数码设计Issue(11)：29-32,4.
苹果叶片病害识别的注意力卷积神经网络研究
Research on Attentional Convolutional Neural Network for Apple Leaf Disease Recognition
摘要
Abstract
The application of artificial intelligence to detect apple leaf diseases is of great significance to the control work.At present,the method of apple leaf disease identification using YOLOv5 has the problem of high leakage and detection rate.In order to solve the above problems,the original algorithm of YOLOv5 is optimized by using CAM(Context Augmentation Module)feature information fusion technology to solve the problems of traditional algorithms in multi-scale feature fusion.In addition,a Transformer-based fusion algorithm is proposed,which focuses attention on valuable information.The experiment proves that the integration of modules such as ATCSP,mAP@0.5从0.396提高到0.463 improves by 16.9%,and the recall rate is improved from 0.383 to 0.439,which is 14.6%.The experimental results show that the algorithm can quickly and accurately improve the recognition rate of apple leaf diseases,as well as improve the accurate localization of the disease.关键词
苹果叶片病害识别/改进YOLOv5/CAM/ATCSPKey words
apple leaf disease recognition/improved YOLOv5/CAM/ATCSP分类
信息技术与安全科学引用本文复制引用
曲逸飞,傅卓军..苹果叶片病害识别的注意力卷积神经网络研究[J].数码设计,2024,(11):29-32,4.基金项目
国家重点研发计划课题(项目编号:2017YFD0301506) （项目编号:2017YFD0301506）
湖南省重点领域研发计划项目(项目编号:2022NK2047) （项目编号:2022NK2047）
