Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks
Breeding higher yielding forage species is limited by current manual harvesting and visual scoring techniques used for measuring or estimation of biomass. Automation and remote sensing for high throughput phenotyping has been used in recent years as a viable solution to this bottleneck. Here, we foc...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2020-02-01
|
Series: | Frontiers in Plant Science |
Subjects: | |
Online Access: | https://www.frontiersin.org/article/10.3389/fpls.2020.00159/full |
_version_ | 1811203568337682432 |
---|---|
author | Christopher J. Bateman Jaco Fourie Jeffrey Hsiao Kenji Irie Angus Heslop Anthony Hilditch Michael Hagedorn Bruce Jessep Steve Gebbie Kioumars Ghamkhar |
author_facet | Christopher J. Bateman Jaco Fourie Jeffrey Hsiao Kenji Irie Angus Heslop Anthony Hilditch Michael Hagedorn Bruce Jessep Steve Gebbie Kioumars Ghamkhar |
author_sort | Christopher J. Bateman |
collection | DOAJ |
description | Breeding higher yielding forage species is limited by current manual harvesting and visual scoring techniques used for measuring or estimation of biomass. Automation and remote sensing for high throughput phenotyping has been used in recent years as a viable solution to this bottleneck. Here, we focus on using RGB imaging and deep learning for white clover (Trifolium repens L.) and perennial ryegrass (Lolium perenne L.) yield estimation in a mixed sward. We present a new convolutional neural network (CNN) architecture designed for semantic segmentation of dense pasture and canopies with high occlusion to which we have named the local context network (LC-Net). On our testing data set we obtain a mean accuracy of 95.4% and a mean intersection over union of 81.3%, outperforming other methods we have found in the literature for segmenting clover from ryegrass. Comparing the clover/vegetation fraction for visual coverage and harvested dry-matter however showed little improvement from the segmentation accuracy gains. Further gains in biomass estimation accuracy may be achievable through combining RGB with complimentary information such as volumetric data from other sensors, which will form the basis of our future work. |
first_indexed | 2024-04-12T02:57:32Z |
format | Article |
id | doaj.art-b8a9706cb160472cb49306c0e0696806 |
institution | Directory Open Access Journal |
issn | 1664-462X |
language | English |
last_indexed | 2024-04-12T02:57:32Z |
publishDate | 2020-02-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Plant Science |
spelling | doaj.art-b8a9706cb160472cb49306c0e06968062022-12-22T03:50:45ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2020-02-011110.3389/fpls.2020.00159482642Assessment of Mixed Sward Using Context Sensitive Convolutional Neural NetworksChristopher J. Bateman0Jaco Fourie1Jeffrey Hsiao2Kenji Irie3Angus Heslop4Anthony Hilditch5Michael Hagedorn6Bruce Jessep7Steve Gebbie8Kioumars Ghamkhar9Lincoln Agritech Limited, Lincoln University, Lincoln, New ZealandLincoln Agritech Limited, Lincoln University, Lincoln, New ZealandLincoln Agritech Limited, Lincoln University, Lincoln, New ZealandRed Fern, Solutions Limited, Christchurch, New ZealandDevelopment Engineering, Lincoln Research Centre, AgResearch, Lincoln, New ZealandDevelopment Engineering, Lincoln Research Centre, AgResearch, Lincoln, New ZealandRed Fern, Solutions Limited, Christchurch, New ZealandDevelopment Engineering, Lincoln Research Centre, AgResearch, Lincoln, New ZealandDevelopment Engineering, Lincoln Research Centre, AgResearch, Lincoln, New ZealandForage, Science, Grasslands Research Centre, AgResearch, Palmerston North, New ZealandBreeding higher yielding forage species is limited by current manual harvesting and visual scoring techniques used for measuring or estimation of biomass. Automation and remote sensing for high throughput phenotyping has been used in recent years as a viable solution to this bottleneck. Here, we focus on using RGB imaging and deep learning for white clover (Trifolium repens L.) and perennial ryegrass (Lolium perenne L.) yield estimation in a mixed sward. We present a new convolutional neural network (CNN) architecture designed for semantic segmentation of dense pasture and canopies with high occlusion to which we have named the local context network (LC-Net). On our testing data set we obtain a mean accuracy of 95.4% and a mean intersection over union of 81.3%, outperforming other methods we have found in the literature for segmenting clover from ryegrass. Comparing the clover/vegetation fraction for visual coverage and harvested dry-matter however showed little improvement from the segmentation accuracy gains. Further gains in biomass estimation accuracy may be achievable through combining RGB with complimentary information such as volumetric data from other sensors, which will form the basis of our future work.https://www.frontiersin.org/article/10.3389/fpls.2020.00159/fullforage yieldcloverryegrassbiomasssemantic segmentationdeep learning |
spellingShingle | Christopher J. Bateman Jaco Fourie Jeffrey Hsiao Kenji Irie Angus Heslop Anthony Hilditch Michael Hagedorn Bruce Jessep Steve Gebbie Kioumars Ghamkhar Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks Frontiers in Plant Science forage yield clover ryegrass biomass semantic segmentation deep learning |
title | Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks |
title_full | Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks |
title_fullStr | Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks |
title_full_unstemmed | Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks |
title_short | Assessment of Mixed Sward Using Context Sensitive Convolutional Neural Networks |
title_sort | assessment of mixed sward using context sensitive convolutional neural networks |
topic | forage yield clover ryegrass biomass semantic segmentation deep learning |
url | https://www.frontiersin.org/article/10.3389/fpls.2020.00159/full |
work_keys_str_mv | AT christopherjbateman assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT jacofourie assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT jeffreyhsiao assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT kenjiirie assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT angusheslop assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT anthonyhilditch assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT michaelhagedorn assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT brucejessep assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT stevegebbie assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks AT kioumarsghamkhar assessmentofmixedswardusingcontextsensitiveconvolutionalneuralnetworks |