Based On LiDAR Ground Rice Biomass And High Precision Of Inversion

With the development of modernization and industrialization,the development of agriculture is inevitably affected.China is a big country with a large population,and the crop output ranks the highest in the world.Therefore,how to grasp the growth of crops and monitor their biomass in real time,so as to provide the basic information for the government,has become one of the research directions.There are many researches on forest biomass at home and abroad,but it is a new research direction to use the ground LiDAR technology to study crop biomass?From the traditional technology to remote sensing technology,and then to the present laser radar technology,people do a lot of research on the scientific monitoring of crop biomass.Laser radar technology has been widely used in military,medical,engineering and forest fields since its development and application.In this paper,we use FARO Focus3 D laser scanner to obtain the point cloud data,which has the advantages of portability,high performance,easy to operate and so on.Three dimensional point cloud data of rice plants can be obtained by using 3D laser scanning technology,and then the point cloud data can be spliced,simplified and denoised.Denoising is the most important in the three processing.According to the characteristics of noise,three methods,radius filtering,statistical filtering and moving least square,are applied to remove noise.And use the F-score method to verify the accuracy of denoising,by statistical filtering in the denoising effect of the most good,the highest precision.In this paper,two parameters of plant height and volume were used to estimate biomass.Strain can be obtained directly in Cloud compare software.The volume is obtained by two methods that 3D dimensional convex hull method and Voxel method respectively,and running the corresponding code in MATLAB can obtain volume.Based on the height biomass model,based on the volume 3D convex hull biomass model,Based on the Voxel biomass model,based on the height and volume of 3D convex hull biomass model and based on the height and volume of Voxel biomass model using SPSS and MATLAB software,a total of 15 regression models were established.The results show that the 3D convex hull volume and height,Voxel volume and heightcomposite regression model established high good decision coefficient is above 0.8.The optimal model is three-dimensional convex hull volume and height of linear equation in two unknowns for a regression model,decision coefficient is reach 0.87.The regression results are ideal.The results showed that the plant height and volume were a feasible method to estimate the biomass of rice.