Plant Disease Detection with Artificial Intelligence Algorithms for Smart Agriculture

Overseeing plant leaf diseases is vital for sustaining crop health, preventing extensive agricultural losses, safeguarding farmers livelihoods and ensuring a stable food supply. The health of the plant leaf is severely harmed by diseases such as early and late blight. Detecting leaf diseases is a time-consuming operation, and many farmers lack understanding about it, resulting in crop failure. Thus, early disease detection is crucial since it affects the quality and quantity of the potatoes. A type of artificial intelligence called deep learning has many uses in several domains and enables accurate and efficient data processing and decision-making. In this study, Leaf illnesses like early blight and late blight are detected and predicted using DL architectures like MobileNetV2, ResNet50, and EfficientNetB0. A comprehensive dataset is employed, containing images of healthy and infected leaves, which aids in model training and evaluation. The results show that MobileNetV2 achieves the highest accuracy of 91% in identifying leaf diseases.