基于迁移学习与注意力机制的花生叶部病害识别算法

打开文本图片集

中图分类号：S565.2；TP183 文献标识码：A 文章编号：2095-5553（2025）07-0226-08

Abstract：Aimingattheproblemsoflowidentificationeficiencyofpeanutleafdiseasesanddiffcultyinone-site identification，apeanutleaf diseaserecognitionalgorithmbasedontheConvolution Block AtentionModule（CBAM） andMobileNetV2isproposed throughtransfer learning and by integrating theatentionmechanismof CBAM. Firstly，adataset offivekindsof peanut leaf disease images，including healthyleaves，black spotdisease leaves， brown spot disease leaves，net spot disease leaves，and mosaic disease leaves，is established.Secondly，a peanut leaf diseaserecognitionmodelisbuiltbyintegratingchannelatentionmechanismandspatialattentionmechanism.Finally， therecognitionaccuracy beforeandaftermodel improvement isanalyzed，compared with VGG16，InceptionV3and ResNet5O，and the detection time of a single image is predicted. Theexperimental results show that theaccuracy rates of MobileNetV2，VGG16，InceptionV3 and ResNet50 are 97.54% ， 97.34% ， 96.06% and 74.88% ，respectively， which are all lower than the accuracy rate of the improved model at 99.41% . The detection time for a single image is 0.061 s. The peanutleaf diseaserecognitionalgorithmbasedon transferlearningandatentionmechanismisalightweightneuralnetwork withhighaccuracyandfewmodelparameters.Itcanbeapliedinpeanutfieldsandusedforon-sitedetectionof peanutdiseasesby using mobile devices，enabling farmers to understand the growth status of peanuts in a timely manner.

Keywords：peanut；leaf disease identification；attention mechanism；transfer learning；MobileNetV2

0 引言

年都会出现花生产量减少和品质下降等问题，进而给种植户带来较大的经济损失。（剩余12630字）