摘要：Due to its light weight and high scanning accuracy, LiDAR (Light Detection and Ranging) has been widely used in surveying, unmanned, and instant map construction. In agriculture, LiDAR can directly collect point cloud data of ground crops. After the processing of denoising, interpolation and other algorithms, the 3D digital model of ground crops will be established to be able to timely observe crop growth. Its high degree of automation. This paper first sets up a hardware system for airborne LiDAR, then the principle of point cloud data generation and the derivation of coordinate system transformation are discussed. The various possible errors in the system are analyzed in detail and the simple control error scheme is given then. After that the method of processing 3D point cloud data is expounded and a combined interpolation method based on existing interpolation method is proposed to improve the effect of 3D reconstruction. At last, the experimental verification was carried out to collect point cloud data of a variety of crops. And the 3D digital model of each crop was established using MATLAB.
