Papaya Leaf Disease Identification using ResNet with Transfer Learning

Papaya (Carica papaya) is an important tropical fruit crop known for its nutritional value and economic significance. It is widely cultivated in many tropical and subtropical regions due to its adaptability to various climatic conditions. Nonetheless, papaya plants can be affected by multiple leaf diseases like brown spot, black spot, powdery mildew, and mealybug, resulting in substantial reductions in yield and quality. Detecting these diseases early and accurately is vital to implement effective management strategies. This research introduces a mobile application that utilizes transfer learning and ResNet models, offering an accessible means to identify papaya leaf diseases. This enables timely interventions and enhances crop management practices. By leveraging pretrained ResNet model, which is a type of CNN, and fine-tuning them with a dataset of papaya leaf images, the system achieves an average accuracy of 88%. The automated system developed in this research could enhance productivity and sustainability in papaya cultivation, benefiting farmers and the agricultural industry.