Journal of System Simulation ›› 2017, Vol. 29 ›› Issue (3): 524-530.doi: 10.16182/j.issn1004731x.joss.201703008

Previous Articles     Next Articles

Skeleton Extraction from Three-dimensional Maize Based on Kinect

Fang Zhili1,2, Wen Weiliang1, Guo Xinyu1, Wu Jianwei3   

  1. 1. Beijing Research Center for Information Technology in Agriculture/ Beijing Key Lab of Digital Plant/ National Engineering Research Center for Information Technology in Agriculture/ Key Laboratory of Agri-information Ministry of Agriculture, P R China/ Beijing Agriculture Internet of Things Engineering Technology Research Center, Beijing 100097, China;
    2. Shanghai Ocean University Information College,Shanghai 201306, China;
    3. Beijing PAIDE Science and Technology Development Co. Ltd., Beijing 100097, China
  • Received:2015-06-01 Revised:2015-08-23 Online:2017-03-08 Published:2020-06-02

Abstract: Aiming at acquiring three-dimensional skeleton of maize plant efficiently, an automatic three-dimensional maize plant skeleton extraction method based on Kinect data was proposed. Kinect data was transformed into point cloud data. Backgroundpoint cloud was deleted by threshold segmentation and the point cloud was stratified according to the height of the 3D model. Each layer was clustered by two circle iteration and the center points of each cluster could be obtained. The set of center points could be considered as the skeleton points of maize. The quadratic B-spline interpolation curve was used to obtain relatively accurate and smooth linear skeleton of maize plant. Experimental result shows that the 3D maize plant linear skeleton obtained by this method has a good consistency of 3D digitizer skeleton and can meet the requirements of rapid experimental data acquirement.

Key words: Kinect, three-dimensional point cloud, maizeskeleton, cluster, quadratic B-spline

CLC Number: