By research area › Breeding, agronomy & horticulture › Method / resource

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

SVM models achieved balanced accuracy (BACC) of 0.74-0.98.

Taxon: GrapevineVitist0
Outcome: BACCe1
Value: 0.74–0.98e1
Qualifier: SVM modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“SVM models achieved F1 scores of0.67–1.00, BACC of 0.74–0.98”
✓ located verbatim in the source · abstract, characters 1565–1625

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Method / resource SVM models achieved AUC values of 0.87-0.94.
Method / resource 1D CNN models achieved F1 of 0.62-0.98.
Method / resource 1D CNN models achieved BACC of 0.79-0.99.
Method / resource 1D CNN models achieved MCC of 0.57-0.97.
Method / resource 1D CNN models achieved AUC of 0.93-1.00.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.

Taxon: GrapevineVitist0
Outcome: MCCe1
Value: 0.65–0.90e1
Qualifier: SVM modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“SVM models achieved F1 scores of0.67–1.00, BACC of 0.74–0.98, MCC of 0.65–0.90”
✓ located verbatim in the source · abstract, characters 1565–1643

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Method / resource SVM models achieved AUC values of 0.87-0.94.
Method / resource 1D CNN models achieved F1 of 0.62-0.98.
Method / resource 1D CNN models achieved BACC of 0.79-0.99.
Method / resource 1D CNN models achieved MCC of 0.57-0.97.
Method / resource 1D CNN models achieved AUC of 0.93-1.00.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

SVM models achieved AUC values of 0.87-0.94.

Taxon: GrapevineVitist0
Outcome: AUC valuese1
Value: 0.87–0.94e1
Qualifier: SVM modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“SVM models achieved F1 scores of0.67–1.00, BACC of 0.74–0.98, MCC of 0.65–0.90, and AUC values of 0.87–0.94”
✓ located verbatim in the source · abstract, characters 1565–1672

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Method / resource SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Method / resource 1D CNN models achieved F1 of 0.62-0.98.
Method / resource 1D CNN models achieved BACC of 0.79-0.99.
Method / resource 1D CNN models achieved MCC of 0.57-0.97.
Method / resource 1D CNN models achieved AUC of 0.93-1.00.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

1D CNN models achieved F1 of 0.62-0.98.

Taxon: GrapevineVitist0
Outcome: F1e1
Value: 0.62–0.98e1
Qualifier: 1D CNN modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“The 1D CNN models also showed high performance (F1: 0.62–0.98”
✓ located verbatim in the source · abstract, characters 1673–1734

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Method / resource SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Method / resource SVM models achieved AUC values of 0.87-0.94.
Method / resource 1D CNN models achieved BACC of 0.79-0.99.
Method / resource 1D CNN models achieved MCC of 0.57-0.97.
Method / resource 1D CNN models achieved AUC of 0.93-1.00.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

1D CNN models achieved BACC of 0.79-0.99.

Taxon: GrapevineVitist0
Outcome: BACCe1
Value: 0.79–0.99e1
Qualifier: 1D CNN modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“The 1D CNN models also showed high performance (F1: 0.62–0.98; BACC: 0.79–0.99”
✓ located verbatim in the source · abstract, characters 1673–1751

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Method / resource SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Method / resource SVM models achieved AUC values of 0.87-0.94.
Method / resource 1D CNN models achieved F1 of 0.62-0.98.
Method / resource 1D CNN models achieved MCC of 0.57-0.97.
Method / resource 1D CNN models achieved AUC of 0.93-1.00.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

1D CNN models achieved MCC of 0.57-0.97.

Taxon: GrapevineVitist0
Outcome: MCCe1
Value: 0.57–0.97e1
Qualifier: 1D CNN modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“The 1D CNN models also showed high performance (F1: 0.62–0.98; BACC: 0.79–0.99;MCC: 0.57–0.97”
✓ located verbatim in the source · abstract, characters 1673–1766

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Method / resource SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Method / resource SVM models achieved AUC values of 0.87-0.94.
Method / resource 1D CNN models achieved F1 of 0.62-0.98.
Method / resource 1D CNN models achieved BACC of 0.79-0.99.
Method / resource 1D CNN models achieved AUC of 0.93-1.00.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

1D CNN models achieved AUC of 0.93-1.00.

