|
| Author and year of publication | Techniques | Type of leaf | Type of disease | Dataset count | Testing accuracy |
|
| [24] | Simple linear cluster and support vector machine (SVM) | Tea plant leaf | Tea anthracnose (Ta), tea brown blight (Tbb), tea netted blister blight (Tnbb), Exobasidium vexans Massee (EVM), and Pestalotiopsis | 1308 | Accuracy-98.5%
Precision-96.8%
Recall-98.6%
value-97.7% |
| [25] | Region proposed network, Chan-Vese algorithm | Plant diseases | Black rot, bacteria plaque, and rust | 4714 | 83.57% |
| [26] | Extreme learning machine, support vector machine | Plant leaf diseases | — | — | LSVM-87.5%
PSVM-92.5%
ELM-95% |
| [27] | ML (support vector machine (SVM), random forest (RF), and stochastic gradient descent (SGD)) and DL (Inception-v3, VGG-16, and VGG-19) | Citrus diseases | Canker, black spot, and greening melanosis | 609 | RF-76.8%, SGD-86.5%, SVM-87%, , and VGG-16–89.5% |
| [28] | Deep convolutional neural network | Citrus pest | Citrus leaf miner, sooty mold, and Pulvinaria | 1774 | Accuracy-99.04 |
| [29] | Representative as light gradient boosting machine (LightGBM) and eXtreme gradient boosting (XGBoost) | Allergen nomenclature (not leaf data) | 583 food allergens, 600 food proteins | — | Accuracy-0.957 |
| [30] | Random forest, K-nearest neighbor | Plant disease | Brown rust, early blight, and late blight | 1000 | SVM-93%
RF-96% |
|
