Application of deep learning to leverage high-throughput phenotyping for crop breeding

Monica Furaste Danilevicz

doi:10.26182/ymmg-d165

Application of deep learning to leverage high-throughput phenotyping for crop breeding

Monica Furaste Danilevicz

Research output: Thesis › Doctoral Thesis

73 Downloads (Pure)

Abstract

High-throughput plant phenotyping (HTPP) platforms can monitor phenotypic variation through multiple sensor types, generating valuable datasets for characterising the phenotype. However, the lack of appropriate analysis methods leads to the underutilisation of HTPP in crop breeding. This thesis has sought to investigate the advantage of using deep learning models to tackle unsolved challenges in using HTPP datasets for crop research by (i) implementing models for yield prediction of crops under breeding trial; (ii) assessing the advantages of integrating varied data types in the deep learning model; and (iii) building a model to identify weeds among a similar looking crop

Original language	English
Qualification	Doctor of Philosophy
Awarding Institution	The University of Western Australia
Supervisors/Advisors	Bennamoun, Mohammed, Supervisor Bayer, Philipp, Supervisor Edwards, Dave, Supervisor Batley, Jacqueline, Supervisor
Thesis sponsors	Australian Government Research Training Program
Award date	21 Aug 2023
DOIs	https://doi.org/10.26182/ymmg-d165
Publication status	Unpublished - 2023

Access to Document

10.26182/ymmg-d165

THESIS - DOCTOR OF PHILOSOPHY - FURASTE DANILEVICZ Monica - 2023
This work is protected by Copyright. You may print or download ONE copy of this document for the purpose of your own non-commercial research or study. Any other use requires permission from the copyright owner. The Copyright Act requires you to attribute any copyright works you quote or paraphrase.

7 Review article
3 Article

Segmentation of Sandplain Lupin Weeds from Morphologically Similar Narrow-Leafed Lupins in the Field
Danilevicz, M. F., Rocha, R. L., Batley, J., Bayer, P. E., Bennamoun, M., Edwards, D. & Ashworth, M. B., Apr 2023, In: Remote Sensing. 15, 7, 1817.
Research output: Contribution to journal › Article › peer-review

Open Access
2 Citations (Scopus)
Expanding Gene-Editing Potential in Crop Improvement with Pangenomes
Tay Fernandez, C. G., Nestor, B. J., Danilevicz, M. F., Marsh, J. I., Petereit, J., Bayer, P. E., Batley, J. & Edwards, D., 1 Feb 2022, In: International Journal of Molecular Sciences. 23, 4, 2276.
Research output: Contribution to journal › Review article › peer-review

Open Access
18 Citations (Scopus)
Genetic and Genomic Resources for Soybean Breeding Research
Petereit, J., Marsh, J. I., Bayer, P. E., Danilevicz, M. F., Thomas, W. J. W., Batley, J. & Edwards, D., 1 May 2022, In: Plants. 11, 9, 1181.
Research output: Contribution to journal › Review article › peer-review

Open Access
9 Citations (Scopus)

Cite this

@phdthesis{919c9f54f6834cfaa33af800d0696313,

title = "Application of deep learning to leverage high-throughput phenotyping for crop breeding",

abstract = "High-throughput plant phenotyping (HTPP) platforms can monitor phenotypic variation through multiple sensor types, generating valuable datasets for characterising the phenotype. However, the lack of appropriate analysis methods leads to the underutilisation of HTPP in crop breeding. This thesis has sought to investigate the advantage of using deep learning models to tackle unsolved challenges in using HTPP datasets for crop research by (i) implementing models for yield prediction of crops under breeding trial; (ii) assessing the advantages of integrating varied data types in the deep learning model; and (iii) building a model to identify weeds among a similar looking crop",

keywords = "Machine Learning, Image Segmentation, Unmanned Aerial Vehicles, Weed Control, Multimodal Deep Learning, Open Source Code, Explainable Artificial Intelligence",

author = "{Furaste Danilevicz}, Monica",

year = "2023",

doi = "10.26182/ymmg-d165",

language = "English",

school = "The University of Western Australia",

}

TY - BOOK

T1 - Application of deep learning to leverage high-throughput phenotyping for crop breeding

AU - Furaste Danilevicz, Monica

PY - 2023

Y1 - 2023

N2 - High-throughput plant phenotyping (HTPP) platforms can monitor phenotypic variation through multiple sensor types, generating valuable datasets for characterising the phenotype. However, the lack of appropriate analysis methods leads to the underutilisation of HTPP in crop breeding. This thesis has sought to investigate the advantage of using deep learning models to tackle unsolved challenges in using HTPP datasets for crop research by (i) implementing models for yield prediction of crops under breeding trial; (ii) assessing the advantages of integrating varied data types in the deep learning model; and (iii) building a model to identify weeds among a similar looking crop

AB - High-throughput plant phenotyping (HTPP) platforms can monitor phenotypic variation through multiple sensor types, generating valuable datasets for characterising the phenotype. However, the lack of appropriate analysis methods leads to the underutilisation of HTPP in crop breeding. This thesis has sought to investigate the advantage of using deep learning models to tackle unsolved challenges in using HTPP datasets for crop research by (i) implementing models for yield prediction of crops under breeding trial; (ii) assessing the advantages of integrating varied data types in the deep learning model; and (iii) building a model to identify weeds among a similar looking crop

KW - Machine Learning

KW - Image Segmentation

KW - Unmanned Aerial Vehicles

KW - Weed Control

KW - Multimodal Deep Learning

KW - Open Source Code

KW - Explainable Artificial Intelligence

U2 - 10.26182/ymmg-d165

DO - 10.26182/ymmg-d165

M3 - Doctoral Thesis

ER -

Application of deep learning to leverage high-throughput phenotyping for crop breeding

Abstract

Access to Document

Fingerprint

Research output

Segmentation of Sandplain Lupin Weeds from Morphologically Similar Narrow-Leafed Lupins in the Field

Expanding Gene-Editing Potential in Crop Improvement with Pangenomes

Genetic and Genomic Resources for Soybean Breeding Research

Cite this