Automatic segmentation of stem and leaf components and individual maize plants in field terrestrial LiDAR data using convolutional neural networks

High-throughput maize phenotyping at both organ and plant levels plays a key role in molecular breeding for increasing crop yields. Although the rapid development of light detection and ranging(Li DAR) provides a new way to characterize three-dimensional(3 D) plant structure, there is a need to develop robust algorithms for extracting 3 D phenotypic traits from Li DAR data to assist in gene identification and selection. Accurate 3 D phenotyping in field environments remains challenging, owing to difficulties in segmentation of organs and individual plants in field terrestrial Li DAR data. We describe a two-stage method that combines both convolutional neural networks(CNNs) and morphological characteristics to segment stems and leaves of individual maize plants in field environments. It initially extracts stem points using the Point CNN model and obtains stem instances by fitting 3 D cylinders to the points. It then segments the field Li DAR point cloud into individual plants using local point densities and 3 D morphological structures of maize plants. The method was tested using 40 samples from field observations and showed high accuracy in the segmentation of both organs(F-score =0.8207) and plants(Fscore =0.9909). The effectiveness of terrestrial Li DAR for phenotyping at organ(including leaf area and stem position) and individual plant(including individual height and crown width) levels in field environments was evaluated. The accuracies of derived stem position(position error =0.0141 m), plant height(R^(2)>0.99), crown width(R^(2)>0.90), and leaf area(R^(2)>0.85) allow investigating plant structural and functional phenotypes in a high-throughput way. This CNN-based solution overcomes the major challenges in organ-level phenotypic trait extraction associated with the organ segmentation, and potentially contributes to studies of plant phenomics and precision agriculture.

Automated extraction of corn leaf points from unorganized terrestrial LiDAR point clouds

Terrestrial LiDAR data can be used to extract accurate structure parameters of corn plant and canopy,such as leaf area,leaf distribution,and 3D model.The first step of these applications is to extract corn leaf points from unorganized LiDAR point clouds.This paper focused on an automated extraction algorithm for identifying the points returning on corn leaf from massive,unorganized LiDAR point clouds.In order to mine the distinct geometry of corn leaves and stalk,the Difference of Normal(DoN)method was proposed to extract corn leaf points.Firstly,the normals of corn leaf surface for all points were estimated on multiple scales.Secondly,the directional ambiguity of the normals was eliminated to obtain the same normal direction for the same leaf distribution.Finally,the DoN was computed and the computed DoN results on the optimal scale were used to extract leave points.The quantitative accuracy assessment showed that the overall accuracy was 94.10%,commission error was 5.89%,and omission error was 18.65%.The results indicate that the proposed method is effective and the corn leaf points can be extracted automatically from massive,unorganized terrestrial LiDAR point clouds using the proposed DoN method.

Effect of layer thickness and voxel size inversion on leaf area density based on the voxel-based canopy profiling method

Voxel-based canopy profiling is commonly used to determine small-scale leaf area.Layer thickness and voxel size impact accuracy when using this method.Here,we determined the optimal combination of layer thickness and voxel size to estimate leaf area density accurately.Terrestrial LiDAR Stonex X300 was used to generate point cloud data for Masson pines(Pinus massoniana).The canopy layer was stratified into 0.10-1.00-m-thick layers,while voxel size was 0.01-0.10 m.The leaf area density of individual trees was estimated using leaf area indices for the upper,middle,and lower canopy and the overall canopy.The true leaf area index,obtained by layered harvesting,was used to verify the inversion results.Leaf area density was inverted by nine combinations of layer thickness and voxel size.The average relative accuracy and mean estimated accuracy of these combined inversion results exceeded 80%.When layer thickness was 1.00 m and voxel size 0.05 m,inversion was closest to the true value.The average relative accuracy was 92.58%,mean estimated accuracy 98.00%,and root mean square error 0.17.The combination of leaf area density and index was accurately retrieved.In conclusion,nondestructive voxel-based canopy profiling proved suitable for inverting the leaf area density of Masson pine in Hetian Town,Fujian Province.