Segmentation of roots in soil with U-Net

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Standard

Segmentation of roots in soil with U-Net. / Smith, Abraham George; Petersen, Jens; Selvan, Raghavendra; Rasmussen, Camilla Ruø.

I: Plant Methods, Bind 16, 13, 2020, s. 1-15.

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Harvard

Smith, AG, Petersen, J, Selvan, R & Rasmussen, CR 2020, 'Segmentation of roots in soil with U-Net', Plant Methods, bind 16, 13, s. 1-15. https://doi.org/10.1186/s13007-020-0563-0

APA

Smith, A. G., Petersen, J., Selvan, R., & Rasmussen, C. R. (2020). Segmentation of roots in soil with U-Net. Plant Methods, 16, 1-15. [13]. https://doi.org/10.1186/s13007-020-0563-0

Vancouver

Smith AG, Petersen J, Selvan R, Rasmussen CR. Segmentation of roots in soil with U-Net. Plant Methods. 2020;16:1-15. 13. https://doi.org/10.1186/s13007-020-0563-0

Author

Smith, Abraham George ; Petersen, Jens ; Selvan, Raghavendra ; Rasmussen, Camilla Ruø. / Segmentation of roots in soil with U-Net. I: Plant Methods. 2020 ; Bind 16. s. 1-15.

Bibtex

@article{2b49b7fa8ce24d91ba7c33949735b84a,
title = "Segmentation of roots in soil with U-Net",
abstract = "Plant root research can provide a way to attain stress-tolerant crops that produce greater yield in a diverse array of conditions. Phenotyping roots in soil is often challenging due to the roots being difficult to access and the use of time consuming manual methods. Rhizotrons allow visual inspection of root growth through transparent surfaces. Agronomists currently manually label photographs of roots obtained from rhizotrons using a line-intersect method to obtain root length density and rooting depth measurements which are essential for their experiments. We investigate the effectiveness of an automated image segmentation method based on the U-Net Convolutional Neural Network (CNN) architecture to enable such measurements. We design a data-set of 50 annotated chicory (Cichorium intybus L.) root images which we use to train, validate and test the system and compare against a baseline built using the Frangi vesselness filter. We obtain metrics using manual annotations and line-intersect counts.",
author = "Smith, {Abraham George} and Jens Petersen and Raghavendra Selvan and Rasmussen, {Camilla Ru{\o}}",
year = "2020",
doi = "10.1186/s13007-020-0563-0",
language = "English",
volume = "16",
pages = "1--15",
journal = "Plant Methods",
issn = "1746-4811",
publisher = "BioMed Central",

}

RIS

TY - JOUR

T1 - Segmentation of roots in soil with U-Net

AU - Smith, Abraham George

AU - Petersen, Jens

AU - Selvan, Raghavendra

AU - Rasmussen, Camilla Ruø

PY - 2020

Y1 - 2020

N2 - Plant root research can provide a way to attain stress-tolerant crops that produce greater yield in a diverse array of conditions. Phenotyping roots in soil is often challenging due to the roots being difficult to access and the use of time consuming manual methods. Rhizotrons allow visual inspection of root growth through transparent surfaces. Agronomists currently manually label photographs of roots obtained from rhizotrons using a line-intersect method to obtain root length density and rooting depth measurements which are essential for their experiments. We investigate the effectiveness of an automated image segmentation method based on the U-Net Convolutional Neural Network (CNN) architecture to enable such measurements. We design a data-set of 50 annotated chicory (Cichorium intybus L.) root images which we use to train, validate and test the system and compare against a baseline built using the Frangi vesselness filter. We obtain metrics using manual annotations and line-intersect counts.

AB - Plant root research can provide a way to attain stress-tolerant crops that produce greater yield in a diverse array of conditions. Phenotyping roots in soil is often challenging due to the roots being difficult to access and the use of time consuming manual methods. Rhizotrons allow visual inspection of root growth through transparent surfaces. Agronomists currently manually label photographs of roots obtained from rhizotrons using a line-intersect method to obtain root length density and rooting depth measurements which are essential for their experiments. We investigate the effectiveness of an automated image segmentation method based on the U-Net Convolutional Neural Network (CNN) architecture to enable such measurements. We design a data-set of 50 annotated chicory (Cichorium intybus L.) root images which we use to train, validate and test the system and compare against a baseline built using the Frangi vesselness filter. We obtain metrics using manual annotations and line-intersect counts.

U2 - 10.1186/s13007-020-0563-0

DO - 10.1186/s13007-020-0563-0

M3 - Journal article

C2 - 32055251

VL - 16

SP - 1

EP - 15

JO - Plant Methods

JF - Plant Methods

SN - 1746-4811

M1 - 13

ER -

ID: 236512165