Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset
Robust and automated segmentation of leaves and other backgrounds is a core prerequisite of most approaches in high-throughput field phenotyping. So far, the possibilities of deep learning approaches for this purpose have not been explored adequately, partly due to a lack of publicly available, appr...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2022-01-01
|
Series: | Frontiers in Plant Science |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2021.774068/full |
_version_ | 1798013999922544640 |
---|---|
author | Radek Zenkl Radu Timofte Norbert Kirchgessner Lukas Roth Andreas Hund Luc Van Gool Achim Walter Helge Aasen |
author_facet | Radek Zenkl Radu Timofte Norbert Kirchgessner Lukas Roth Andreas Hund Luc Van Gool Achim Walter Helge Aasen |
author_sort | Radek Zenkl |
collection | DOAJ |
description | Robust and automated segmentation of leaves and other backgrounds is a core prerequisite of most approaches in high-throughput field phenotyping. So far, the possibilities of deep learning approaches for this purpose have not been explored adequately, partly due to a lack of publicly available, appropriate datasets. This study presents a workflow based on DeepLab v3+ and on a diverse annotated dataset of 190 RGB (350 x 350 pixels) images. Images of winter wheat plants of 76 different genotypes and developmental stages have been acquired throughout multiple years at high resolution in outdoor conditions using nadir view, encompassing a wide range of imaging conditions. Inconsistencies of human annotators in complex images have been quantified, and metadata information of camera settings has been included. The proposed approach achieves an intersection over union (IoU) of 0.77 and 0.90 for plants and soil, respectively. This outperforms the benchmarked machine learning methods which use Support Vector Classifier and/or Random Forrest. The results show that a small but carefully chosen and annotated set of images can provide a good basis for a powerful segmentation pipeline. Compared to earlier methods based on machine learning, the proposed method achieves better performance on the selected dataset in spite of using a deep learning approach with limited data. Increasing the amount of publicly available data with high human agreement on annotations and further development of deep neural network architectures will provide high potential for robust field-based plant segmentation in the near future. This, in turn, will be a cornerstone of data-driven improvement in crop breeding and agricultural practices of global benefit. |
first_indexed | 2024-04-11T15:10:14Z |
format | Article |
id | doaj.art-44196e1a9887437c94a170bb7ea74547 |
institution | Directory Open Access Journal |
issn | 1664-462X |
language | English |
last_indexed | 2024-04-11T15:10:14Z |
publishDate | 2022-01-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Plant Science |
spelling | doaj.art-44196e1a9887437c94a170bb7ea745472022-12-22T04:16:39ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2022-01-011210.3389/fpls.2021.774068774068Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat DatasetRadek Zenkl0Radu Timofte1Norbert Kirchgessner2Lukas Roth3Andreas Hund4Luc Van Gool5Achim Walter6Helge Aasen7Group of Crop Science, Department of Environmental Systems Science, Institute of Agricultural Sciences, ETH Zurich, Zurich, SwitzerlandComputer Vision Lab, Department of Information Technology and Electrical Engineering, ETH Zurich, Zurich, SwitzerlandGroup of Crop Science, Department of Environmental Systems Science, Institute of Agricultural Sciences, ETH Zurich, Zurich, SwitzerlandGroup of Crop Science, Department of Environmental Systems Science, Institute of Agricultural Sciences, ETH Zurich, Zurich, SwitzerlandGroup of Crop Science, Department of Environmental Systems Science, Institute of Agricultural Sciences, ETH Zurich, Zurich, SwitzerlandComputer Vision Lab, Department of Information Technology and Electrical Engineering, ETH Zurich, Zurich, SwitzerlandGroup of Crop Science, Department of Environmental Systems Science, Institute of Agricultural Sciences, ETH Zurich, Zurich, SwitzerlandRemote Sensing Team, Division of Agroecology and Environment, Agroscope, Zurich, SwitzerlandRobust and automated segmentation of leaves and other backgrounds is a core prerequisite of most approaches in high-throughput field phenotyping. So far, the possibilities of deep learning approaches for this purpose have not been explored adequately, partly due to a lack of publicly available, appropriate datasets. This study presents a workflow based on DeepLab v3+ and on a diverse annotated dataset of 190 RGB (350 x 350 pixels) images. Images of winter wheat plants of 76 different genotypes and developmental stages have been acquired throughout multiple years at high resolution in outdoor conditions using nadir view, encompassing a wide range of imaging conditions. Inconsistencies of human annotators in complex images have been quantified, and metadata information of camera settings has been included. The proposed approach achieves an intersection over union (IoU) of 0.77 and 0.90 for plants and soil, respectively. This outperforms the benchmarked machine learning methods which use Support Vector Classifier and/or Random Forrest. The results show that a small but carefully chosen and annotated set of images can provide a good basis for a powerful segmentation pipeline. Compared to earlier methods based on machine learning, the proposed method achieves better performance on the selected dataset in spite of using a deep learning approach with limited data. Increasing the amount of publicly available data with high human agreement on annotations and further development of deep neural network architectures will provide high potential for robust field-based plant segmentation in the near future. This, in turn, will be a cornerstone of data-driven improvement in crop breeding and agricultural practices of global benefit.https://www.frontiersin.org/articles/10.3389/fpls.2021.774068/fulldeep learningbreedingmachine learningremote sensingrandom forrestsupport vector classification |
spellingShingle | Radek Zenkl Radu Timofte Norbert Kirchgessner Lukas Roth Andreas Hund Luc Van Gool Achim Walter Helge Aasen Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset Frontiers in Plant Science deep learning breeding machine learning remote sensing random forrest support vector classification |
title | Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset |
title_full | Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset |
title_fullStr | Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset |
title_full_unstemmed | Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset |
title_short | Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset |
title_sort | outdoor plant segmentation with deep learning for high throughput field phenotyping on a diverse wheat dataset |
topic | deep learning breeding machine learning remote sensing random forrest support vector classification |
url | https://www.frontiersin.org/articles/10.3389/fpls.2021.774068/full |
work_keys_str_mv | AT radekzenkl outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT radutimofte outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT norbertkirchgessner outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT lukasroth outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT andreashund outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT lucvangool outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT achimwalter outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset AT helgeaasen outdoorplantsegmentationwithdeeplearningforhighthroughputfieldphenotypingonadiversewheatdataset |