Deep machine learning provides state-of-the-art performance in image-based plant phenotyping

In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, hence the motivation for finding a fully automated approach. Deep lear...

Full description

Bibliographic Details
Main Authors: Pound, Michael P., Atkinson, Jonathan A., Townsend, Alexandra J., Wilson, Michael H., Griffiths, Marcus, Jackson, Aaron S., Bulat, Adrian, Tzimiropoulos, Georgios, Wells, Darren M., Murchie, Erik H., Pridmore, Tony P., French, Andrew P.
Format: Article
Published: Oxford University Press 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/47289/
_version_ 1848797508742938624
author Pound, Michael P.
Atkinson, Jonathan A.
Townsend, Alexandra J.
Wilson, Michael H.
Griffiths, Marcus
Jackson, Aaron S.
Bulat, Adrian
Tzimiropoulos, Georgios
Wells, Darren M.
Murchie, Erik H.
Pridmore, Tony P.
French, Andrew P.
author_facet Pound, Michael P.
Atkinson, Jonathan A.
Townsend, Alexandra J.
Wilson, Michael H.
Griffiths, Marcus
Jackson, Aaron S.
Bulat, Adrian
Tzimiropoulos, Georgios
Wells, Darren M.
Murchie, Erik H.
Pridmore, Tony P.
French, Andrew P.
author_sort Pound, Michael P.
building Nottingham Research Data Repository
collection Online Access
description In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, hence the motivation for finding a fully automated approach. Deep learning is an emerging field that promises unparalleled results on many data analysis problems. Building on artificial neural networks, deep approaches have many more hidden layers in the network, and hence have greater discriminative and predictive power. We demonstrate the use of such approaches as part of a plant phenotyping pipeline. We show the success offered by such techniques when applied to the challenging problem of image-based plant phenotyping and demonstrate state-of-the-art results (>97% accuracy) for root and shoot feature identification and localization. We use fully automated trait identification using deep learning to identify quantitative trait loci in root architecture datasets. The majority (12 out of 14) of manually identified quantitative trait loci were also discovered using our automated approach based on deep learning detection to locate plant features. We have shown deep learning–based phenotyping to have very good detection and localization accuracy in validation and testing image sets. We have shown that such features can be used to derive meaningful biological traits, which in turn can be used in quantitative trait loci discovery pipelines. This process can be completely automated. We predict a paradigm shift in image-based phenotyping bought about by such deep learning approaches, given sufficient training sets.
first_indexed 2025-11-14T20:05:00Z
format Article
id nottingham-47289
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:05:00Z
publishDate 2017
publisher Oxford University Press
recordtype eprints
repository_type Digital Repository
spelling nottingham-472892020-05-04T19:09:25Z https://eprints.nottingham.ac.uk/47289/ Deep machine learning provides state-of-the-art performance in image-based plant phenotyping Pound, Michael P. Atkinson, Jonathan A. Townsend, Alexandra J. Wilson, Michael H. Griffiths, Marcus Jackson, Aaron S. Bulat, Adrian Tzimiropoulos, Georgios Wells, Darren M. Murchie, Erik H. Pridmore, Tony P. French, Andrew P. In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, hence the motivation for finding a fully automated approach. Deep learning is an emerging field that promises unparalleled results on many data analysis problems. Building on artificial neural networks, deep approaches have many more hidden layers in the network, and hence have greater discriminative and predictive power. We demonstrate the use of such approaches as part of a plant phenotyping pipeline. We show the success offered by such techniques when applied to the challenging problem of image-based plant phenotyping and demonstrate state-of-the-art results (>97% accuracy) for root and shoot feature identification and localization. We use fully automated trait identification using deep learning to identify quantitative trait loci in root architecture datasets. The majority (12 out of 14) of manually identified quantitative trait loci were also discovered using our automated approach based on deep learning detection to locate plant features. We have shown deep learning–based phenotyping to have very good detection and localization accuracy in validation and testing image sets. We have shown that such features can be used to derive meaningful biological traits, which in turn can be used in quantitative trait loci discovery pipelines. This process can be completely automated. We predict a paradigm shift in image-based phenotyping bought about by such deep learning approaches, given sufficient training sets. Oxford University Press 2017-10-01 Article PeerReviewed Pound, Michael P., Atkinson, Jonathan A., Townsend, Alexandra J., Wilson, Michael H., Griffiths, Marcus, Jackson, Aaron S., Bulat, Adrian, Tzimiropoulos, Georgios, Wells, Darren M., Murchie, Erik H., Pridmore, Tony P. and French, Andrew P. (2017) Deep machine learning provides state-of-the-art performance in image-based plant phenotyping. GigaScience, 6 (10). pp. 1-10. ISSN 2047-217X Phenotyping; Deep learning; Root; Shoot; QTL; Image analysis https://doi.org/10.1093/gigascience/gix083 doi:10.1093/gigascience/gix083 doi:10.1093/gigascience/gix083
spellingShingle Phenotyping; Deep learning; Root; Shoot; QTL; Image analysis
Pound, Michael P.
Atkinson, Jonathan A.
Townsend, Alexandra J.
Wilson, Michael H.
Griffiths, Marcus
Jackson, Aaron S.
Bulat, Adrian
Tzimiropoulos, Georgios
Wells, Darren M.
Murchie, Erik H.
Pridmore, Tony P.
French, Andrew P.
Deep machine learning provides state-of-the-art performance in image-based plant phenotyping
title Deep machine learning provides state-of-the-art performance in image-based plant phenotyping
title_full Deep machine learning provides state-of-the-art performance in image-based plant phenotyping
title_fullStr Deep machine learning provides state-of-the-art performance in image-based plant phenotyping
title_full_unstemmed Deep machine learning provides state-of-the-art performance in image-based plant phenotyping
title_short Deep machine learning provides state-of-the-art performance in image-based plant phenotyping
title_sort deep machine learning provides state-of-the-art performance in image-based plant phenotyping
topic Phenotyping; Deep learning; Root; Shoot; QTL; Image analysis
url https://eprints.nottingham.ac.uk/47289/
https://eprints.nottingham.ac.uk/47289/
https://eprints.nottingham.ac.uk/47289/