Semantic segmentation of plant roots from RGB (mini-) rhizotron images—generalisation potential and false positives of established methods and advanced deep-learning models
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Standard
Semantic segmentation of plant roots from RGB (mini-) rhizotron images—generalisation potential and false positives of established methods and advanced deep-learning models. / Baykalov, Pavel; Bussmann, Bart; Nair, Richard; Smith, Abraham George; Bodner, Gernot; Hadar, Ofer; Lazarovitch, Naftali; Rewald, Boris.
I: Plant Methods, Bind 19, Nr. 1, 122, 2023.Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Semantic segmentation of plant roots from RGB (mini-) rhizotron images—generalisation potential and false positives of established methods and advanced deep-learning models
AU - Baykalov, Pavel
AU - Bussmann, Bart
AU - Nair, Richard
AU - Smith, Abraham George
AU - Bodner, Gernot
AU - Hadar, Ofer
AU - Lazarovitch, Naftali
AU - Rewald, Boris
N1 - Publisher Copyright: © 2023, The Author(s).
PY - 2023
Y1 - 2023
N2 - Background: Manual analysis of (mini-)rhizotron (MR) images is tedious. Several methods have been proposed for semantic root segmentation based on homogeneous, single-source MR datasets. Recent advances in deep learning (DL) have enabled automated feature extraction, but comparisons of segmentation accuracy, false positives and transferability are virtually lacking. Here we compare six state-of-the-art methods and propose two improved DL models for semantic root segmentation using a large MR dataset with and without augmented data. We determine the performance of the methods on a homogeneous maize dataset, and a mixed dataset of > 8 species (mixtures), 6 soil types and 4 imaging systems. The generalisation potential of the derived DL models is determined on a distinct, unseen dataset. Results: The best performance was achieved by the U-Net models; the more complex the encoder the better the accuracy and generalisation of the model. The heterogeneous mixed MR dataset was a particularly challenging for the non-U-Net techniques. Data augmentation enhanced model performance. We demonstrated the improved performance of deep meta-architectures and feature extractors, and a reduction in the number of false positives. Conclusions: Although correction factors are still required to match human labelled root lengths, neural network architectures greatly reduce the time required to compute the root length. The more complex architectures illustrate how future improvements in root segmentation within MR images can be achieved, particularly reaching higher segmentation accuracies and model generalisation when analysing real-world datasets with artefacts—limiting the need for model retraining.
AB - Background: Manual analysis of (mini-)rhizotron (MR) images is tedious. Several methods have been proposed for semantic root segmentation based on homogeneous, single-source MR datasets. Recent advances in deep learning (DL) have enabled automated feature extraction, but comparisons of segmentation accuracy, false positives and transferability are virtually lacking. Here we compare six state-of-the-art methods and propose two improved DL models for semantic root segmentation using a large MR dataset with and without augmented data. We determine the performance of the methods on a homogeneous maize dataset, and a mixed dataset of > 8 species (mixtures), 6 soil types and 4 imaging systems. The generalisation potential of the derived DL models is determined on a distinct, unseen dataset. Results: The best performance was achieved by the U-Net models; the more complex the encoder the better the accuracy and generalisation of the model. The heterogeneous mixed MR dataset was a particularly challenging for the non-U-Net techniques. Data augmentation enhanced model performance. We demonstrated the improved performance of deep meta-architectures and feature extractors, and a reduction in the number of false positives. Conclusions: Although correction factors are still required to match human labelled root lengths, neural network architectures greatly reduce the time required to compute the root length. The more complex architectures illustrate how future improvements in root segmentation within MR images can be achieved, particularly reaching higher segmentation accuracies and model generalisation when analysing real-world datasets with artefacts—limiting the need for model retraining.
KW - Automatic image segmentation
KW - Data augmentation
KW - Deep learning
KW - False positives
KW - Fine roots
KW - Image processing
KW - Minirhizotron
KW - Neural networks
KW - Root segmentation
KW - U-Net
UR - http://www.scopus.com/inward/record.url?scp=85175806863&partnerID=8YFLogxK
U2 - 10.1186/s13007-023-01101-2
DO - 10.1186/s13007-023-01101-2
M3 - Journal article
C2 - 37932745
AN - SCOPUS:85175806863
VL - 19
JO - Plant Methods
JF - Plant Methods
SN - 1746-4811
IS - 1
M1 - 122
ER -
ID: 372962578