Research on Attentional Convolutional Neural Network for Apple Leaf Disease Recognition
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.
apple leaf disease recognitionimproved YOLOv5CAMATCSP
国家科技期刊平台
NSTL
万方数据
