Soybean Leaf Disease Detection Using Convolutional Neural Networks and Google Net Integration

Soybean holds significant importance in global agriculture due to its high protein and oil content, and its demand is increasing alongside the shift toward plant-based dietary preferences. In India, soybean cultivation is expanding rapidly due to an increase in veganism; however, it continues to be susceptible to a range of foliar diseases, including Bacterial Blight, Soybean Rust, Yellow Mosaic, and Brown Spot. This research proposes a deep learning-based method for the detection of soybean leaf diseases, utilizing an ensemble of Convolutional Neural Networks (CNNs) and the GoogleNet model. Through transfer learning, the system was trained on a well-structured dataset of 5,479 annotated soybean leaf images. By combining both architectures, the ensemble approach achieved enhanced accuracy in disease classification. The experimental findings affirm the model's effectiveness in precisely recognizing prevalent soybean leaf diseases. Future efforts will focus on expanding the dataset, integrating advanced deep learning models, and creating a mobile-based solution for real-time disease diagnosis directly in the field.