Identification of Banana Leaf Diseases: A Collaboration of Deep Learning Models

The analysis in this research paper investigates the usage of Convolutional Neural Networks (CNNs), particularly the ResNet-18 and ResNet-34 architectures, for predicting diseases in banana leaves. The research involves a broad dataset containing images of both healthy banana leaves and leaves infected by various diseases such as segatoka and xamthomonas. This research illustrates the viability of utilizing CNNs, specifically ResNet-18 and ResNet-34, in addressing the challenge of banana leaf disease prediction. By harnessing the power of deep learning, the agricultural sector can benefit from advanced tools that contribute to improved disease management, increased yield, and enhanced food security. Ultimately, the trained models demonstrate promising results in predicting banana leaf diseases, showcasing the potential to assist farmers and agricultural experts in early disease detection and management.