for the field management and freezing damage ratings at site FIP (all persons ETH Zurich, Zürich, Switzerland). We thank the anonymous reviewers for the thorough evaluation and constructive suggestions. Additional Points Source Code . Maintained source code for processing is publicly available on the ETH Zurich GitLab server ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools ); archived versions can be found in Roth [ 62 , 63 ]. The GitLab repository includes following preprocessing steps: (1) Image mask generation (Standalone Agisoft Metashape Script) ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/ImageProjectionAgisoft ). (2) Image segmentati
Open resource ↗PhenoFly_data_processing_tools · lines:218-251Unverified paper record
Repeated Multiview Imaging for Estimating Seedling Tiller Counts of Wheat Genotypes Using Drones
Plant Phenomics · 7 Sept 2020 · 10.34133/2020/3729715
Abstract
Early generation breeding nurseries with thousands of genotypes in single-row plots are well suited to capitalize on high throughput phenotyping. Nevertheless, methods to monitor the intrinsically hard-to-phenotype early development of wheat are yet rare. We aimed to develop proxy measures for the rate of plant emergence, the number of tillers, and the beginning of stem elongation using drone-based imagery. We used RGB images (ground sampling distance of 3 mm pixel -1 ) acquired by repeated flights (≥ 2 flights per week) to quantify temporal changes of visible leaf area. To exploit the information contained in the multitude of viewing angles within the RGB images, we processed them to multiview ground cover images showing plant pixel fractions. Based on these images, we trained a support vector machine for the beginning of stem elongation (GS30). Using the GS30 as key point, we subsequently extracted plant and tiller counts using a watershed algorithm and growth modeling, respectively. Our results show that determination coefficients of predictions are moderate for plant count ( R 2 = 0.52), but strong for tiller count ( R 2 = 0.86) and GS30 ( R 2 = 0.77). Heritabilities are superior to manual measurements for plant count and tiller count, but inferior for GS30 measurements. Increasing the selection intensity due to throughput may overcome this limitation. Multiview image traits can replace hand measurements with high efficiency (85-223%). We therefore conclude that multiview images have a high potential to become a standard tool in plant phenomics.
Plant phenotyping relevance
ドローン多視点画像と画像解析により、コムギの出芽・分げつ数・茎伸長を推定する手法を開発・評価しており、植物表現型取得が研究の中心である。
abstractWe aimed to develop proxy measures for the rate of plant emergence, the number of tillers, and the beginning of stem elongation using drone-based imagery.
abstractBased on these images, we trained a support vector machine for the beginning of stem elongation (GS30). Using the GS30 as key point, we subsequently extracted plant and tiller counts using a watershed algorithm and growth modeling, respectively.
abstractWe therefore conclude that multiview images have a high potential to become a standard tool in plant phenomics.
Code and data availability
The paper publicly releases its authors' phenotyping processing code (multiview image generation, segmentation, early growth trait extraction) on ETH GitLab and secondary plot-based phenotype data (BLUEs, BLUPs, repeatability, heritability) on the ETH Research Collection. Raw UAS images are only available upon request,
code for processing is publicly available on the ETH Zurich GitLab server ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools ); archived versions can be found in Roth [ 62 , 63 ]. The GitLab repository includes following preprocessing steps: (1) Image mask generation (Standalone Agisoft Metashape Script) ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/ImageProjectionAgisoft ). (2) Image segmentation with random forest ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/ActiveLearningSegmentation ). (3) Multiview image generation ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/MultiViewImage ). The Gi
Open resource ↗PhenoFly_data_processing_tools/ImageProjectionAgisoft · lines:218-251tools ); archived versions can be found in Roth [ 62 , 63 ]. The GitLab repository includes following preprocessing steps: (1) Image mask generation (Standalone Agisoft Metashape Script) ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/ImageProjectionAgisoft ). (2) Image segmentation with random forest ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/ActiveLearningSegmentation ). (3) Multiview image generation ( https://gitlab.ethz.ch/crop_phenotyping/PhenoFly_data_processing_tools/MultiViewImage ). The GitLab repository furthermore includes the following trait extraction method: (4) Early growth trait extraction ( https://gitlab.ethz.ch/crop_p
Open resource ↗PhenoFly_data_processing_tools/ActiveLearningSegmentation · lines:218-251This is an automatically classified, unverified record. Curator approval is required before any resource enters the Catalog.