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
. 2023 Jun 26:14:1146490.
doi: 10.3389/fpls.2023.1146490. eCollection 2023.

A fast phenotype approach of 3D point clouds of Pinus massoniana seedlings

Honghao Zhou  1   2 , Yang Zhou  1   2 , Wei Long  3 , Bin Wang  3 , Zhichun Zhou  3 , Yue Chen  4
Affiliations

A fast phenotype approach of 3D point clouds of Pinus massoniana seedlings

Honghao Zhou et al. Front Plant Sci. .

Abstract

The phenotyping of Pinus massoniana seedlings is essential for breeding, vegetation protection, resource investigation, and so on. Few reports regarding estimating phenotypic parameters accurately in the seeding stage of Pinus massoniana plants using 3D point clouds exist. In this study, seedlings with heights of approximately 15-30 cm were taken as the research object, and an improved approach was proposed to automatically calculate five key parameters. The key procedure of our proposed method includes point cloud preprocessing, stem and leaf segmentation, and morphological trait extraction steps. In the skeletonization step, the cloud points were sliced in vertical and horizontal directions, gray value clustering was performed, the centroid of the slice was regarded as the skeleton point, and the alternative skeleton point of the main stem was determined by the DAG single source shortest path algorithm. Then, the skeleton points of the canopy in the alternative skeleton point were removed, and the skeleton point of the main stem was obtained. Last, the main stem skeleton point after linear interpolation was restored, while stem and leaf segmentation was achieved. Because of the leaf morphological characteristics of Pinus massoniana, its leaves are large and dense. Even using a high-precision industrial digital readout, it is impossible to obtain a 3D model of Pinus massoniana leaves. In this study, an improved algorithm based on density and projection is proposed to estimate the relevant parameters of Pinus massoniana leaves. Finally, five important phenotypic parameters, namely plant height, stem diameter, main stem length, regional leaf length, and total leaf number, are obtained from the skeleton and the point cloud after separation and reconstruction. The experimental results showed that there was a high correlation between the actual value from manual measurement and the predicted value from the algorithm output. The accuracies of the main stem diameter, main stem length, and leaf length were 93.5%, 95.7%, and 83.8%, respectively, which meet the requirements of real applications.

Keywords: 3D point cloud; Pinus massoniana seedlings; phenotyping; skeletonization; slicing.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Nondestructive 3D image acquisition setup. (A) Curtain. (B) Plant placement. (C) Experimental platform. (D) Rotary table. (E) Yunteng691 bracket. (F) Aruze Kinect camera. (G) Computer.
Figure 2
Figure 2
Image of point cloud preprocessing. (A) Real image. (B) Depth image. (C) 3D point cloud image. (D) Background removed point cloud image. (E) Experimental platform removed point cloud image; parts of discrete points are marked by red circles. (F) Discrete points removed point cloud image. (G) Soil detect image; point clouds in the soil are marked in red. (H) Registered point cloud image of 0 degrees. (I) Registered point cloud image of 180 degrees.
Figure 3
Figure 3
Image of the stem and leaf segmentation. (A) Point cloud slices along the Z-axis and X-axis with adjacent slice layers distinguished by different colors. (B) Diagram of the relationship between the skeleton points. The green point, red point, and blue point represent the junction, vertex, and internal node, respectively. (C) Skeleton points of the main stem with the canopy; the red circle represents the canopy. (D) The green line represents a normal composed of a fifth skeleton point and the adjacent skeleton points above it. The gray normal represents a projection plane perpendicular to the normal. (E) The red part represents the projection of the top canopy on the tangent plane. (F) Skeleton point of the main stem after removing the canopy. (G) Main stem skeleton points before interpolation. (H) Main stem skeleton points after interpolation. (I) Point cloud after removing the main stem.
Figure 4
Figure 4
Algorithm for skeletonization.
Figure 5
Figure 5
Algorithm for main stem skeleton point extraction.
Figure 6
Figure 6
Algorithm for main stem point cloud restoration.
Figure 7
Figure 7
Image of the morphological traits extraction. (A) Upper plane. (B) Lower plane. (C) 2D projection of the main stem point cloud, short-axis of the ellipse as the stem diameter. (D) Main stem skeleton points for calculating the main stem length. (E) 2D points projected onto the lower plane. (F) Convex boundaries and links between the boundaries and center point.
Figure 8
Figure 8
Stem and leaf segmentation procedure visualization of different heights. (A) Input cloud point. (B) Slice in the X-axis and Z-axis. (C) Extracting alternative skeleton points of the main stem according to the centroid and generating MST according to the skeleton points. (D) DAG longest path algorithm searches for the skeleton points of the main stem with a canopy. (E) Main stem skeleton points after removing the canopy. (F) Main stem skeleton points after interpolation expansion. (G) After removing the main stem point cloud, the plant point cloud only contains leaves.
Figure 9
Figure 9
Comparison of the six phenotypic parameters system estimates and manual measurements. (A–H) line of regression represents the straight line fitted by the estimated value. r = 1 represents estimated value is equal to actual value and regards as a reference line.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Besl P. J., Jain R. C. (1988). Segmentation through variable-order surface fitting. IEEE Trans. Pattern Anal. Mach. Intell. 10 (2), 167–192. doi: 10.1109/34.3881 - DOI
    1. Bradski G., Kaehler A. (2008). Learning OpenCV: computer vision with the OpenCV library (O'Reilly Media, Inc; ). https://ieeexplore.ieee.org/document/5233425/
    1. Cao J., Tagliasacchi A., Olson M., Zhang H., Su Z. (2010). “Point cloud skeletons via laplacian based contraction,” in 2010 Shape Modeling International Conference. 187–197. doi: 10.1109/SMI.2010.25 - DOI
    1. Cupec R., Vidović I., Filko D., Đurović P. (2020). Object recognition based on convex hull alignment. Pattern Recognit. 102, 107199. doi: 10.1016/j.patcog.2020.107199 - DOI
    1. Dalitz C., Schramke T., Jeltsch M. (2017). Iterative hough transform for line detection in 3D point clouds. Image Process. On Line 7, 184–196. doi: 10.5201/ipol.2017.208 - DOI

LinkOut - more resources