Implementation of MobileNetV2 Architecture In Rice Disease Detection System Using Digital Images

Rice plant diseases pose a significant threat, reducing crop yields. Especially in agricultural areas such as the Tanjung Morawa region of Deli Serdang, accurate and rapid early detection is very difficult. Three main disease types, Bacterial Leaf Blight (BLB), Brown Spot, and Leaf Smut, are identified by the MobileNetV2 architecture in the rice disease detection system. The transfer learning method was employed to enhance training using local public data. Although metrics such as accuracy, precision, recall, and F1 score were used to measure performance, evaluation of the model showed that accuracy and detection efficiency had improved compared to traditional methods. By integrating this system into a Flask-based web application, users can upload photos of rice leaves and receive detection results directly. It is expected that this research will make a significant contribution to the development of intelligent agricultural technologies that will help farmers find rice diseases early and treat them correctly.