Article | Year | Single/ Multi task | Crop | Part | Diseases | Severity Grade/Level | Dataset | Models used | Results |
[49] | 2021 | Single | Wheat | Spike/fruit | Fusarium head blight (FHB) | Grade 0: [0 - 1%], Grade 1: (1% - 2.5%], Grade 2: (2.5% - 5%], Grade 3: (5% - 7.5%], Grade 4: (7.5% - 10%], Grade 5: (10% - 12.5%], Grade 6: (12.5% - 15%], Grade 7: (15% - 17.5%], Grade 8: (17.5% - 20%], Grade 9: (20% - 25%], Grade 10: (25% - 30%], Grade 11: (30% - 40%], Grade 12: (40% - 50%], Grade 13: (50% - 60%], Grade 14: (60% - 100%] | Self-collected dataset of 690 images | Mask-RCNN | accuracies of 77.76% and 98.81%, respectively for FHB detection and severity assessment. |
[20] | 2021 | single | Rice | Leaf | Bacterial Leaf Streak | Level 0: no lesion; Level 1: lesion < 10%; Level 2 = 11% - 25% lesion; Level 3: 26% - 45% lesion; Level 4: 46% - 65% lesion; Level 5: >65% lesion. | Self-collected dataset of 109 images | Proposed Deep Learning algorithm named BLSNet and based on Unet | Average accuracy: 94% |
[26] | 2021 | single | Coffee, Soybean and Wheat | Leaf | Coffee leaf miner, Soybean rust and Wheat tan spot | percentage | Three self-collected datasets: Coffee 406 images; Soybean 208 images and Wheat 152 images. | Unet, SegNet, PSPNet, FPN, DeepLabV3 (Xception) and DeepLabV3 (MobileNetV2) | Average precision values are ranged from 90.4% to 95.6% and recall values are ranged from 89.4% to 94.7%. |
[8] | 2021 | Multi-task | Tea | Leaf | Leaf blight | Mild and severe | Self-collected dataset of 398 images | Faste R-CNN and VGG16 | detection average precision and the severity grading accuracy improved by more than 6% and 9%, respectively, compared to existing solutions. |