Skip to main content
. 2024 Mar 13;15:1356260. doi: 10.3389/fpls.2024.1356260

Table 7.

Chilli vegetable classification using AI.

References AI Methods Dataset Disease Accuracy
(Naik et al., 2022) 12 pre-trained DL networks Plant leaf datasets from 43 different groups Down curl, gemini virus, cercospora, leaf spot, yellow leaf, and up curl SECNN=99.28%
(Ahmad Loti et al., 2021) SVM, RF, and ANN Self-collected data Bacterial, cercospora, mosaic, mottle virus, leaf curl, healthy, aphids-infestation, and whitefly-infestation SVM=92.10%
(Sachdeva et al., 2021) DCNN with Bayesian PlantVillage Black spot, botrytis blight, leaf spot, powdery mildew, and rust spores DCNN=98.9%
(Aminuddin et al., 2022) CNN, ResNet-18 Self-collected data Healthy and diseased chilli leaf CNN=97%
(Mustafa et al., 2023) Optimized-CNN Augmented dataset with 20000 images Health status of chilli plant leaves CNN=99.99%
(Karadağ et al., 2020) ANN, NB, and KNN 80 leaves samples collected by GATEAM Turkey Healthy, fusarium, mycorrhizal fungus, combine fusarium, and mycorrhizal KNN=100%
(Kanaparthi and Ilango, 2023) Optimized-SqueezeNet Kaggle dataset Gemni virus and mosaic SqueezeNet =100%
(KM et al., 2023) YOLOv5 Kaggle dataset Leaf spot and leaf curl YOLOv5 = 75.64%