Design of silkworm disease detection algorithm based on deep learning
In the context of silkworm disease recognition,the YOLOv8 model is susceptible to interference from background information due to changes in the morphology and appearance of silkworm disease at different stages.To address this issue,a new feature extraction module called DCNv2-Block was proposed to enhance the model's ability to capture characteristics of silkworm disease at different stages.Experimental results demonstrated that compared with the original YOLOv8 model,the proposed im-proved method achieved the mAP0.5 of 96.5%,representing 1.1 percentage point increase.This enhancement enables better cap-ture of symptoms of silkworm diseases,improves accuracy in identifying silkworm diseases,and facilitates timely detection and control measures.
YOLOv8detection of silkworm diseasevariable convolutiondeep learning
万方数据
