Table 3.
Comparison of several DL approaches in terms of various performance metrics.
| DL Architectures/Algorithms | Datasets | Selected Plant/s | Performance Metrics (and Their Results) | Refs |
|---|---|---|---|---|
| CNN | PlantVillage | Maize | CA (92.85%) | [56] |
| AlexNet, GoogLeNet, ResNet | PlantVillage | Tomato | CA by ResNet which gave the best value (97.28%) | [57] |
| LeNet | PlantVillage | Banana | CA (98.61%), F1 (98.64%) | [32] |
| AlexNet, ALexNetOWTBn, GoogLeNet, Overfeat, VGG | PlantVillage and in-field images | Apple, blueberry, banana, cabbage, cassava, cantaloupe, celery, cherry, cucumber, corn, eggplant, gourd, grape, orange, onion | Success rate of VGG (99.53%) which is the best among all | [58] |
| AlexNet, VGG16, VGG 19, SqueezeNet, GoogLeNet, Inceptionv3, InceptionResNetv2, ResNet50, Resnet101 | Real field dataset | Apricot, Walnut, Peach, Cherry | F1(97.14), Accuracy (97.86 ± 1.56) of ResNet | [35] |
| Inceptionv3 | Experimental field dataset | Cassava | CA (93%) | [59] |
| CNN | Images taken from the research center | Cucumber | CA (82.3%) | [60] |
| Super-Resolution Convolutional Neural Network (SCRNN) | PlantVillage | Tomato | Accuracy (~90%) | [61] |
| CaffeNet | Downloaded from the internet | Pear, cherry, peach, apple, grapevine | Precision (96.3%) | [27] |
| AlexNet and GoogLeNet | PlantVillage | Apple, blueberry, bell pepper, cherry, corn, peach, grape, raspberry, potato, squash, soybean, strawberry, tomato | CA (99.35%) of GoogLeNet | [25] |
| AlexNet, GoogLeNet, VGG- 16, ResNet-50,101, ResNetXt-101, Faster RCNN, SSD, R-FCN, ZFNet | Image taken in real fields | Tomato | Precision (85.98%) of ResNet-50 with Region based Fully Convolutional Network(R-FCN) | [68] |
| CNN | Bisque platform of Cy Verse | Maize | Accuracy (96.7%) | [70] |
| DCNN | Images were taken in real field | Rice | Accuracy (95.48%) | [78] |
| AlexNet, GoogLeNet | PlantVillage | Tomato | Accuracy (0.9918 ± 0.169) of GoogLeNet | [72] |
| VGG-FCN-VD16 and VGG-FCN-S | Wheat Disease Database 2017 | Wheat | Accuracy (97.95%) of VGG-FCN-VD16 | [73] |
| VGG-A, CNN | Images were taken in real field | Radish | Accuracy (93.3%) | [74] |
| AlexNet | Images were taken in real field | Soybean | CA (94.13%) | [77] |
| AlexNet and SqueezeNet v1.1 | PlantVillage | Tomato | CA (95.65%) of AlexNet | [62] |
| DCNN, Random forest, Support Vector Machine and AlexNet | PlantVillage dataset, Forestry Image dataset and agricultural field in China | Cucumber | CA (93.4%) of DCNN | [66] |
| Teacher/student architecture | PlantVillage | Apple, bell pepper, blueberry, cherry, corn, orange, grape, potato, raspberry, peach, soybean, strawberry, tomato, squash | Training accuracy and loss (~99%,~0–0.5%), validation accuracy and loss (~95%, ~10%) | [67] |
| Improved GoogLeNet, Cifar-10 | PlantVillage and various websites | Maize | Top-1 accuracy (98.9%) of improved GoogLeNet | [86] |
| MobileNet, Modified MobileNet, Reduced MobileNet | PlantVillage dataset | 24 types of plant | CA (98.34%) of reduced MobileNet | [89] |
| VGG-16, ResNet-50,101,152, Inception-V4 and DenseNets-121 | PlantVillage | Apple, bell pepper, blueberry, cherry, corn, orange, grape, potato, raspberry, peach, soybean, strawberry, tomato, squash | Testing accuracy (99.75%) of DenseNets | [63] |
| User defined CNN, SVM, AlexNet, GoogLeNet, ResNet-20 and VGG-16 | Images were taken in real field | Apple | CA (97.62%) of proposed CNN | [87] |
| AlexNet and VGG-16 | PlantVillage | Tomato | CA (AlexNet) | [64] |
| LeafNet, SVM, MLP | Images were taken in real field | Tea leaf | CA (90.16%) of LeafNet | [88] |
| 2D-CNN-BidGRU | Real wheat field | wheat | F1 (0.75) and accuracy (0.743) | [111] |
| OR-AC-GAN | Real environment | Tomato | Accuracy (96.25%) | [112] |
| 3D CNN | Real environment | Soybean | CA (95.73%), F1-score (0.87) | [84] |
| DCNN | Real environment | Wheat | Accuracy (85%) | [114] |
| ResNet-50 | Real environment | Wheat | Balanced Accuracy (87%) | [79] |
| GPDCNN | Real environment | Cucumber | CA (94.65%) | [81] |
| VGG-16, AlexNet | PlantVillage, CASC-IFW | Apple, banana | CA (98.6%) | [82] |
| LeNet | Real environment | Grapes | CA (95.8%) | [83] |
| PlantDiseaseNet | Real environment | Apple, bell-pepper, cherry, grapes, onion, peach, potato, plum, strawberry, sugar-beets, tomato, wheat | CA (93.67%) | [90] |
| LeNet | PlantVillage | Soybean | CA (99.32%) | [71] |
| VGG-Inception | Real environment | Apple | Mean average accuracy (78.8%) | [85] |
| Resnet-50, Inception-V2, MobileNet-V1 | Real environment | Banana | Mean average accuracy (99%) of ResNet-50 | [69] |
| Modified LeNet | PlantVillage | Olives | True positive rate (98.6 ± 1.47%) | [65] |