Rice disease detection method, product, medium and equipment

• Main Authors: LU WEIAN, LIU YONGCHUN
• Format: Patent
• Language: Chinese; English
• Published: 24.05.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses a rice disease detection method, a product, a medium and equipment, and relates to the field of crop disease identification, and the method comprises the steps: obtaining a training data set comprising a plurality of rice image samples marked with disease positions and disease types; the method comprises the following steps: on the basis of a YOLOv8n network, replacing all bottleneck layers in a C2f module in a backbone network of the network with a Ghost bottleneck layer, embedding an EMA attention mechanism behind the C2f module of a small target detection layer in a neck network of the network, and introducing an NWD loss function to obtain a GEN-YOLO network model; training and optimizing the model by using a training data set; and inputting a to-be-detected rice image into the optimized model, and obtaining a disease position and a disease type in the to-be-detected rice image by using the optimized model. According to the invention, YOLOv8 can be used for rice disease detection,