Research Article

Using a Camera System for the In-Situ Assessment of Cordon Dieback due to Grapevine Trunk Diseases

Table 2

Results of training GTD detection algorithm when applied to a completely unseen test set (Block 4).

IdentifierTraining dataTraining hyperparametersTrunks detected (%)Cordons detected (%)Class accuracy (%)Variation accuracy (±5%)Variation accuracy (±10%)

Model 1V2022 training and validation data. No augmentations appliedSee Supplementary Table 199.4215.796.4311.6814.02
Model 2V2022 training and validation data. No augmentations. 85% left-right flip appliedSee Supplementary Table 299.4282.5624.1256.0971.49
Model 3V2022 training and validation data. Exposure and blur augmentations appliedSee Supplementary Table 3100.0088.6624.9963.3577.00
Model 4V2021 and V2022 training and validation data. No augmentations. 85% left-right flip appliedSee Supplementary Table 497.0985.1726.7460.7474.96
Model 5V2021 and V2022 balanced training and validation data. No augmentationsSee Supplementary Table 598.8497.125.7264.3981.52
Model 6V2021 and V2022 balanced training and validation data. Exposure and blur augmentations applied. 85% left-right flip appliedSee Supplementary Table 6100.0099.4226.1663.5684.18