Abstract
Abstract Purpose The discrimination of grapevine cultivars is an important yet underexploredtopic in precision viticulture. This study evaluated three aspects of grapevinecultivar classification and model training strategies: (1) the influence ofseasonal variability and data collection strategy; (2) model generalisabilityacross sample types; and (3) the combined impact of temporal and sample-typevariation on classification performance. Methods Spectral data from leaf and canopy samples were used to classify six grapevinecultivars: Currant, Merbein, Muscat, Selma Pete, Sugra-39, and Sultana. Elevendataset configurations were evaluated using ten machine learning and deeplearning classifiers. Performance was assessed using F1 score, balancedaccuracy (BACC), Matthews Correlation Coefficient (MCC), and Area Under theReceiver Operating Characteristic curve (AUC). The Sum of Ranking Differences(SRD) method identified the most robust classifiers across training strategies. Results Leaf spectra collected in December—coinciding with the fruit set phenologicalstage—provided the highest classification accuracy. Models showed limitedgeneralisability across data scales (i.e., sample types), with substantialdeclines in accuracy when trained on one type and tested on another. Combiningleaf and canopy spectra in the training data improved performance but remainedlower than when models were trained and tested on the same sample type. SRDanalysis identified Support Vector Machine (SVM) and 1D Convolutional NeuralNetwork (CNN) as the most robust classifiers. 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.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). Conclusion These findings highlight the importance of temporal and sample-typeconsiderations when developing spectral classification models for grapevinecultivars. Overall, the results provide a foundation for a scalable andcommercially viable approach to cultivar mapping, supporting more precisevineyard management.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 12
Claim · model-written summary
Leaf spectra collected in December, at the fruit set stage, gave the highest cultivar classification accuracy.
VitisNot reportedValue
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“Leaf spectra collected in December—coinciding with the fruit set phenological stage—provided the highest classification accuracy”
Claim · model-written summary
Models showed limited generalisability across sample types, with substantial declines in accuracy when trained on one type and tested on another.
- Outcome
- limited generalisabilitye1
- Qualifier
- substantial declines in accuracy when trained on one type and tested on anothere1
Not reportedValue
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“Models showed limited generalisability across data scales (i.e., sample types), with substantial declines in accuracy when trained on one type and tested on another”
Claim · model-written summary
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.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“Combining leaf and canopy spectra in the training data improved performance but remained lower than when models were trained and tested on the same sample type”
Claim · model-written summary
SRD analysis identified Support Vector Machine and 1D Convolutional Neural Network as the most robust classifiers.
- Outcome
- most robust classifierse1
- Qualifier
- Support Vector Machine (SVM) and 1D Convolutional Neural Network (CNN)e1
Not reportedValue
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“SRD analysis identified Support Vector Machine (SVM) and 1D Convolutional Neural Network (CNN) as the most robust classifiers”
Claim · model-written summary
SVM models achieved F1 scores of 0.67-1.00.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“SVM models achieved F1 scores of0.67–1.00”
Claim · model-written summary
SVM models achieved balanced accuracy (BACC) of 0.74-0.98.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“SVM models achieved F1 scores of0.67–1.00, BACC of 0.74–0.98”
Claim · model-written summary
SVM models achieved Matthews Correlation Coefficient (MCC) of 0.65-0.90.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“SVM models achieved F1 scores of0.67–1.00, BACC of 0.74–0.98, MCC of 0.65–0.90”
Claim · model-written summary
SVM models achieved AUC values of 0.87-0.94.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
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”
Claim · model-written summary
1D CNN models achieved F1 of 0.62-0.98.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“The 1D CNN models also showed high performance (F1: 0.62–0.98”
Claim · model-written summary
1D CNN models achieved BACC of 0.79-0.99.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“The 1D CNN models also showed high performance (F1: 0.62–0.98; BACC: 0.79–0.99”
Claim · model-written summary
1D CNN models achieved MCC of 0.57-0.97.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
e1“The 1D CNN models also showed high performance (F1: 0.62–0.98; BACC: 0.79–0.99;MCC: 0.57–0.97”
Claim · model-written summary
1D CNN models achieved AUC of 0.93-1.00.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Grapevine Cultivar Discrimination”
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”