Taxon: GrapevineVitist0
Outcome: AUCe1
Value: 0.93–1.00e1
Qualifier: 1D CNN modelse1

Verbatim

t0
“Grapevine Cultivar Discrimination”
✓ located verbatim in the source · title, characters 114–147

e1
“The 1D CNN models also showed high performance (F1: 0.62–0.98; BACC: 0.79–0.99;MCC: 0.57–0.97; AUC: 0.93–1.00”
✓ located verbatim in the source · abstract, characters 1673–1782

Title

Robust Spectral Classification Under Sample Type and Seasonal Variability: A Proximal Remote Sensing Approach for Grapevine Cultivar Discrimination

Kyle Loggenberg; Albert Strever; Zahn Münch Precision Agriculture 2026. doi:10.1007/s11119-026-10333-w

Other claims from this paper (11)

Method / resource Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
Method / resource Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
Method / resource Combining leaf and canopy spectra in training improved performance but it remained lower than when models were trained and tested on the same sample type.
Method / resource SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
Method / resource SVM models achieved F1 scores of 0.67-1.00.
Method / resource SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Method / resource SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Method / resource SVM models achieved AUC values of 0.87-0.94.
Method / resource 1D CNN models achieved F1 of 0.62-0.98.
Method / resource 1D CNN models achieved BACC of 0.79-0.99.
Method / resource 1D CNN models achieved MCC of 0.57-0.97.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

FIBA U-Net achieved the best performance across all evaluation metrics when compared with Eff U-Net, ResNeSt50, ConvNeXt and ViT for segmenting maize ear damage.

Taxon: maizeZea mayst0
Outcome: performance across all evaluation metricse1
Direction: beste1
Comparator: Eff U-Net, ResNeSt50, ConvNeXt, and ViTe0

Verbatim

t0
“maize ear harvesters”
✓ located verbatim in the source · title, characters 31–51

e0
“We compare the performance of FIBA U-Net with Eff U-Net, ResNeSt50, ConvNeXt, and ViT”
✓ located verbatim in the source · abstract, characters 829–914

e1
“FIBA U-Net achieves the best performance across all evaluation metrics”
✓ located verbatim in the source · abstract, characters 1015–1085

Title

Harvesting damage detection in maize ear harvesters based on the Fused Inverted Bottleneck Attention U-Net

Jun Fu; Chengyuan Zhang; Qiankun Fu; et al. Artificial Intelligence in Agriculture 2026. doi:10.1016/j.aiia.2026.02.005

Other claims from this paper (3)

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The method reached an overall accuracy of up to 98.61% in field tests.

Taxon: maizeZea mayst0
Outcome: overall accuracye0
Value: 98.61%e0
Setting: field testse0

Verbatim

t0
“maize ear harvesters”
✓ located verbatim in the source · title, characters 31–51

e0
“The method achieved an overall accuracy of up to 98.61% in field tests”
✓ located verbatim in the source · abstract, characters 1087–1157

Title

Harvesting damage detection in maize ear harvesters based on the Fused Inverted Bottleneck Attention U-Net

Jun Fu; Chengyuan Zhang; Qiankun Fu; et al. Artificial Intelligence in Agriculture 2026. doi:10.1016/j.aiia.2026.02.005

Other claims from this paper (3)

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Damage detection rates at the bottom of the grain collection box differed from actual damage rates by only 0.31% under partial coverage.

Taxon: maizeZea mayst0
Outcome: damage detection ratese0
Value: 0.31%e0
Qualifier: under partial coveragee0
Comparator: actual damage ratese0

Verbatim

t0
“maize ear harvesters”
✓ located verbatim in the source · title, characters 31–51

e0
“The damage detection rates for the bottom of the grain collection box differed from the actual damage rates by only 0.31% under partial coverage and 0.42% under full coverage”
✓ located verbatim in the source · abstract, characters 1159–1333

Title

Harvesting damage detection in maize ear harvesters based on the Fused Inverted Bottleneck Attention U-Net

Jun Fu; Chengyuan Zhang; Qiankun Fu; et al. Artificial Intelligence in Agriculture 2026. doi:10.1016/j.aiia.2026.02.005

Other claims from this paper (3)

Method / resource 289,819 claims

289,819 claims