Field-Based High-Throughput Phenotyping for Maize Plant Using 3D LiDAR Point Cloud Generated With a "Phenomobile" - PubMed Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 May 7:10:554.
doi: 10.3389/fpls.2019.00554. eCollection 2019.

Field-Based High-Throughput Phenotyping for Maize Plant Using 3D LiDAR Point Cloud Generated With a "Phenomobile"

Affiliations

Field-Based High-Throughput Phenotyping for Maize Plant Using 3D LiDAR Point Cloud Generated With a "Phenomobile"

Quan Qiu et al. Front Plant Sci. .

Abstract

With the rapid rising of global population, the demand for improving breeding techniques to greatly increase the worldwide crop production has become more and more urgent. Most researchers believe that the key to new breeding techniques lies in genetic improvement of crops, which leads to a large quantity of phenotyping spots. Unfortunately, current phenotyping solutions are not powerful enough to handle so many spots with satisfying speed and accuracy. As a result, high-throughput phenotyping is drawing more and more attention. In this paper, we propose a new field-based sensing solution to high-throughput phenotyping. We mount a LiDAR (Velodyne HDL64-S3) on a mobile robot, making the robot a "phenomobile." We develop software for data collection and analysis under Robotic Operating System using open source components and algorithm libraries. Different from conducting phenotyping observations with an in-row and one-by-one manner, our new solution allows the robot to move around the parcel to collect data. Thus, the 3D and 360° view laser scanner can collect phenotyping data for a large plant group at the same time, instead of one by one. Furthermore, no touching interference from the robot would be imposed onto the crops. We conduct experiments for maize plant on two parcels. We implement point cloud merging with landmarks and Iterative Closest Points to cut down the time consumption. We then recognize and compute the morphological phenotyping parameters (row spacing and plant height) of maize plant using depth-band histograms and horizontal point density. We analyze the cloud registration and merging performances, the row spacing detection accuracy, and the single plant height computation accuracy. Experimental results verify the feasibility of the proposed solution.

Keywords: 3D LiDAR; field-based; high-throughput phenotyping; maize; mobile robot; point cloud.

PubMed Disclaimer

Figures

FIGURE 1
FIGURE 1
Velodyne HDL64E-S3 on the robot.
FIGURE 2
FIGURE 2
Landmark in the experimental field.
FIGURE 3
FIGURE 3
Two experimental parcels. (A) Is parcel-1; (B) is parcel-2.
FIGURE 4
FIGURE 4
Experimental layouts of parcel-1 and parcel-2. (A) Is parcel-1; (B) is parcel-2.
FIGURE 5
FIGURE 5
Flowchart for the general framework of data processing.
FIGURE 6
FIGURE 6
Different point types in DBSCAN. Because p and q both have more than five neighbors within E circle adjacent area, they are Core Points; b has less than five neighbors, but it lies in the adjacent area of q, so b is a Border Point; n is a Noise Point.
FIGURE 7
FIGURE 7
Flowchart for DBSCAN and Octree based landmark detection.
FIGURE 8
FIGURE 8
Flowchart for virtual-landmark-based registration.
FIGURE 9
FIGURE 9
Flowchart for row spacing computation.
FIGURE 10
FIGURE 10
Flowchart for plant height computation.
FIGURE 11
FIGURE 11
Locations of all manually measured plant height samples.
FIGURE 12
FIGURE 12
Manual sampling method for row spacing. The dash lines are two adjacent local row-lines.
FIGURE 13
FIGURE 13
Manual sampling method for plant height.
FIGURE 14
FIGURE 14
Landmark detection result for plot 2 in Parcel-1.
FIGURE 15
FIGURE 15
Merged clouds for three different steps. (A) Is the merged cloud for “step = 1”; (B) is the merged cloud for “step = 2”; (C) is the merged cloud for “step = 3.”
FIGURE 16
FIGURE 16
Cloud registration and merging process. (A) Shows the unregistered two point clouds; (C) shows the landmark points and noise points after cutting off most plant points; (D) shows the virtual landmarks generated after DBSCAN; (E) shows the SAC-IA registration result; (F) shows the IPC registration result; (G) shows the merged two clouds; (B) and (H) show the magnified views of the blue rectangle areas in (A) and (G), respectively.
FIGURE 17
FIGURE 17
Row detection result for Parcel-2.
FIGURE 18
FIGURE 18
Data analysis for manually sampled and calculated results of row spacing. (A) Is the regression line, the R square, and the RMSE for all the 57 row spacing values; (B) is the regression line, the R square, and the RMSE for the row spacing values in Area A; (C) is the regression line, the R square, and the RMSE for the row spacing values in Area B; (D) is the regression line, the R square, and the RMSE for the row spacing values in Area C.
FIGURE 19
FIGURE 19
Single plant detection for parcel-1. The detected single plant cloud is in red and marked out with a green bounding box.

Similar articles

Cited by

References

    1. Andrade F. H., Calviño P., Cirilo A., Barbieri P. (2002). Yield responses to narrow rows depend on increased radiation interception. Agron. J. 94 975–980.
    1. Andrade-Sanchez P., Gore M. A., Heun J. T., Thorp K. R., Carmo-Silva A. E., French A. N., et al. (2014). Development and evaluation of a field-based high-throughput phenotyping platform. Funct. Plant Biol. 41 68–79. 10.3389/fpls.2018.01362 - DOI - PubMed
    1. Andújar D., Ribeiro A., Fernández-Quintanilla C., Dorado J. (2016). Using depth cameras to extract structural parameters to assess the growth state and yield of cauliflower crops. Comput. Electron. Agric. 122 67–73. 10.1016/j.compag.2016.01.018 - DOI
    1. Araus J. L., Cairns J. E. (2014). Field high-throughput phenotyping: the new crop breeding frontier. Trends Plant Sci. 19 52–61. 10.1016/j.tplants.2013.09.008 - DOI - PubMed
    1. Asaari M. S. M., Mishra P., Mertens S., Dhondt S., Inzé D., Wuyts N., et al. (2018). Close-range hyperspectral image analysis for the early detection of stress responses in individual plants in a high-throughput phenotyping platform. ISPRS J. Photogramm. Remote Sens. 138 121–138. 10.1016/j.isprsjprs.2018.02.003 - DOI