CottonLeafVision: An Explainable and Robust Deep Learning Framework for Cotton Leaf Disease Classification
Researchers have developed advanced deep learning frameworks for classifying plant diseases from leaf images, achieving high accuracy rates. One study focused on lemon leaf disease, utilizing ensemble models like InceptionV3 and MobileNetV2, reaching 99.27% accuracy with adversarial training for robustness. Another framework, CottonLeafVision, employed models such as DenseNet201, InceptionV3, and VGG19 to classify cotton leaf diseases, with DenseNet201 achieving 98% accuracy. A hybrid approach combining ResNet-50 with Vision Transformers also demonstrated strong performance, reaching 98.58% accuracy for multi-class plant disease identification, with interpretability techniques like Grad-CAM used across these studies to highlight disease-relevant regions. AI
IMPACT These frameworks offer improved accuracy and interpretability for precision agriculture, aiding in early disease detection and management.