Table 1 Comparative analysis.
From: Advancing plant leaf disease detection integrating machine learning and deep learning
Sl. No | Reference | Dataset used | Algorithms | Disease detection/identification | Description/outcome |
|---|---|---|---|---|---|
1 | Custard Apple Leaf diseases | kNN and SVM | Leaf parameter analysis, detection of N, P, K deficiencies, and leaf diseases | Accuracy-99.5% | |
2 | Citrus Leaf Disease | SVM, SGD, RF, Inceptionv3, VGG16, VGG19 | Canker, Blackspot, Greening, Melanose, Healthy | Classification Accuracy of RF-76.8%, SGD-86.5%, SVM-87%, VGG19-87.4%, Inceptionv3-89%, VGG16-89.5% | |
3 | Custard Apple disease | Expert System | Anthracnose, Leaf spot, Diplodia rot, Black canker, Spiral nematode, and Stunt nematode, IPM for Custard Apple | Expert System Developed | |
4 | 2 Apple Leaf and 1 Coffee Leaf | DL | A general framework for recognizing plant diseases | The proposed method achieves 99.79%, 92.59%, and 97.12% classification accuracies on the three datasets | |
5 | Alliance of Bioversity International and CIAT Banana Image Library | Total generalized variation fuzzy C means, CNN | Disease Classification | Sesitivity-89.04%, Specificity-96.38%, and Accuracy- 93.45% | |
6 | Banana Plant Leaf Disease | Histogram pixel localization, region-based edge normalization, Gabor-based binary patterns, convolution RNN, Convolutional Recurrent Neural Network–Region-Based Convolutional Neural Network (CRNN–RCNN) | Disease Classification | Precision-97.7%, recall-97.7%, sensitivity-98.69%, accuracy-98% | |
7 | PlantVillage | VGG16-based model | Disease Classification into 9 major types | Accuracy-99.7% and area under the curve (AUC)-93.3% | |
8 | Citrus fruits and leaves dataset | Ant Colony Optimization with Convolution Neural Network (ACO-CNN) | Classification of leaves | Accuracy-99.98% | |
9 | Potato Leaf Disease | Swin Transformer DL | Classification of healthy and unhealthy leaves | Accuracy(training)-99.70% | |
10 | PlantVillage | Modified DenseNet-201 | Potato Late Blight (PLB), Potato Early Blight (PEB), Potato Leaf Roll (PLR), Potato Verticillium_wilt (PVw) and Potato Healthy (PH) class | Accuracy-97.2% | |
11 | Papaya Fruit Diseases, Guava Leaves and Fruits Dataset, Citrus Fruits and Leaves Dataset | MobileNet-v2, VGG16, DenseNet121 | Detect disease in fruits via images | MobileNetv2 model, the disease prediction accuracy for papaya, guava, and citrus was 99.4%, 98.8%, and 95.8% and the recall values were 99.4%, 98.8%, and 93.8%, respectively.VGG16, the disease prediction accuracy for papaya, guava, and citrus was 97.7%, 99.6%, and 94.2% and the recall values were 96.5%, 99.6%, and 89.2%, respectively.DenseNet121, the disease prediction accuracy for papaya, guava, and citrus was 99.4%, 97.6%, and 99.2%, and the recall values were 98.8%, 97.6%, and 99.2%, respectively | |
12 | – | VGG-16, VGG-19, InceptionV3, and DenseNet-121 | Leaf Disease Identification | Accuracy-91.75% on DenseNet-121 |
