Enhanced Potato Leaf Disease Detection through Multi-Modal Fusion of Graph Neural Networks and ResNet18

• Published in: Proceedings (International Confernce on Computational Intelligence and Communication Networks) pp. 1498 - 1504
• Main Authors: Bajpai, Abhishek, Pal, Yogesh, Sahu, Shalinee, Srivastava, Vivek
• Format: Conference Proceeding
• Language: English
• Published: IEEE 22.12.2024
• Subjects: Computational modeling; Crops; Deep learning; Disease detection; Diseases; Feature extraction; Graph Neural Network; Graph neural networks; Manuals; Plant diseases; Potato leaf diseases; Productivity; ResNet18; Transfer learning
• ISSN: 2472-7555
• DOI: 10.1109/CICN63059.2024.10847571

In agricultural technology, artificial intelligence is crucial for early detection and management of plant diseases, especially in potato crops. Diseases like early and late blight can cause significant yield losses if not dealt with promptly. Manual diagnosis of these diseases is often time-consuming and laborious, highlighting the need for automated solutions to improve disease detection efficiency. A new study introduces an innovative approach that combines Graph Neural Network (GNN) and ResNet18 to identify potato leaf diseases accurately. This model uses deep learning and transfer learning techniques to extract relevant features from leaf images and ensure precise disease classification. The results show that this model can potentially revolutionize disease detection in potato crops, thus enhancing agricultural productivity through proactive disease management.