Research on Landscape Environment with 3D-Reconstruction and Volume Measurement of Fruit Tree Canopy Based on Kinect

Abstract

With the demand for precision management of orchards, 3-D reconstruction of fruit tree canopy is receiving more attention. Using depth image data from Kinect sensor, this article attempts to find the 3-D coordinates of the sensed point on the canopy to get the reconstructed fruit tree canopy quickly. By slicing the point clouds to find its surface-line features and the outer envelope within the same slices is used to reconstruct the fruit tree canopy and calculate the canopy’s volume. Tests show that the measuring error for regular cuboid’s volume is about 4.2% and the repeated measuring error for citrus tree’s volume about 6.9%, indicating that applying Kinect for measuring volume of fruit tree canopy has reasonably high accuracy and reliability. The Kinect, as well as this technique, may well be used in landscape measurement on monitering.