基于深度学习的苹果树病虫害检测系统  

Apple Tree Insect Pest and Plant Disease Detection System Based on Deep Learning

作　　者：林少聪 Lin Shaocong(School of Information Engineering,Xiamen Ocean Vocational College,Xiamen 361000,China)

机构地区：[1]厦门海洋职业技术学院信息工程学院,福建厦门361000

出　　处：《洛阳师范学院学报》2024年第11期28-31,共4页Journal of Luoyang Normal University

基　　金：厦门海洋职业技术学院校级课程思政示范课([2022]175);厦门海洋职业技术学院校级一流课程资助项目([2023]3)。

摘　　要：病虫害对苹果产量构成严重影响,采用深度学习技术实现苹果树病虫害自动识别,对提高苹果产量具有积极作用.通过EasyDL平台,构建了基于深度学习的苹果树病虫害检测系统.经过数据集的整理、模型的训练与优化,以及模型部署等步骤,该系统已能自动识别5种苹果叶病虫害.实验结果显示,模型准确率、F1-score、精确率和召回率分别为98.98%、98.98%、99.00%和98.96%,效果较好.Insect pest and plant disease have a serious impact on apple yield.It is of positive significance to improve apple yield to use deep learning technology to realize automatic identification of apple tree insect pest and plant disease.This paper builds an apple tree insect pest and plant disease detection system based on deep learning through EasyDL platform.After data set sorting,model training and optimization,model deployment and other steps,the system realizes automatic identification of five kinds of apple leaf insect pests and plant diseases.The experimental results show that the Top1 accuracy of the model reaches 98.98%,the F1-score reaches 98.98%,the precision reaches 99.00%,the recall reaches 98.96%,thus the system achieves the expected effect.

关 键 词：深度学习 苹果 病虫害 百度EasyDL 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]