基于YOLOv5-CBE的农作物害虫智能检测方法

打开文本图片集

中图分类号：TP391；TP311 文献标识码：A文章编号:1006-8228(2025)10-06-05

IntelligentDetectionofCropPestsvia YOLOv5-CBE

Chen Xiaoqiang，Wang Hong,Xiao Shao (Ji'anCollege,Ji'an,Jiangxi,China)

Abstract:Croppestsareacriticalconstraintonglobalfoodsecurityandthesustainabledevelopmentofagriculture.Conventional chemicalcontrolandmaualmonringntailologicalpolution,loweficencyandighsujectivityTableadate andautomatedpestdetection，weproposeYOLOv5-CBE，animprovedYOLOv5model.CBAMisembeddedinthebackbone, BiFPNservesastheweightedbidirectionalfeaturepyramidandEIoUlossisadopted.ExperimentsshowthatYOLOv5-CBE retains high inference speed while raising mAP(∅0.5 by 7.9 percentage points,reaching 86.5%

Keywords:YOLOv5-CBE;Automated Detection;CBAM;BiFPN;EIoU

0引言

农作物害虫是影响农业生产的重要因素之一，每年因虫害造成的农作物产量损失达到 20%-40% ，对农作物产量构成了严重威胁。（剩余8760字）