Boukhana et al., “Geometric models for plant leaf area estimation from 3D pointclouds: a comparative study”, GVC, 2022.

place_holder

Abstract

Measuring leaf area is a critical task in plant biology. Meshing techniques, parametric surface modelling and implicit surface modelling allow estimating plant leaf area from acquired 3D point clouds. However, there is currently no consensus on the best approach because of little comparative evaluation. In this paper, we provide evidence about the performance of each approach, through a comparative study of four meshing, three parametric modelling and one implicit modelling methods. All selected methods are freely available and easy to use. We have also performed a parameter sensitivity analysis for each method in order to optimise its results and fully automate its use. We identified nine criteria affecting the robustness of the studied methods. These criteria are related to either the leaf shape (length/width ratio, curviness, concavity) or the acquisition process (e.g. sampling density, noise, misalignment, holes). We used synthetic data to quantitatively evaluate the robustness of the selected approaches with respect to each criterion. In addition we evaluated the results of these approaches on five tree and crop datasets acquired with laser scanners or photogrammetry. This study allows us to highlight the benefits and drawbacks of each method and evaluate its appropriateness in a given scenario. Our main conclusion is that fitting a B'ezier surface is the most robust and accurate approach to estimate plant leaf area in most cases.

Overview

In this paper, we propose a thorough comparative study of eight leaf reconstruction method from point clouds selected from the three main approaches: (1) mesh reconstruction, (2) explicit modelling, (3) implicit modelling. Our aim is to help plant biologists to choose the most suitable method during leaf area estimation from 3D point clouds. As a consequence, we have selected methods that are easily accessible to the practitioners, being freely available in open-source software or libraries. These methods and the corresponding surface models, are detailed in the article. We evaluated the selected methods on both synthetic and real datasets and compared their results.

2.5D $\alpha$ shape BPA Incrmental B-Spline Bézier Trimmed Poisson
2.5D point cloud and mesh Alpha shape Ball pivoting algorithm Incremental reconstruction B-Spline Bézier leaf model Trimmed Bézier Poisson
2.5D point cloud and mesh Alpha shape Ball pivoting algorithm Incremental reconstruction B-Spline Bézier leaf model Trimmed Bézier Poisson
2.5D $\alpha$ shape BPA Incrmental B-Spline Bézier Trimmed Poisson
2.5D point cloud and mesh Alpha shape Ball pivoting algorithm Incremental reconstruction B-Spline Bézier leaf model Trimmed Bézier Poisson
2.5D point cloud and mesh Alpha shape Ball pivoting algorithm Incremental reconstruction B-Spline Bézier leaf model Trimmed Bézier Poisson

Software released along with the article

Licence

Creative Commons License
The software released along with the article are licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. For legal details, the code of this licence is detailled at: https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode

Source code

The source code released along with the article is available at https://gitlab.unistra.fr/plant-leaf-area-estimation-gvc-2022/surface-models-source-code

Citation

If you use our work or use it for comparison, please cite our article:

@inproceedings
{
    boukhana2021,
}

Creative Commons License
The Plant leaf area estimation website by Joris Ravaglia is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.