基于迁移学习的农作物病虫害检测方法研究与应用  

Research and Application of Crop Diseases Detection Method Based on Transfer Learning

作　　者：余小东 杨孟辑 张海清[3] 李丹[1] 唐毅谦[1] 于曦[1,2] YU Xiaodong;YANG Mengji;ZHANG Haiqing;LI Dan;TANG Yiqian;YU Xi(College of Computer Science,Chengdu University,Chengdu 610106,China;Key Laboratory of Pattern Recognition and Intelligent Information Processing in Sichuan,Chengdu University,Chengdu 610106,China;School of Software Engineering,Chengdu University of Information Technology,Chengdu 610225,China)

机构地区：[1]成都大学计算机学院,成都610106 [2]成都大学模式识别与智能信息处理四川省高校重点实验室,成都610106 [3]成都信息工程大学软件工程学院,成都610225

出　　处：《农业机械学报》2020年第10期252-258,共7页Transactions of the Chinese Society for Agricultural Machinery

基　　金：国家自然科学基金项目(61602064);欧盟Erasmus+SHYFTE项目(598649-EPP-1-2018-1-FR-EPPKA2-CBHE-JP)。

摘　　要：为了提高农作物病虫害严重程度(健康、一般、严重)的分类效果,采用迁移学习方式并结合深度学习提出了一种基于残差网络(ResNet 50)的CDCNNv2算法。通过对10类作物的3万多幅病虫害图像进行训练,获得了病虫害严重程度分类模型,其识别准确率可达91.51%。为了验证CDCNNv2模型的鲁棒性,分别与使用迁移学习的ResNet 50、Xception、VGG16、VGG19、DenseNet 121模型进行对比试验,结果表明,CDCNNv2模型比其他模型的平均精度提升了2.78~10.93个百分点,具有更高的分类精度,病虫害严重程度识别的鲁棒性增强。基于该算法所训练的模型,结合Android技术开发了一款实时在线农作物病虫害等级识别APP,通过拍摄农作物叶片病虫害区域图像,能够在0.1~0.5 s之内获取识别结果(物种病害种类严重程度)及防治建议。Classifying the severity of crop diseases is the staple basic element of the plant pathology for making disease prevent and control strategies.In order to achieve better results in the classification of the severity(healthy,general or severe)of crop diseases,a CDCNNv2 algorithm based on residual network(ResNet 50)and deep transfer learning was proposed.By training more than 30,000 crop disease images which were divided into 10 species,a model for the classification of disease severity was obtained,and the recognition accuracy could reach 91.51%.For verifying the robustness of the CDCNNv2 model,comparative experiments were carried out with ResNet 50,Xception,VGG16,VGG19 and DenseNet 121 that used transfer learning.The experimental results showed that the average accuracy of the CDCNNv2 model was increased by 2.78~10.93 percentage points,which had higher classification accuracy and strengthened the robustness of crop disease severity identification.At the same time,based on the model trained by this algorithm,combined with Android technology,a real-time and online crop diseases severity recognition APP was developed.By photographing the disease areas of the crop leaves,the recognition results(speciesdiseaseseverity)and recommendations for prevention and treatment can be obtained between 0.1 s and 0.5 s.In addition,other related supporting functions such as disease encyclopedia made the APP more complete,which can provide effective solutions and ideas for the prevention and treatment of crop diseases.

关 键 词：农作物病虫害 病虫害等级分类 深度迁移学习 ResNet 50 移动应用程序 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP181[自动化与计算机技术—计算机科学与技术]