simplified skeleton will be helpful for the structural analysis and understanding of plants of interest and the measurements of their traits such as the areas, perimeters of leaves, curvatures, and the lengths between different nodes. To this end, we propose a novel method to voxelize the given plant point cloud, extract the skeleton voxels, and find the nearest neighbors to connect the skeleton points as a connected representation. A number of different plant point clouds are used to validate and compare the proposed voxelization thinning method with a state-of-the-art one. Better results have been obtained.

[1]  John P. Lewis,et al.  Pose Space Deformation: A Unified Approach to Shape Interpolation and Skeleton-Driven Deformation , 2000, SIGGRAPH.

[2]  Sunil Arya,et al.  An optimal algorithm for approximate nearest neighbor searching fixed dimensions , 1998, JACM.

[3]  Tiow Seng Tan,et al.  Decomposing polygon meshes for interactive applications , 2001, I3D '01.

[4]  Nico Blodow,et al.  Fast Point Feature Histograms (FPFH) for 3D registration , 2009, 2009 IEEE International Conference on Robotics and Automation.

[5]  Taku Komura,et al.  Topology matching for fully automatic similarity estimation of 3D shapes , 2001, SIGGRAPH.

[6]  Heiner Kuhlmann,et al.  Surface feature based classification of plant organs from 3D laserscanned point clouds for plant phenotyping , 2013, BMC Bioinformatics.

[7]  Tong-Yee Lee,et al.  Skeleton extraction by mesh contraction , 2008, SIGGRAPH 2008.

[8]  Yonghuai Liu,et al.  Accurate Multi-View Stereo 3D Reconstruction for Cost-Effective Plant Phenotyping , 2014, ICIAR.

[9]  Daniel Cohen-Or,et al.  L1-medial skeleton of point cloud , 2013, ACM Trans. Graph..

[10]  Timothy S. George,et al.  Field Phenotyping and Long-Term Platforms to Characterise How Crop Genotypes Interact with Soil Processes and the Environment , 2014 .

[11]  Attila Kuba,et al.  A 3D 6-subiteration thinning algorithm for extracting medial lines , 1998, Pattern Recognit. Lett..

[12]  D. Cohen-Or,et al.  Interactive topology-aware surface reconstruction , 2007, ACM Trans. Graph..