Abstract:
Mango is an economically significant tropical fruit, vulnerable to various diseases that can significantly impact yield and quality. In this study, we propose a mango leaf disease analysis system using Convolutional Neural Networks (CNNs) to enable timely and accurate disease detection and classification. The approach leverages a combination of a pre-trained VGG16 model and custom trainable layers, optimizing the model's performance while minimizing false positives and false negatives. Through rigorous training on a diverse dataset of mango leaf images, the CNN model demonstrates high precision and efficiency in identifying diseases such as anthracnose, powdery mildew, and bacterial spot. The system's practical implications extend beyond mango farming, potentially benefiting other agricultural settings through transfer learning. Overall, this research contributes to the advancement of agricultural artificial intelligence, empowering farmers and experts to make informed decisions for disease management and sustainable crop production.
