MangoLeafNet: Transforming Mango Leaf Disease Management with Deep Learning
J. Relin Francis Raj, Sivakami Raja, Ezhil E. Nithila, K. Haribabu, R. Santhana Krishnan, A. Essaki Muthu
Abstract
Effective mango leaf disease detection is critical for optimizing crop health and yield, utilizing advanced models for accurate diagnosis. To discover important mango leaf diseases, this study uses two different datasets to investigate MangoLeafNet, an advanced deep-learning model based on EfficientN etB0. Four images are included in each dataset: healthy leaves, powdery mildew, Anthracnose, and sooty mold (also known as sooty mold). MangoLeafNet was chosen for its effective compound scaling methodology and evaluated for its classification performance. Metrics like accuracy, precision, recall, and Fl score were used to assess the Model's performance. In order to evaluate MangoLeafNet's comparative efficacy, it was compared with VGG16, MobileNetV1, and AlexNet, among other models. Because of its sophisticated feature extraction capabilities, the Model can classify data precisely, which makes it an invaluable tool for controlling diseases in the mango agricultural sector.