Published Paper Details:

This research presents a novel approach for detecting diseases in soybean plants using Convolutional Neural Networks (CNNs) applied to RGB images from the PlantVillage dataset. The study employs transfer learning and data augmentation techniques to enhance model performance while reducing the need for costly Internet of Things (IoT) sensors or advanced imaging systems. Our method successfully identifies common soybean diseases, including Frogeye Leaf Spot, Septoria Brown Spot, Cercospora Leaf Blight, and Bacterial Blight, achieving a remarkable test accuracy of 97.6%. Additionally, we integrate explainable AI techniques, such as Grad-CAM, to provide insights into the model's decision-making process, allowing stakeholders to visualize critical features in the images. The results underscore the efficacy of deep learning strategies in improving early disease detection, ultimately contributing to increased soybean crop yields and economic sustainability. This work lays the foundation for scalable and accessible solutions for precision agriculture, promoting sustainable farming practices in the soybean industry.